I
I
]

Marine Biological Laboratory Library

Woods Hole, Mass.

Presented by

ESTATE OF HERBERT W. RAND
January 9, 1964

A.

I
I

o-

1!

ru

o

-D
m

o

O

a
m
o

ENTOMOLOGY

FOLSOM

:

BLAKISTON'S SCIENCE SERIES

ENTOMOLOGY

WITH SPECIAL REFERENCE TO

ITS BIOLOGICAL AND ECONOMIC ASPECTS

BY

JUSTUS WATSON FOLSOM, Sc.D. (HARVARD)

ASSISTANT PROFESSOR OF ENTOMOLOGY AT THE UNIVERSITY OF ILLINOIS

ScconD IRcviscD JEfcitton
TliHitb four plates an5 304 eit=3Ficuircs

PHILADELPHIA:

P. BLAKISTON'S SON & CO.
1012 WALNUT STREET

1913

COPYRIGHT, 1913, BY P. BLAKISTON'S SON & Co.

PREFACE

This book gives a comprehensive and concise account of insects.
Though planned primarily for the student, it is intended also for the
general reader.

The book was written in an effort to meet the growing demand for a
biological treatment of entomology.

The existence of several excellent works on the classification of insects
(notably Comstock's Manual, Kellogg's American Insects and Sharp's
Insects) has enabled the author to omit the multitudinous details of
classification and to introduce much material that hitherto has not
appeared in text-books.

As a rule, only the commonest kinds of insects are referred to in the
text, in order that the reader may easily use the text as a guide to personal
observation.

All the illustrations have been prepared by the author, and such as
have been copied from other works are duly credited.

To Dr. S. A. Forbes the author is especially indebted for the use of
literature, specimens and drawings belonging to the Illinois State Labora-
tory of Natural History.

Permission to copy several illustrations from Government publica-
tions was received from Dr. L. O. Howard, Chief of the Bureau of Ento-
mology; Dr. C. Hart Merriam, Chief of the Division of Biological
Survey, and Dr. Charles D. Walcott, Director of the U. S. Geological
Survey. Several desired books were obtained from F. M. Webster, of
the Bureau of Entomology.

Acknowledgments for the use of figures are due also to Dr. E. P.
Felt, State Entomologist of New York; Dr. E. A. Birge, Director of the
Wisconsin Geological and Natural History Survey; Prof. E. L. Mark
and Prof. Roland Thaxter, of Harvard University; Prof. J. H. Comstock
of Cornell University; Prof. C. W. Woodworth of the University of
California; Prof. G. Macloskie of Princeton University; Prof. W. A.
Locy of Northwestern University; Prof. J. G. Needham of Cornell Uni-
versity; Di. George Dimmock of Springfield, Mass.; Dr. Howard Ayers
of Cincinnati, Ohio; Dr. W. M. Wheeler of the American Museum of
Natural History, New York City; Dr. W. L. Tower of the University

VI PREFACE

of Chicago; Dr. A. G. Mayer, Director of the Marine Biological Lab-
oratory, Tortugas, Fla. ; James H. Emerton of Boston, Mass. ; Dr. and
Mrs. G. W. Peckham of Milwaukee, Wis.; Dr. William Trelease, Director
of the Missouri Botanical Garden; Dr. Henry Skinner, as editor of "En-
tomological News"; and the editors of "The American Naturalist. "

Acknowledgments are further due to the Boston Society of Natural
History, the American Philosophical Society and the Academy of Science
of St. Louis.

Courteous permission to use certain figures was given also by The
Macmillan Co.; Henry Holt & Co.; Ginn & Co.; Prof. Carl Chun of
Leipzig; F. Diimmler of Berlin, publisher of Kolbe's Einfuhrung; and
Gustav Fischer of Jena, publisher of Hertwig's Lehrbuch and Lang's
Lehrbuch.

In the revised edition much new matter has been added; particularly,
an entire chapter on the transmission of diseases by insects. A few new
illustrations have been introduced and several of the old figures improved.
The bibliography has been increased by the addition of about one hun-
dred titles of important works.

CONTENTS

CHAPTER PAGE

I. CLASSIFICATION i

II. ANATOMY AND PHYSIOLOGY 21

III. DEVELOPMENT 118

IV. ADAPTATIONS OF AQUATIC INSECTS 152

V. COLOR AND COLORATION 159

VI. ADAPTIVE COLORATION 178

VII. INSECTS IN RELATION TO PLANTS 195

VIII. INSECTS IN RELATION TO OTHER ANIMALS 215

IX. TRANSMISSION OF DISEASES BY INSECTS 234

X. INTERRELATIONS OF INSECTS 253

XI. INSECT BEHAVIOR 283

XII. DISTRIBUTION 300

XIII. INSECTS IN RELATION TO MAN 325

LITERATURE 339

INDEX 387

82478

Vll

ENTOMOLOGY

CHAPTER I

CLASSIFICATION

|

At the outset it is essential to know where insects stand in relation to
other animals.

Arthropoda. Comparing an insect, a centipede and a crayfish
with one another, they are found to have certain fundamental characters
in common. All are bilaterally symmetrical, are composed of a linear
series of rings, or segments, bearing paired, jointed appendages, and have
an external skeleton, consisting largely of a peculiar substance known as
chitin.

If the necessary dissections are made, it can be seen that in each of
these types the alimentary canal is axial in position; above it extends the

FIG. i. Diagram to express the fundamental structure of an arthropod, a. Antenna; al,
alimentary canal; b, brain; d, dorsal vessel; ex, exoskeleton; /, limb; n, nerve chain; s,
suboesophageal ganglion. After SCHMEIL.

dorsal blood vessel and below lies the ventral ladder-like series of seg-
mental ganglia and paired nerve cords, or commissures; between the
commissures that connect the brain and the subcesophageal ganglion
passes the oesophagus. These relations appear in Figs, i and 163.
Furthermore, the sexes are almost invariably separate and the primary
sexual organs consist of a single pair.

No animals but arthropods have all these characters, though the
segmented worms, or annelids, have some of them for example the
segmentation, dorsal heart and ventral nervous chain. On account of
these correspondences and for other weighty reasons it is believed that

ENTOMOLOGY

arthropods have descended from annelid-like ancestors. Annelids,
however, as contrasted with arthropods, have segments that are essen-
tially alike, have no external skeleton and never have paired limbs that
are jointed.

Classes of Arthropoda. Excepting the king-crab, trilobites and a

few other aberrant forms of uncertain
position, the members of the series, or
phylum, Arthropoda fall into six dis-
tinct classes, namely, Crustacea, Arach-
nida, Malacopoda, Diplopoda, Chi-
lopoda and Insecta. These classes are
characterized as follows:

Crustacea. Aquatic, as a rule.
Head and thorax often united into a
cephalothorax. Numerous paired
appendages, typically biramous (Y-
shaped); abdominal limbs often pres-
ent. Two pairs of antennae. Res-
piration branchial (by means of gills)
or cutaneous (directly through the
skin). The exoskeleton contains car-
bonate and phosphate of lime in addi-
tion to chitin. Example, cray-fish.

Arachnida. Terrestrial. Usually
two regions, cephalothorax and abdo-
men ; though various Acarina have but
one and Solpugida have all three-
head, thorax and abdomen. Cephalo-
thorax unsegmented, bearing two pairs
of oral appendages and four pairs of legs. Abdomen segmented or
not, limbless. Respiration tracheal, by means of book-leaf tracheae,
tubular tracheae, or both; stigmata almost always abdominal, at most
four pairs. Heart abdominal in position. Example, Buthus (Fig. 2).

Malacopoda. Terrestrial. Vermiform (worm-like), unsegmented
externally. One pair of antennae, a pair of jaws and a pair of oral slime
papillae. Legs numerous, paired, imperfectly segmented. Respiration
by means of tubular tracheae, the stigmata of which are scattered over the
surface of the body. Numerous nephridia (excretory) are present and
these are arranged segmentally in pairs. Two separate longitudinal

FIG. 2. A scorpion. Buthus.
size.

Natural

CLASSIFICATION

nerve cords, connected by transverse commissures. Integument delicate.
A single genus, Peripatus (Fig. 3), comprising many species.

Diplopoda. Terrestrial. Two regions, head and body. Body
usually cylindrical, with numerous segments, most of which are double
and bear two pairs of short limbs, which are inserted near the median
ventral line. Eyes simple, antennae short, mouth parts consisting of a

FIG. 3. Peripatus capcnsis. Natural size. After MOSELEY.

pair of mandibles and a compound plate, or gnathochilarium. Genital
openings separate, anterior in position (on the second segment of the
body). Example, Spirobolus (Fig. 4).

Chilopoda. Terrestrial. Two regions, head and body. Body long
and flattened, with numerous segments, each of which bears a pair of
long six- or seven-jointed limbs, which are not inserted near the median
line. Eyes simple and numerous (agglomerate in Scutigera), antennas
long. A pair of mandibles and two pairs
of maxillae. A single genital opening, on
the preanal segment. Example, Scolo-
pendra (Fig. 5).

Insecta (Hexapoda). Primarily ter-
restrial. Three distinct regions head,
thorax and abdomen. Head with a pair
of compound eyes in most adults, one pair
of antennas and three pairs of mouth parts
mandibles, maxillae and labium besides
which a hypopharynx, or tongue, is pres-
ent. Thorax with a pair of legs on each
of its three segments and usually a pair
of wings on each of the posterior two seg-
ments; though there may be only one pair of wings (as in Diptera and
male Coccidae) ; the prothorax never bears wings. Abdomen typically
with ten segments (seldom more) and without legs, excepting in some
larvae (as those of Lepidoptera, Tenthredinidse and Panorpidas). Stig-
mata paired and segmentally arranged. A metamorphosis (direct or
indirect) occurs except in Thysanura and Collembola.

FIG. 4. A diplopod, Spirobolus
Hiarginatus. Natural size.

ENTOMOLOGY

Relationships. The interrelationships of the classes of Arthropoda
form an obscure and highly debatable subject.

Crustacea and Insecta agree in so many morphological details that
their resemblances can no longer be dismissed as results of a vague
"parallelism," or "convergence" of development, but are inexplicable
except in terms of community of origin, as Carpenter has insisted.

Arachnida are extremely unlike other
arthropods but find their nearest allies
among Crustacea, particularly the fossil
forms known as trilobites.

Malacopoda, as represented by Perip-
atus, are often spoken of as bridging the
gulf that separates Insecta, Chilopoda and
Diplopoda from Annelida. Peripatus in-
deed resembles the chaetopod annelids in
its segmentally arranged nephridia, dermo-
muscular tube, coxal glands and soft integu-
ment, and resembles the three other classes
in its trachea?, dorsal vessel, lacunar circu-
lation, mouth parts and salivary glands.
These resemblances, however, are by no
means close, and Peripatus does not form
a direct link between the other tracheate
arthropods and the annelid stock, but is
best regarded as an offshoot from the base
of the arthropodan stem.

In speaking of annelid ancestors, none
of the recent annelids are meant, of course,
but reference is made to the primordial
stock from which recent annelids themselves
have been derived.

Though Diplopoda and Chilopoda have
long been grouped together under the name
Myriopoda, they really have so little in common, beyond the numerous
limb-bearing segments and the characters that are possessed by all tra-
cheate arthropods, that their differences entitle them to rank as separate
classes. Chilopoda as a whole are more nearly related to Insecta than
are Diplopoda, as regards segmentation, mouth parts, tracheae, genital
openings and other characters.

Scolopendrclla, now placed either among Diplopoda or else in a class

FIG. 5. A centipede, Scolo-
pendra heros. About two-thirds
the maximum length.

CLASSIFICATION

by itself, Symphyla, presents a remarkable combination of cliplopodan
and insectean characters. Scolopendrella (Fig. 6) and the thysanuran
Campodea have the same kind of head, with its long moniliform antennae,
and agree in the general structure of the mouth parts; the number of

FIG. 6. Section of Scolopendrella immaciilata. b, Brain; c, coxal gland; /, fore intestine;
/;, hind intestine; ;, mid-intestine; n, nerve chain; 0, opening of silk gland; od, oviduct; ov,
ovary; s, silk gland; n, urinary tube. After PACKARD.

body segments is nearly the same, the legs and claws are essentially
alike, and cerci and paired abdominal stylets are present in the two
genera, not to mention the correspondences of internal organization.
Indeed, it is highly probable, as Packard maintained, that the most
primitive insects, Thysanura (and consequently all
other insects), originated from a form much like
Scolopendrella. A singular thysanuran, Anajapy.v
I'csiciilosus (Fig. 7), has been discovered by Silvestri,
who regards it as being in many respects the most
primitive insect known, combining as it does char-
acters of Symphyla, Diplopoda and Campodea.

Silvestri discovered a peculiar arthropod, Acer-
entomon doderoi for which he made a new order
Protura. Berlese added two genera to this order,
namely, Eosentomon and Acerentulus; and according
to good authority Protapieron indicum Schepotieff
belongs to the former genus. Silvestri, followed
by Borner, put Protura among Apterygota; but
Berlese, who grouped these forms under the name
of Myrieutomata, found that they had myriopodan
as well as insectean affinities; and Rimsky-Kor-
sakow argues that Myrientomata cannot be rightly

regarded as insects, but logically constitute a class by themselves; and
that this class does not form a direct link between myriopods and
insects, but that all these groups came from the same ancestral stock.

Fi c . 7 . A n aja pyx
irsiciilosus. Length, 2
mm. After SILVESTRI.

ENTOMOLOGY

The following diagram (Fig. 8) expresses very crudely one view as
to the annelid origin of the chief classes of Arthropoda.

The naturalness of the phylum Arthropoda has been questioned by
Kingsley and Packard. The latter author divided Arthropoda into
five independent phyla, holding that "there was no common ancestor
of the Arthropoda as a whole, and that the group is a polyphyletic one."
This iconoclastic view, however, by emphasizing unduly the structural
differences among arthropods, tends to conceal the many deep-seated
resemblances that exist between the classes of Arthropoda.

Carpenter, in a most sagacious summary of the whole subject
of arthropod relationships, has brought together no little evidence in

favor of a revised form of the
INSECTA

CHILOPODA

CRUSTACEA

ARACHNIDA

DIPLOPODA

ARTHROPODA

MALACOPODA
ANNELIDA

old Miillerian theory of crus-
tacean origins. He traces all
the classes of Arthropoda back
to common arthropodan ances-
tors with a definite number of
segments and distinctly crus-
tacean in character; then traces
these primitive arthropods back
to forms like the nauplius larva
of Crustacea, and these in turn
to a hypothetical form like the
trochosphere larva of recent
polychaste annelids.

Orders of Insects. Lin-
naeus arranged insects in seven
orders, namely, Coleoptera,

Hemiptera, Lepidoptera, Neuroptera, Hymenoptera, Diptera and Aptera.
The wingless insects termed Aptera were soon found to belong to diverse
orders and the name has now become so ambiguous as to meet with little
approbation.

From the Linnaean group Hemiptera, the Orthoptera were set apart ;
the old order Neuroptera, a heterogeneous and unnatural group, has been
split into several distinct orders, and many other changes in the classifica-

i

tion ha,ve been necessary.

Without entering any further into the history of the subject, it is
sufficient to say that increasing discrimination on the part of entomolo-
gists has been followed by a gradual increase in the number of orders,
until our present system has been attained.

FIG. 8.-

-Diagram to indicate the origin of Ar-
thropoda.

CLASSIFICATION 7

Owing to the incomplete condition of entomological knowledge, how-
ever, the best system as yet proposed is but tentative and more or less
open to objection. The most competent and widely approved classifica-
tions are those of Brauer and Packard, and the system here adopted is
essentially that of Brauer, with certain important modifications made by
Packard.

In the course of the following synopsis of the orders of insects it is
necessary to use some terms, as metamorphosis and thysanuriform, in
anticipation of their subsequent definition.

i. Thysanura. No metamorphosis. Mouth parts mandibulate,
either free (ectognathous) or enclosed in the head (entognathous).

FIG. Q. Campodea. Length, 3 mm.

FIG. 10. Lepisma. Length, 10 mm.

Wings invariably absent. Thoracic segments simple and similar. Ab-
dominal segments ten, with two to eight pairs of rudimentary limbs
and two or three anal cerci. Eyes aggregate, compound or absent.
Antennae multiarticulate. Integument thin. Examples, Campodea (Fig.
9), Japyx, Ma-chilis, Lepisma (Fig. 10). Some one hundred and seventy-
five species are known.

2. Collembola. No metamorphosis. Mouth parts entognathous
and typically mandibulate, with occasional secondary suctorial modifica-
tions. Wings invariably absent. Thoracic segments simple and similar

8

ENTOMOLOGY

or prothorax reduced. Body cylindrical or globular; abdomen with six
segments. Ventral tube and furcula usually present, rarely rudimentary.
Eyes ocelliform or absent. Antennae of four segments in most genera;

five or six in a few genera. Integument delicate.
Examples, Achorutes (Fig. n), Sminthurus (Fig.
12). About seven hundred species have been de-
scribed.

Under the term Apterygota (Apterygogenea,
Brauer; Synaptera, Packard) the Thysanura and
Collembola, as primitively wingless insects, are
conveniently distinguished from all other insects,
or Pterygota (Pterygogenea, Brauer).

3. Orthoptera. -Metamorphosis direct.
Mouth parts mandibulate. Wings two pairs as a
rule, though not infrequently reduced or absent;
front wings coriaceous (tegmina) ; hind pair mem-
branous, ample, closely reticulate, plicate along the
numerous radiating principal veins. Abdomen
with ten or eleven segments. Eight families: For-
ficulidae, Hemimeridae(Fig. 13), Blattidae, Mantidae,
Phasmidae (Fig. 241), Acridiidae (Fig. 14), Locusti-
dae, Gryllidae. Over ten thousand species are
known.

Some authors prefer to separate Forficulidae from Orthoptera as
a distinct order, for which Brauer and Packard preserve the old term
Derma ptera of Leach, while Comstock uses West wood's term Euplexoptera.

Hemimeridae consist at pres-
ent of two African species whose
affinities appear to lie with For-
ficulidae, but deserve further
study.

4. Platyptera. Metamor-
phosis direct. Mouth parts man-
dibulate. Wings, if present, two
pairs, delicate, membranous, equal

FIG. ii. The snow
flea, Achorutes nivicola.
Length, 2 mm.

FIG. 12. Sminthurus hortensis.

1.2 mm.

Length,

or hind pair smaller, and with the

principal veins few and simple. Integument usually thin. Nymphs

thysanuriform. Two suborders.

Suborder Corrodentia. Including three families, as follows:

CLASSIFICATION

Termi tides. Eyes facetted. Antennae 9-31 jointed. Mouth parts
prognathous or hypognathous. 1 Prothorax large. Wings elongate, alike,

FIG. 13. Hemimcrus talpoidcs. Length,
11.5 mm. After HAXSEN.

FIG. 14. Scliistoccrca amcrimna. Slightly re-
duced.

membranous, delicate, with indefinite reticulation and with a character-
istic basal suture. Abdomen elongate, with ten segments and a pair of
short, two- jointed anal cerci. In-
tegument delicate. Social in
habit. Example, Termes (Fig.
277). Over one hundred species
are known.

Comstock places Termitidae in
an order by themselves, Isoptera.

EmbiidcE. Eyes facetted. An-
tennae 15-32 jointed. Mouth
parts prognathous. Thorax elon-

FIG. 15. Oligotoma michacli. Length, 10.5
mm. After Me LACHLAN.

gate, prothorax reduced. Wings

(sometimes absent) elongate,

membranous, delicate, with few

and feebly developed longitudinal and cross veins. Abdomen elongate,

with ten or possibly eleven segments, and a pair of stout biarticu-

late cerci. Integument delicate. Not social in habit. Examples,

1 Prognathous, directed forward; hypognath-nts, directed downward.

10

ENTOMOLOGY

FIG. 16. Psocus venosus. Length, 5 mm.

Embia, Oligoloma (Fig. 15). Some twenty species, all from warm
climates.

These insects are most nearly related to Termitidae and Psocidae.
Psocidce. Eyes facetted. Antennae 13-50 jointed. Mouth parts
hypognathous. Prothorax reduced. Wings present, rudimentary or

absent; front pair the larger;
veins few and irregular. Abdomen
with nine or ten segments and no
cerci. Integument delicate. Ex-
ample, Psocus (Fig. 1 6). About
two hundred species.

Comstock raises Psocidae to the
rank of an order, for which he em-
ploys, in a new sense, Brauer's term Corrodentia.

Suborder Mallophaga. Wingless flattened insects, of parasitic
habit. Head large. Eyes consisting of a few isolated ocelli or else ab-
sent. Antennae 3-5 jointed. Mouth parts prognathous. Prothorax dis-
tinct; mesothorax often and metathorax usually transferred to the ab-
dominal region. Abdominal segments eight to
ten in number; no cerci. Parasitic upon birds
and a few mammals. Example, Menopon (Fig.
17). More than fifteen hundred species have
been described.

Packard's order Platyptera originally included
Perlidae. Brauer's order Corrodentia consisted
of Termitidae, Psocidae and Mallophaga; Perlidae
being set apart as an order (Plecoptera} and Em-
biidae being transferred doubtfully to Orthop-
tera.

Enderlein's thorough studies confirm the
view that Termitidae, Embiidae, Psocidae and
Mallophaga constitute a single order.

5. Plecoptera. - - Metamorphosis direct.
Antennae long, multiarticulate. Mouth parts
mandibulate. Prothorax large. Wings two

pairs, membranous, coarsely and complexly reticulate; equal or else hind
wings larger and with an ample plicate anal area. Abdomen with ten
segments and usually a pair of long multiarticulate cerci. Nymphs thy-
sanuriform, aquatic; adults unique in having tracheal gills. Example,
Pteronarcys (Fig. 18). A single family, Perlidae, comprising two hundred
species.

FIG. 17. A chicken
louse, Menopon. Length,
2 mm.

CLASSIFICATION

II

6. Ephemerida. -Metamorphosis direct. Antennae bristle-like.
Mouth parts mandibulate, but atrophied in the adult. Prothorax small.
Wings membranous, minutely reticulate; hind pair much the smaller,

A B

FIG. 18. Pteronarcys regalis. A, nymph (after NEWPORT); B, imago. Slightly reduced.

B
FIG. 19. Hcxageniavariabilis. A, nymph; B, imago. Natural size.

12

ENTOMOLOGY

rarely absent. Abdomen slender, with ten segments and three or two
very long multiarticulate cerci. Integument delicate. Nymphs thysa-

A B

FIG. 20. Libclliila piilcliclla. A, Last nymphal skin; B, imago. Slightly reduced.

nuriform,
species.

aquatic. Example, Hexagenia (Fig. 19). Three hundred

7. Odonata. Metamor-
phosis direct. Antennae in-
conspicuous, bristle-shaped.
Mouth parts mandibulate.
Prothorax small. Wings
four, elongate, subequal, simi-
lar, membranous, minutely
reticulate, with a costal joint,
or nodus. Abdomen slender,
with ten segments. Nymphs
thysanuriform, aquatic. Ex-
ample, Libellula (Fig. 20).
About two thousand species
have been described.

8. Thysanoptera (Phy-

sopoda). Metamorphosis direct, but including a subpupa stage.
Mouth parts suctorial. Prothorax long. Tarsus terminating in a
bladder-like organ. Wings present, rudimentary or absent, the two pairs
narrow, equal, similar, with few or no veins and fringed with long hairs.
Abdomen with ten segments. Minute insects. Example, Euthrips
(Fig. 21). About one hundred and fifty species have been des-
cribed.

9. Hemiptera. Metamorphosis direct (excepting male Coccidae).

FIG. 21. Euthrips tritici. Length, 1.2 mm.

CLASSIFICATION 13

Antennse usually few-jointed. Mouth parts suctorial. Prothorax usu-

FIG. 22. Benacus grisciis. Slightly reduced.

FIG. 23. Head louse,
Pediculus capitis, female.
Length, 2 mm.

FlG. 24. Chrysopa
plorabunda. Slightly

reduced.

ally large. Wings usually present, except in the parasitic forms. Eigh-
teen thousand species. Three suborders:

Suborder Heteroptera. Wings four, folded
flat; front wings thickened basally, membranous
apically (hemelytra), overlapping obliquely; hind
wings membranous. Head not deflexed. Ex-
ample, Benacus (Fig. 22). About twelve thou-
sand species.

Suborder Homoptera. Wings four, sloping

roof-like, similar and membranous or front pair somewhat coriaceous

throughout. Head deflexed. Example,
Cicada (Fig. 207). Six thousand species.
Suborder Parasita. Wingless. Eyes
simple or none. Thoracic segments inti-
mately united; tarsus with a single claw.
Integument thin. Parasites upon mammals.
Example, Pediculus (Fig. 23). Some fifty
species are known.

10. Neuroptera. Metamorphosis in-
direct. Antennae conspicuous. Mouth
parts mandibulate. Prothorax large.

Wings almost always four, membranous, subequal or else hind pair

FIG. 25. Bitlacus slrigosits. Na-
tural size.

14 ENTOMOLOGY

smaller, complexly reticulate, not plicate. Larvae thysanuriform or in
some cases cruciform, and aquatic or terrestrial. Example, Chrysopa
(Fig. 24). About six hundred species have been named.

11. Mecoptera. Metamorphosis indirect. Mouth parts mandib-
ulate, at the end of a deflexed rostrum, or beak. Prothorax small.
Wings four, elongate, membranous, naked, coarsely reticulate, or else
rudimentary or absent. Larvae cruciform, caterpillar-like, with numer-
ous prolegs, carnivorous. Example, Bittacus (Fig. 25). A single family,
Panorpidae, comprising but few known species.

12. Trichoptera. Metamorphosis indirect. Antennae filiform.
Mouth parts of imago rudimentary or imperfectly suctorial ; mandibles

FIG. 26. Molanna cinerea. A, Larva; B, imago. X 4 diameters. After FELT.

rudimentary or absent. Prothorax small. Wings four, membranous,
hairy, veins moderate in number, cross veins few; hind pair almost
always the larger, with plicate anal area. Larvae suberuciform, aquatic,
usually case-forming. Example, Molanna (Fig. 26). Between five and
six hundred species are known.

13. Lepidoptera. Metamorphosis indirect. Mouth parts suctorial,
mandibles absent or rudimentary (except in a few generalized species).
Prothorax small. Wings four, similar, membranous, clothed with scales,
veins moderate in number, cross veins few. Larvae cruciform (caterpil-
lars), phytophagous (almost never carnivorous), mandibulate. Some
fifty thousand species have been described. Two suborders, not sharply
separated from each other.

Suborder Heterocera. Antennae of various forms, but not ter-

CLASSIFICATION 15

minating in a distinct knob or club. Frenulum usually present. Chiefly
nocturnal in habit. Example, Callosamia (Fig. 237).

Suborder Rhopalocera. Antennae simple, terminating in a dis-
tinct club and without conspicuous lateral processes. Frenulum absent.
Diurnal normally. Examples, Papilio (Fig. 27), Anosia (Fig. 244, A).

14. Coleoptera. Metamorphosis indirect. Mouth parts mandibu-
late. Pro thorax large, as a rule. Wings four; front pair horny (elytra),
meeting in a straight line; hind pair membranous, often folded. Larvae
thysanuriform or cruciform. Example, Hydro philus (Fig. 28). About
one hundred and fifty thousand species.

15. Diptera. Metamorphosis indirect. Mouth parts typically suc-
torial, but modified for piercing, lapping, rasping, etc. Prothorax small.

B

FIG. 27. Papilio troilus. A, Larva; B, larva suspended for pupation; C, chrysalis. Nat-
ural size.

One pair of wings (mesothoracic) , membranous, transparent, with few
veins; wings rudimentary or absent, however, in most of the parasitic
species; hind wings represented by a pair of knobbed threads, or balan-
cers. Larvae cruciform, with the head frequently reduced to a mere
vestige with or without a pair of mandibles, and usually without true
legs, though pseudopods may be present. Example, Tipula (Fig. 29).
About forty thousand described species.

16. Siphonaptera (Aphaniptera) . Metamorphosis indirect. Head
small. Eyes simple or absent. Mouth parts suctorial. Body laterally
compressed. Thoracic segments subequal. Wings absent or at most
quite rudimentary. Larvae with a head, mandibulate, apodous. Para-

i6

ENTOMOLOGY

sitic insects. Example, Ctenocephalus (Fig. 30). One hundred and fifty
species.

17. Hymenoptera. Metamorphosis indirect. Mouth parts at the
same time mandibulate and suctorial. Prothorax usually small. Wings

FIG. 28. Hydrophilns triangnlaris. Natural size.

four, similar, membranous, transparent, with a few irregular veins and
cells; hind pair the smaller. Females with an ovipositor, modified for
sawing, boring or stinging. Larvae cruciform, mandibulate, caterpillar-

j\

FIG. 29. Tipitla. A, larva; B, cast pupal skin; C, imago. Slightly reduced.

like, with head and legs, or else maggot-like and apodous. Twenty-five
or thirty thousand species. Two suborders.

Suborder Terebrantia (Phytophaga, Sessiliventres). Abdomen
broadly attached to [the thorax. Ovipositor modified for boring, sawing

CLASSIFICATION

or cutting. Larvae with complex mouth parts and frequently abdominal
legs. Phytophagous. Example, Tremex (Fig. 31).

Suborder Aculeata (Heterophaga, Petiolata). Abdomen petio-
late or subpetiolate; first abdominal segment transferred to the thorax.
Ovipositor often modified to form a sting. Larvae apodous. Example,
Apis (Fig. 281).

Interrelations of the Orders. The modern classification aims to
express relationships, and these are most clearly to be ascertained by a
comparative study of the facts of anatomy and development.

FIG. 30. Cat and dog flea, Ctenocephalus canis.
A, Larva (after KUXCKEL D'HERCULAIS); B, adult.
Length of adult. 2 mm.

FIG. 31. Tremex columfra. A,
Imago; B, larva (with parasitic larva
of Thalcssa attached). Natural size.
After RILEY.

The most generalized, or primitive, insects are the Thysanura. Sub-
tracting their special, or adaptive, peculiarities, their remaining characters
may properly be regarded as inheritances from some vanished ancestral
type of arthropod. This primordial type, then, probably had three
simple and equal thoracic segments differing but slightly from the ten
abdominal segments; three pairs of legs and no wings; three pairs of
exposed biting mouth parts; a pair of long, many-jointed antennas and a
pair of cerci of the same description; a thin naked integument; a simple
straight alimentary canal distinctly divided into three primary regions;
a ganglion and a pair of spiracles for each of the three thoracic and the
first eight abdominal segments, if not all the latter; no metamorphosis;
functional abdominal legs and active terrestrial habits.

The existing form that best meets these requirements is Scolopendrella,
which is not an insect, however, but belongs among or near the diplopods.
The most primitive of known insects are Anajapyx and Campodea,
3

1 8 ENTOMOLOGY

through which other insects trace their origin to the stock from which
Symphyla and Diplopoda arose.

Collembola, though specialized in several important ways, all have
the same peculiar kind of entognathous mouth parts as Campodea and
Japyx, for which reason and many others it is believed that Collembola
are an offshoot from the thysanuran stem. Collembola, however, are
not nearly so primitive as Thysanura, for the former have fewer ab-
dominal segments than the latter, exhibit much greater concentration of
the nervous system, and are uniquely specialized in several respects,
notably as regards the ventral tube and the furcula, or springing organ.

Returning to Thysanura the genera Machilis and Lepisma show de-
cided orthopteran affinities; thus their eyes are compound and their
mouth parts strongly orthopteran; indeed, the likeness of Lepisma to a
young cockroach is striking, as is also that of Japyx to a young forficulid.

In short, as Hyatt and Arms express it, "The generalized form of
Thysanura, and the manner in which it reappears in the larvae of other
insects, is the natural key of the classification."

Orthoptera probably arose directly from the original thysanuriform
stem.

Platyptera, as a whole, are most nearly related to Orthoptera on the
one hand and to Plecoptera on the other. Termitidae have strong orthop-
teran affinities and Embiidae have even been placed in the order Orthop-
tera, though the latter family is most nearly allied to Termitidse and
Psocidae. These two are approached rather closely by Mallophaga and
exhibit, by the way, some collembolan characters, as Enderlein has
pointed out.

Plecoptera, which Packard placed in his group Platyptera, are better
regarded as a distinct order with some orthopteran and many ephemerid
and odonate affinities. The strong resemblance between nymphs of
Plecoptera, Ephemerida and Odonata indicates community of origin.

Ephemerida and Odonata are well circumscribed orders, most nearly
related to each other, but sharply separated, nevertheless, by differences
in the wings, mouth parts and other organs. Ephemerida are almost
unique among insects in having a pair of genital openings a primitive
condition.

Thysanoptera form a distinct order, which is usually placed next to
Hemiptera, chiefly on account of the suctorial mouth parts, though even
in this respect there is no close agreement between the two orders.

Hemiptera stand alone and give few hints of their ancestry.- They
are least unlike Orthoptera and possibly originated with Thysanoptera

CLASSIFICATION 19

from some mandibulate and winged form. The conversion of manclibu-
late into suctorial organs may be seen within the order Collembola, but it
is highly improbable that Hemiptera arose from forms like Collembola.
Hemiptera are exceptional among insects with a direct metamorphosis in
their highly developed type of suctorial mouth parts.

Metamorphosis offers, upon the whole, the broadest criteria for the
separation of insects into primary groups. All the orders considered thus
far are characterized either by no metamorphosis or by a slight, or so-
called '"direct," or "incomplete," transformation. The following orders,
on the contrary, are distinguished by an "indirect," or "complete,"
metamorphosis, which appears in Neuroptera and attains its maximum
development in Diptera and Hymenoptera.

With Neuroptera the eruciform type of larva appears, as a derivative
of the earlier thysanuriform type. The larva of Mantispa, as Packard
has shown, actually passes, during its individual development, from the
primary, thysanuriform stage to the secondary, eruciform condition.

Mecoptera form an isolated order, though their caterpillar-like larvae,
with eleven or twelve pairs of legs, suggest affinities with Lepidoptera
and, more remotely, with the tenthredinid Hymenoptera.

Trichoptera, while much like Mecoptera in structure and metamor-
phosis, are undoubtedly closely related to Lepidoptera; in view of the ex-
tensive and deep-seated resemblances between caddis flies and the most
generalized moths (Micropterygidas) there is little doubt that Trichoptera
and Lepidoptera originated from the same stock.

The origin of the coherent group Coleoptera is by no means clear, al-
though thysanuriform larvae occur frequently in this order. Packard
suggests that both beetles and earwigs arose from some thysanuroid form
or that the primitive coleopterous larva sprang from some metabolous
neuropteroid form. In any linear arrangement of the orders the position
of Coleoptera is largely arbitrary, and here the order is intruded between
Lepidoptera and Diptera simply for want of a more satisfactory place.

Lepidoptera, Trichoptera and Mecoptera are probably branches from
one stem. Lepidoptera, Diptera and Kymenoptera are regarded by
Packard as having had a common origin from metabolic Neuroptera.

Among Diptera, such larvae as those of Culicidas are comparatively
primitive, according to Packard, and larvae of Muscidae are secondary, or
adaptive, forms.

Siphonaptera used to be regarded as Diptera and are probably an off-
shoot from the dipteran stem.

The most primitive hymenopterous larvae are those of the sawflies

20 ENTOMOLOGY

(Tenthredinidae), judging from their resemblance to mecopterous and
lepidopterous larvae; and the simple, maggot-like form of the larvae of
ants, bees, wasps and parasitic Hymenoptera is due to secondary modi-
fications in correlation with their sedentary mode of life.

In Diptera and Hymenoptera the phenomenon of metamorphosis
attains its greatest complexity, as was remarked. Opinions differ as to
which of these two orders is the more specialized. Hymenoptera are
commonly called the "highest" insects, when their remarkable psycho-
logical development is taken into account; but from a purely structural
standpoint it is hard to say which order is the more complex indeed, the
two orders are specialized in so many different ways that no precise com-
parison can be made between them.

The following diagram (Fig. 32) is a graphic summary of what has
just been said in regard to the genealogy of the orders of insects. The

PLECOPTERA
PLATYPTERA

SIPHONAPTERA
THYSANOPTERA

ORTHOPTERA k ^-HEMIPTERA

COLLEMBOLA

GOLEOPTERA THYSANURA

FIG. 32. Genealogical diagram of the orders of insects.

positions of Hemiptera and Coleoptera are most open to criticism. The
central group (T) is the hypothetical thysanuroid source of all insects,
including Thysanura themselves. Though Thysanura and Collembola
show no traces of wings, even in the embryo, it should be borne in mind
that all the other insects probably had winged ancestors and that it is
more reasonable to assume a single winged group as a starting point than
to suppose that wings originated independently in several different
groups of insects.

CHAPTER II

ANATOMY AM) PHYSIOLOGY

i. SKELETON

Number and Size of Insects. The number of insect species al-
ready known is about 300,000 and it is safe to estimate the total number
of existing species as at least one million.

Among the largest living species are the Venezuelan beetle, Dynastes
hercules, which is 155 mm. long, and the Venezuelan grasshopper, Acri-
dium latreillci, which has a length of 166 mm. and an alar expanse of 240
mm. Among Lepidoptera, Attacus atlas of Indo-China spreads 240 mm.;
Attacus cccsar of the Philippines, 255 mm.; and the Brazilian noctuid
Erebus agrippina, 280 mm. Some of the exotic wood-boring larvae attain
a length of 150 mm.

The giants among insects have been found in the Carboniferous, from
which Brongniart described a phasmid (Titanophasma) as being one-fourth
of a meter long.

At the other extreme are beetles of the family Trichopterygidae, some
of which are only 0.25 mm. in length, as are also certain hymenopterous
egg-parasites of the families Chalcididae and Proctotrypidae.

Thus, as regards size, insects occupy an intermediate place among
animals; though some insects are smaller than the largest protozoans and
others are larger than the smallest vertebrates.

Segmentation. One of the fundamental characteristics of arthro-
pods is their linear segmentation. The subject of the origin of this seg-
mentation is far from simple, as it involves some of the most difficult
questions of heredity and variation. As arthropod segmentation is
usually regarded as an inheritance from annelid-like ancestors, the sub-
ject resolves itself into the question of the origin of the segmented from
the unsegmented "worms." Cope, Packard and others give the me-
chanical explanation which is here summarized. In a thin-skinned, un-
segmented worm, the flexures of the body initiated by the muscular sys-
tem would throw the integument into folds, much as in the leech, and
with the thickening of the integument, segmentation would appear from
the fact that the deposit of chitin would be least at the places of greatest
flexure, i. e., the valleys of the folds, and greatest at the places of least

21

22 ENTOMOLOGY

flexure, i. e., the crests of the folds. This explanation, which has been
elaborated in some detail by the Neo-Lamarckians, applies also to the
segmentation of the limbs, as well as the body.

Head. In an insect several of the most anterior pairs of primary
appendages have been brought together to co-operate as mouth parts and
sense organs, and the segments to which they belong have become com-
pacted into a single mass the head in which the original segmentation
is difficult to trace. The thickened cuticula of the head forms a skull,
which serves as a fulcrum for the mouth parts, furnishes a base of attach-
ment for muscles and protects the brain and other organs.

While the jaws of most insects can only open and shut, transversely,
their range of action is enlarged by movements of the entire head, which
are permitted by the articulation between the head and thorax.

As a rule, one segment overlaps the one next behind; but the head,
though not a single segment of course, never overlaps the prothorax in the
typical manner, but is usually received into that segment. This condi-
tion, which may possibly have been brought about simply by the back-
ward pull of the muscles that move the head, has certain mechanical
advantages over the alternative condition, in securing, most economically,
freedom of movement of the head and protection for the articulation itself.

The size and strength of the skull are usually proportionate to the
size and power of the mouth parts. In some insects almost the entire
surface of the head is occupied by the eyes, as in Odonata (Fig. 20, B)
and Diptera (Fig. 39). In muscid and many other dipterous larvas, or
"maggots," the head is reduced to the merest rudiment.

Though commonly more or less globose or ovate, the head presents
innumerable forms; it often bears unarticulated outgrowths of various
kinds, some of which are plainly adaptive, while others are apparently
purposeless and often fantastic.

Sclerites and Regions of the Skull. The dorsal part of the skull
(Fig. 33) consists almost entirely of the epicranium, which bears the com-
pound eyes; it is usually a single piece, or sclerite, though in some of the
simpler insects it is divided by a Y-shaped suture. The middle of the
face, where the median ocellus often occurs, is termed the front; ordinarily
this is simply a region, though a frontal sclerite exists in some insects.
Just above the front, and forming the summit of the head, is the region
known as the vertex; it often bears ocelli. The clypeus is easily recog-
nized as being the sclerite to which the upper lip, or labrum, is hinged,
though the clypeus is not invariably delimited as a distinct sclerite.
The cheeks of an insect are known as the genes, and post-gen<z sometimes

ANATOMY AND PHYSIOLOGY 23

occur. On the under side of the head is the gula, which bears the under
lip, or labhim. That part of the skull nearest the prothorax is termed
the occiput; usually it is not delimited from the epicranium, though in
some insects it is continuous with the post-genae to form a distinct sclerite.
The occiput surrounds the opening known as the occipital foramen,
through which the oesophagus and other organs pass into the thorax.
The membrane of the neck in Orthoptera and some other insects contains
small cervical sclerites, dorsal, lateral or ventral in position; these, in the
opinion of Comstock, pertain to the last segment of the head. Besides

oc

B

B

FIG. 33. Skull of a grasshopper, Melanoplus differ entiatis. a, Antenna; c, clypeus; e,
compound eye; /, front; g, gena; /, labrum; Ip, labial palpus; m, mandible; mp, maxillary
palpus; o, ocelli; oc, occiput; pg, post-gena; v, vertex.

those described, a few other cephalic sclerites may occur, small and in-
conspicuous, but nevertheless of considerable morphological importance.
Tentorium. In the head is a chitinous supporting structure known
as the tentorium. This consists of a central plate from which diverge two
pairs of arms extending to the skull (Fig. 34). The central plate lies
between the brain and the subcesophageal ganglion and under the. oeso-
phagus, which passes between the anterior pair of arms. The tentorium
braces the skull, affords muscular attachments and holds the cephalic
ganglia and the oesophagus in place. It is not a true internal skeleton,
but arises from the same ectodermal layer which produces the external

24 ENTOMOLOGY

cuticula; though authors are not agreed as to the details of the develop-
ment.

Eyes. The eyes are of two kinds simple and compound. The latter,
or eyes proper, conspicuous on each side of the head, are of common
occurrence except in the larvae of most holometabolous insects, in some

fe s;: :.:*>_._;

$frl*.- : ;i,f?:t?g3fg<j- -, y-Sf 51 *^ jF.

FIG. 34. Skull of a grasshopper, Dissos- FIG. 35. Head of a gyrinid beetle, Dincutus,
li'lni Carolina, o, occipital foramen; /, /, to show divided eye.

anterior arms of tentorium.

generalized forms (as Collembola) and in parasitic insects. The com-
pound eyes (Fig. 40) are convex and often hemispherical, though their
outline varies greatly; thus it may be oval (Orthoptera) or triangular
(Xotonecta), while in the aquatic beetles of the family Gyrinidae (Fig. 35)
each eye has a dorsal and a ventral lobe, enabling the insect to see upward

FIG. 36. Agglomerate eyes of a male coccid,
Lcachia fusdpennis. After SIGNORET.

FIG. 37. Facets of a compound eye of
Mclanopliis. Highly magnified.

and downward at the same time; so also in Oberea and other terrestrial
beetles of the same family. Superficially, a compound eye is divided
into minute areas, or facets, which though circular in the agglomerate
type of eye (Fig. 36) are commonly more or less hexagonal (Fig. 37), as
the result of mutual pressure. These facets are not necessarily equal in

ANATOMY AND PHYSIOLOGY

2.S

size, for in dragon flies the dorsal facets are frequently larger than the
ventral. In diameter the facets range from .016 mm. (Lyccena) to .094
mm. (Cerambyx). Their number is often enormous; thus the house fly
(Musca domestica) has 4,000 to each eye, a butterfly (Papilio] 17,000, a
beetle (Mordella) 25,000 and a sphingid moth
27,000; on the other hand, ants have from 400
down, the worker ant of Edton having at most
a single facet on each side of the head.

Ocelli. The simple eyes, or ocelli, appear
as small polished lenses, either lateral or dorsal
in position. Lateral ocelli (Fig. 38) occur in
the larvae of most holometabolous insects and
in parasitic forms. Dorsal ocelli, supplemen-
tary to the compound eyes, occur on or near
the vertex, and are more commonly three in
number, arranged in a triangle, as in Odonata,
Diptera (Fig. 39) and Hymenoptera (Fig. 40)
as well as many Orthoptera and Hemiptera.
Few beetles have ocelli and almost no butter-
flies (Lerema accius with its one ocellus being the only exception known),
though not a few moths have two ocelli.

As explained beyond, the compound eyes are adapted to perceive
form and movements and the ocelli to form images of objects at close
range or simply to distinguish between light and darkness.

FIG. 38. Head of a cater-
pillar, Samia I'fi i-opiti, to show
lateral ocelli.

FIG. 39. Ocelli and compound eyes of a fly, PJionnia retina. A, male; B, female.

Sexual Differences in Eyes. In most Diptera (Fig. 39) and in
Hymenoptera (Fig. 40) and Ephemeridae as well, the eyes of the male are
larger and closer together (Iwloptic) than those of the female (dichoptic).
This difference is attributed to the fact that the male is more active than
the female, especially in the matter of seeking out the opposite sex.
Among ants of the same species the different forms may differ greatly

26

ENTOMOLOGY

in the number of lateral facets. Thus in Formica pratensis, according to
Forel, the worker has about 600 facets in each eye, the queen 800-900
and the male 1,200.

FIG. 40. Ocelli and compound eyes of the honey bee, Apis mcllifera. A, queen; B, drone.

After CHESHIRE.

B

D

K

FIG. 41. Various forms of antennae. A , filiform, Euschistus;. B, setaceous, Plathemis; C,
momliiorm,Catogenus; D, geniculate, Bombus; f, flagellum; p, pedicel; 5, scape; E, irregular,
Phormia; a, arista; F, setaceous, Galerita; G, clavate, Anosia; H, pectinate, male Ptilodactyla;
/, lamellate, Lachnosterna; J, capitate, Megalodacne; K, irregular, Dineutus.

ANATOMY AND PHYSIOLOGY

Blind Insects. Many larvae, surrounded by an abundance of food
and living often in darkness, need no eyes and have none ; this is true of
the dipterous "maggots" and many other sedentary larvae, particularly
such as are internal parasites (Tachinidas, Ichneumonidae), or such as
feed within the tissues of plants (many Buprestidae, Cerambycidae and
Curculionidas) . Subterranean or cavernicolous insects are either eyeless
or else their eyes are more or less degenerate, according to the amount of
light to which they have access. The statement is made that blind in-
sects never have functional wings.

Antennae. The antennae, never more than a single pair (though
embryonic "second antennas" occur in Thysanura and Collembola), are
situated near the compound eyes and frequently between them. With
rare exceptions the antennae have always several and usually many seg-
ments. In form these organs are exceedingly varied, though many of
them may be referred to the types represented in Figs. 41-43.

Though homologous in all insects, the antennas are by no means equiv-
alent in function. They are commonly tactile (grasshoppers, etc.) or
olfactory (beetles, moths) and occasionally auditory (mosquito), as
described beyond, but may be adapted for other than sensory functions.
Thus the antennae of the aquatic beetle Hydro pJiilus are used in connection
with respiration and those of the male
Meloe to hold the female.

Sexual Differences in Anten-
nae. In moths of the family Saturnii-
dae (S. cecropia, C. promethea, etc.) the
pectinate antennae of the male are
larger and more feathered than those
of the female, and differ also in having
more segments (Fig. 42). Here the
antennae are chiefly olfactory, and the
reason for their greater development in
the male appears from the fact that the
male seeks out the female by means of
the sense of smell and depends upon
his antennae to perceive the odor ema-
nating from the opposite sex.

The plumose antennae of the male mosquito (Fig. 43) are highly de-
veloped organs of hearing, and are used to locate the female; they have
delicate nbrillae of various lengths, some of which are thrown into sym-
pathetic vibration by the note of the female (p. 85).

FIG. 42. Antennae of a moth, Samia
cecropia. A, male; B, female.

28

ENTOMOLOGY

Meloe has just been mentioned. In Sminthurus malmgrenii (Collem-
bola) the antennae of the male are provided with hooks and otherwise
adapted to grasp those of the female at copulation.

Though systematists have recorded many instances of antennal
antigeny, the interpretation of these sexual differences has received very
little attention; though a beginning in the subject has been made by
Schenk, whose results will be referred to in connection with the sense
organs.

Mouth Parts. On account of their great range of differentiation,
the mouth parts are of fundamental importance to the systematist, par-
ticularly for the separation of insects into orders. Most of the orders fall

FIG. 43. Antennae of mosquito, Cnlcx pi picas.

O B

l

A, male; B, female.

into two groups according as the mouth parts are either biting (mandibu-
late) or sucking (suctorial]. Collembola and Hymenoptera, however,
combine both functions; Diptera, though suctorial, exhibit various modi-
fications for piercing, lapping or rasping; Thysanoptera are partly man-
dibulate but chiefly suctorial; and adult Ephemerida and Trichoptera
have but rudimentary mouth parts.

The mandibulate orders are Thysanura, Collembola (primarily),
Orthoptera, Platyptera, Plecoptera, Ephemerida (rudimentarily in adult),
Odonata, Neuroptera, Mecoptera and Coleoptera.

The mouth parts of an insect consist typically of labrum, mandibles,
maxillcE, labium and hypo pharynx (Fig. 44), though these organs differ
greatly in different orders of insects. The mandibulate, or primary type,

ANATOMY AND PHYSIOLOGY

2 9

from which the suctorial, or secondary type, has been derived, will be
considered first.

Mandibulate Type. The labrum, or upper lip. in biting insects is a
simple plate, hinged to the clypeus and moving up and down; though
capable of protrusion and retraction to some extent. It covers the man-
dibles in front and pulls food back to these organs. On the roof of the
pharynx, under the labrum and clypeus, is the epipharynx; this consists
of teeth, tubercles or bristles, which serve in some insects merely to hold
food, though as a rule the epipharynx in mandibulate insects bears end-
organs of taste (Packard).

FIG. 44.- Mouth parts of a cockroach, Ischnoptcra peniisyfaanica. A, labrum; B, mandi-
ble; C, hypopharynx; D, maxilla; E, labium; c, cardo; g (of maxilla), galea; g (of labium).
glossa; /, lacinia; //, labial palpus; m, mentum; mp, maxillary palpus; p, paraglossa; pf,
palpifer; pg, palpiger; 6-, stipes; sm, submentum. B, D and E are in ventral aspect.

The mandibles, or jaws proper, move in a transverse plane, being
closed by a pair of strong adductor muscles and opened by a pair of weaker
abductors. The mandible is almost always a single solid piece. In
herbivorous insects (Fig. 45, A] it is compact, bluntly toothed, and often
bears a molar, or crushing, surface behind the incisive teeth. In car-
nivorous species (B) the mandible is usually long, slender and sharply
toothed, without a molar surface. Often, as in soldier ants, the man-
dibles are used as piercing weapons; in bees (C) they are used for various
industrial purposes; in some beetles they are large, grotesque in form and

ENTOMOLOGY

apparently purposeless. The mandibles of Onthophagus (D) and many
other dung beetles consist chiefly of a flexible lamella, admirably adapted
for its special purpose. In Euphoria (Fig. 262), which feeds on pollen

FIG. 45. Various forms of mandibles. A,Mdanoplus; B, Cicindela; C,Apis; D, Onthoph-
agus; E, Chrysopa; F-I, soldier termites (after HAGEN).

and the juices of fruits, the mandibles, and the other mouth parts as
well, are densely clothed with hairs. In the larva of Chrysopa, the inner
face of the mandible (Fig. 45, E) has a longitudinal groove against which
the maxilla fits to form a canal, through which the blood of plant lice is

sucked into the cesophagus. In termites (F-I)
the mandibles assume curious and often inexpli-
cable forms.

Next in order are the maxilla, or under jaws,
which are less powerful than the mandibles and
more complex, consisting as they do of several
sclerites (Figs. 44, 46). Essentially, the maxilla
/ consists of three lobes, namely, palpus, galea and
lacinia, which are borne by a stipes, and hinged to
the skull by means of a cardo. The palpus, always
lateral in position, is usually four- or five-jointed
and is tactile, olfactory or gustatory in function.
The lacinia is commonly provided with teeth or
spines. The maxillae supplement the mandibles
by holding the food when the latter open, and help
to comminute the food. Additional maxillary
sclerites, of minor importance, often occur.

The labium, or under lip, may properly be lik-
ened to a united pair of maxillae, for both are formed
on the same three-lobed plan. This correspondence
is evident in the cockroach, among other generalized insects. Thus, in
this insect (Fig. 44) :

FIG. 46. Maxilla of
Harpalus caliginosus,
ventral aspect. c,
cardo; g, galea; /,
lacinia; p, palpus; pf,
palpifer; s, stipes; sg,
subgalea.

ANATOMY AND PHYSIOLOGY

LABIUM = MAXILLA

palpus = palpus
paraglossa = galea

glossa = lacinia
palpiger = palpifer
tncntum sti piles
submcntum with gula = cardines

In most mandibulate orders the glossas unite to form a single me-
dian organ, as in Ear pains (Fig. 47, g). The
labium forms the floor of the pharynx and
assists in carrying food to the mandibles and
maxillae.

The use of the term "second maxillae" for the
labium of an insect is open to objection, as it im-
plies an equivalence with the second maxillae of
Crustacea which is by no means established.

The tongue, or hypo pharynx, is a median
fleshy organ (Fig. 44) which is usually united
more or less with the base of the labium. In

insects in general, the salivary glands open at

, , . FIG. 47. Labium of Har-

the base of the hypopharynx. In the most gen- pa i us ca i; g i n osus, ventral

' eralized insects, Thysanura and Collembola, the as P ect ; s, united glossae,

J termed the glossa; m, men-

hypopharynx is a compound organ, consisting of turn; p, palpus; pg, palpiger;

a median ventral lobe, or lingua, and two dorso- ^ n tJ^ agl T?e median pot
lateral lobes, termed superlingucR by the author, tion of the labium beyond

... , the mentum is termed the

Superlmguas occur in a few other mandibulate u g ui a .
orders (Orthoptera, Fig. 48; Ephemerida, Fig.

49), but have not yet been recognized in the more
specialized orders of insects.

Suctorial Types. Owing to their greater com-
plexity, suctorial mouth parts are not nearly so well
understood as the mandibulate organs, but enough
has been learned to enable us to homologize the two
types, even though morphologists still disagree in
regard to minor details of interpretation.

The suctorial, or haustellate, orders are Collem-
bola (in part), Thysanoptera (in part), Hemiptera,
Trichoptera (imperfectly), Lepidoptera, Diptera,
Siphonaptera and Hymenoptera (which have func-
tional mandibles, however).

FIG. 48. Hypo-
pharynx of Hcmimcrus
talpoides. I, lingua; s,
superlingua. After
HANSEN.

3 2

ENTOMOLOGY

Hemiptera. The beak, or rostrum, in Hemiptera consists (Fig. 50)
of a conspicuous, one- to four-jointed labium, which ensheathes hair-like
mandibles and maxillae and is covered above at its base by a short

labrum. The mandibles and maxillae
are sharply-pointed, piercing organs
and the former frequently bear retrorse
barbs just behind the tip; the two
maxillae lock together to form a suck-
ing tube. Though primarily a sheath,
the labium bears at its extremity sen-
sory hairs, which are doubtless used to
test the food. This general description
applies to all Hemiptera except the

parasitic forms, which present special modifications. A pharyngeal
pumping apparatus is present, which is similar in its general plan to that

FIG. 49. Hypopharynx of an ephe-
merid, Hcptagcnia. '/, lingua; si, si,
superlinguae. After VAYSSIERE.

D

FIG. 50. Mouth parts of a hemipteron, Bcnaciis grisens. A , dorsal aspect; B, transverse
section; C, extremity of mandible; D, transverse section of mandibles and maxillae; c, canal;
/, labrum; //, labium; in, mandible; ;w.v, maxilla;.

of Lepidoptera and Diptera, as presently described, though it differs as
regards the smaller details of construction.

ANATOMY AND PHYSIOLOGY

33

Lepidoptera. In Lepidoptera, excepting Enocephala, the labrum is
reduced (Fig. 51) and the mandibles are either rudimentary or absent
(Rhopalocera). The two maxillae
are represented by their galeae,
which form a conspicuous pro-

FIG. 51. Head of a sphingid moth, Phlcge-
thontius scxta. a, antenna; c, clypeus; c, eye;
/, labrum; in, mandible; p, pilifer; pr, pro-
boscis.

boscis; the grooved inner faces

of the galeae (or laciniae, according

to Kellogg) form the sucking tube,

which opens into the oesophagus.

The labium is reduced, though

the labial palpi (Fig. 52) are well

developed. The so-called rudi-
mentary mandibles of Anosia and

other forms have been shown by

Kellogg to be lateral projections

of the labrum (Fig. 51) and he

terms them pi-lifers.

The exceptional structure of

the mouth parts in the generalized genus Eriocephala (Micropteryx) sheds

much light on the morphology of these organs
in other Lepidoptera, as Walter and Kellogg
have shown. In this genus there are func-
tional mandibles; the maxilla presents palpus,
galea, lacinia, stipes and cardo, though there
is no proboscis; the labium has well developed
submentum, mentum and palpi; a hypo-
pharynx is present.

The sucking apparatus, as described by
Burgess, is essentially like that of Diptera.
Five muscles, originating at the skull and in-
serted on the wall of a pharyngeal bulb, serve
to dilate the bulb that it may suck in fluids,
while numerous circular muscles serve by con-
tracting successively to squeeze the contents
of the bulb back into the stomach; a hypo-
pharyngeal valve prevents their return forward.
Diptera. In the female mosquito the
mouth parts (Fig. 53) are long and slender.

As Dimmock has found, the labrum and epipharynx combine 1 to form a

FIG. 52. Head of a but-
terfly, Vanessa, a, antenna;;
I, labial palpus; p, proboscis.

1 Kulagin, however, describes them as remaining separate.

34

ENTOMOLOGY

sucking tube; the mandibles and maxillae are delicate, linear, piercing
organs, the latter being barbed distally; maxillary palpi are present; the
hypopharynx is linear also and serves to conduct saliva ; the labium forms
a sheath, enclosing the other mouth parts when they are not in use; a
pair of sensory lobes, termed labella. occur at the extremity of the labium.
The oesophagus is dilated to form a bulb, or sucking organ, from which
muscles pass outward to the skull ; when these contract, the bulb dilates
and can suck in fluids, as blood or water, which are forced back into the

FIG. 53. Mouth parts of female mosquito, Culcx pipicns. A , dorsal aspect; B, transverse
section; C, extremity of maxilla; D, extremity of labrum-epipharynx; a, antenna; e, com-
pound eye; h, hypopharynx; /, labrum-epipharynx; II, labium; m, mandible; mx, maxilla;
p, maxillary palpus. B, after DIMMOCK.

stomach by the elasticity of the bulb itself, according to Dimmock; the
regurgitation of the food is prevented by a valve.

The male mosquito rarely if ever sucks blood, and its mouth parts
differ from those of the female in that the mandibles are aborted and the
maxillae slightly developed, but with long palpi, while the hypopharynx
coalesces with the labium and there is no cesophageal bulb.

Hymenoptera. In the honey bee, which will serve as a type, the
labrum (Fig. 54) is simple ; the mandibles are well developed instruments
for cutting and other purposes and the remaining mouth parts form a

ANATOMY AND PHYSIOLOGY

35

highly complex suctorial apparatus, as follows. The tongue is a long
flexible organ, terminating in a ''spoon" (Fig. 127) and clothed with hairs
of various kinds, for gathering nectar or for sensory or mechanical pur-
poses. The maxillae and labial palpi form a tube embracing the tongue,
while the epipharynx fits into the space between the bases of the maxillae
to complete this tube. Through this canal nectar is driven, by the ex-
pansion and contraction of the tube itself, according to Cheshire, except
that when only a small quantity of nectar is taken, this passes from the
spoon into a fine "central duct," or also into the "side ducts," which are
specially fitted to convey quantities of fluid too small for the main tube.
For a detailed account of the highly complex and exquisitely adapted
mouth parts of the honey bee, the reader is referred to Cheshire's admir-
able work or to Packard's Text-Book.

FIG. 54. Mouth parts of the honey bee, Apis mdlifera. a, base of antenna; br, brain; c,
clypeus; h, hypopharynx; /, labrum; Ip, labial palpus; Amentum; wo, mouth; mx, maxilla;
sin, submentum. After CHESHIRE.

Segmentation of the Head. The determination of the number of
segments entering into the composition of the insect head has been a
difficult problem. As no segment bears more than one pair of primary
appendages, there are at least as many segments in the head as there are
pairs of primary appendages. On this basis, then, the antennae, man-
dibles, maxillae and labium may be taken to indicate so many segments;
but in order to decide whether the eyes, labrum and hypopharynx repre-
sent segments, other than purely anatomical evidence is necessary. The
key to the subject is furnished by embryology. At an early stage of
development the future segments are marked off by transverse grooves
on the ventral surface of the embryo, and the pairs of segmental appen-
dages are all alike (Fig. 195), or equivalent, though later they differen-
tiate into antennae, mouth parts, legs, etc. Moreover, the nervous sys-
tem exhibits a segmentation which corresponds to that of the entire in-

ENTOMOLOGY

S

sect; in other words, each pair of primitive ganglia, constituting a
romere, indicates a segment. Now in front of the oesophagus three primi-
tive segments appear, each with its neuromere (Fig. 55) : first in position,
an ocular segment, destined to bear the compound eyes; second, an anten-
nal segment; third, an intercalary (premandibular) segment, which in
the generalized orders Thysanura and Collembola bears a transient pair
of appendages that are probably homologous with the second antennae of
Crustacea. In the adult, the ganglia of these three segments have united

to form the brain, and
the original simplicity
and distinctness have
been lost. The labrum,
by the way, does not rep-
resent a pair of append-
ages, but arises as a single
median lobe. Behind the
oesophagus, three embry-
onic segments are clearly
distinguishable, each
with its pair of appen-
dages, namely, mandibu-
lar, maxillary and labial.
Finally, the hypo-
pharynx, or rather a
part of it, claims a place
in the series of segmental
appendages, as the
author has maintained;
for in Collembola its two
dorsal constituents, or
superlinguce, develop

essentially as do the other paired appendages and, moreover, a super-
lingual neuromere (Fig. 55) exists. The four primitive ganglia immedi-
ately behind the mouth eventually combine to form the subcesophageal
ganglion.

To summarize the head of an insect is composed of at least six seg-
ments, namely, ocular, antennal, intercalary, mandibular, maxillary and
labial; and at most seven, since a superlingual segment occurs between
the mandibular and maxillary segments in Collembola and probably
Thysanura, though it has not yet been discovered in the more specialized
insects.

FIG. 55. Paramedian section of an embryo of the col-
lembolan Anurida maritima, to show the primitive cephalic
ganglia, i, ocular neuromere; 2, antennal; j, intercalary;
4, mandibular; 5, superlingual; 6, maxillary; 7, labial; 8,
prothoracic; 9, mesothoracic; a, antenna; /, labrum; //,
labium; I 1 , 1 2 , 1 3 , thoracic legs; m, mandible; mx, maxilla.
After FOLSOM.

\\ATOMY AND PHYSIOLOGY 37

Thorax. The thorax, or middle region, comprises the three segments
next behind the head, which are termed, respectively, pro-, meso- and
metathorax. In aculeate Hymenoptera, however, the thoracic mass in-
cludes also the first abdominal segment, then known as the propodeum, or
median segment. Each of the three thoracic segments bears a pair of
legs in almost all adult insects, but only the meso- and metathorax may
bear wings.

The differentiation of the thorax as a distinct region is an incidental
result of the development of the organs of locomotion, particularly the
wings. Thus in legless (apodous) larvae the thoracic and abdominal seg-
ments are alike; when legs are present, but no wings, the thoracic seg-
ments are somewhat enlarged ; and when wings occur, the size of a wing-
bearing segment depends on the volume of the wing muscles, which in
turn is proportionate to the size of the wings. When wings are absent
(as in Thysanura and Collembola) or the two
pairs equal in area (as in Termitidae, Odonata,
Trichoptera and most Lepidoptera) the meso-
and metathorax are equal. If the fore wings
exceed the hind ones (Ephemeridae, Hymenop-
tera) the mesothorax is proportionately larger
than the metathorax; as also in Diptera, where

no hind wings occur. If the fore wings are

FIG. 56. Diagram of the

small (Coleoptera) or almost absent (Stylopidae) principal sclerites of a thoracic
the mesothorax is correspondingly smaller than

the metathorax. The prothorax, which never parapteron; ps, postscutellum;

s, scutum; si, scutellum; st,

bears wings, may be enlarged dorsally to torm sternum. After COMSTOCK.
a protective shield, as in Orthoptera, Hemip-

tera and Coleoptera; or, on the contrary, may be greatly reduced, as in
Ephemerida, Odonata, Lepidoptera and Hymenoptera. In the primitive
Apterygota the prothorax may become reduced (many Collembola) or
slightly enlarged (Lepisma).

The dorsal wall of a thoracic segment is termed the notum, or tergum;
the ventral wall, the sternum; and each lateral wall, a pleuron; the re-
striction of these terms to particular segments of the thorax being indi-
cated by the prefixes pro-, meso- or meta-. These parts are usually divided
by sutures into distinct pieces, or sclerites, as represented diagrammati-
.cally in Fig. 56. Thus the tergum of a wing-bearing segment is regarded
as being composed of four sclerites (tergites, Fig. 57), namely and in order,
prcescutum, scutum, scutellum and postscutellum. The scutum and
scutellum are commonly evident, but the two other sclerites are usually

ENTOMOLOGY

small and may be absent. Each pleuron consists chiefly of two sclerites
(pleurites, Fig. 58), separated from each other by a more or less oblique
suture. The anterior of these two, which joins the sternum, is termed
the episternum; the other, the epimeron. The former is divided into two
sclerites in Odonata and both are so divided in Neuroptera.

The sternum, though usually a single plate, is in some instances divided
into halves, as in the cockroach, or even into five sclerites (Forficulidae)
To these should be added the patagia of Lepidoptera a pair of erec-
tile appendages of the prothorax; and the paraptera, or tegula, of Lepi-
doptera and Hymenoptera a pair of small
sclerites at the bases of the front wings.

Each thoracic segment bears a pair of
spiracles in the embryo and in some adults
as well (Campodea, Heteroptera), but in
most imagines there are only two pairs of
thoracic spiracles, the suppressed pair
being usually the prothoracic.

The sclerites of the thorax owe their
origin probably to local strains on the in-
tegument, brought about by the muscles
of the thorax. Thus the primitively wing-
less Thysanura and Collembola have no
hard thoracic sclerites, though certain
creases about the bases of the legs may
be regarded as incipient sutures, produced
mechanically by the movements of the
legs. In soft nymphs and larvae, the
sclerites do not form until the wings
develop; and in forms that have nearly or quite lost their wings, as
Pediculidae, Mallophaga, Siphonaptera and some parasitic Diptera, the
sclerites of the thorax tend to disappear. Furthermore, the absence of
sclerites in the prothorax is probably due to the lack of prothoracic
wings, notwithstanding the so-called obsolete sutures of the pronotum
in grasshoppers.

Endoskeleton. An insect has no internal skeleton, strictly speaking,
though the term endoskeleton is used in reference to certain ingrowths of
the external cuticula which serve as mechanical supports or as protections
for some of the internal organs. The tentorium of the head has already
been referred to. In the thorax three kinds of chitinous ingrowths may
be distinguished according to their positions: (i) phragmas, or dorsal

FIG. 57. Dorsal aspect of the
thorax of a beetle, Hydrous piceus.
I, pronotum; 2, mesopraescutum; j,
mesoscutum; 4, mesoscutellum; 5,
mesopostscutellum; 6, metaprae-
scutum; 7, metascutum; 8, meta-
scutellum; 9, metapostscutellum.
After NEWPORT.

ANATOMY AND PHYSIOLOGY

39

projections ; (2) apodemes, lateral ; (3) apophyses, ventral. The
phragmas (Fig. 59) are commonly three large plates, pertaining to the
meso- and metathorax, and serving for the origin of indirect muscles of
flight in Lepidoptera, Diptera, Hymenoptera and other strong-winged

FIG. 58. Ventral aspect of a carabid beetle, Galerita janus. i, prosternum; 2, proepi-
sternum; j, proepimeron; 4, coxal cavity; 5, inflexed side of pronotum; 6, mesosternum; 7,
mesoepisternum; 8, mesoepimeron ; p, metastcrnum; 10, antecoxal piece; n, metaepi-
sternum; 12, metaepimeron; 15, inflexed side of elytron; a, sternum of an abdominal segment;
an, antenna; c, coxa; /, femur; lp, labial palpus; md, mandible; mp, maxillary palpus; /,
trochanter; tb, tibia; Is, tarsus.

orders. The apodemes are comparatively small ingrowths, occurring
sometimes in all three thoracic segments, though usually absent in the
prothorax. The apophyses occur in each thoracic segment as a pair of
conspicuous processes, which either remain separate or else unite more or
less; leaving, however, a passage for the ventral nerve cord.

a

a

40 ENTOMOLOGY

These endoskeletal processes serve chiefly for the origin of muscles
concerned with the wings or legs, and are absent in such wingless forms as
Thysanura, Pediculidas and Mallophaga.

Some ambiguity attends the use of these terms. Thus some writers
use the term apodemes for apophyses and others apply the term apo-
deme to any of the three kinds of ingrowths.

Legs. In almost all adult insects and in most larvae each of the three
thoracic segments bears a pair of legs. The leg is articulated to the

sternum, episternum and epimeron and consists
of five segments (Fig. 60), in the following order :
'coxa, trochanter, femur, tibia, tarsus. The coxa,
or basal segment, often has a posterior sclerite,
the trochantine. 1 The trochanter is small, and
in parasitic Hymenoptera consists of two sub-
segments. The femur is usually stout and con-
spicuous, the tibia commonly slender. The
tarsus, rarely single-jointed, consists usually of
five segments, the last of which bears a pair of
claws in the adults of most orders of insects and
a single claw in larvae; between the claws in
most imagines is a pad, usually termed the
pulvillus, or em podium.

Adaptations of Legs. The legs exhibit
a great variety of adaptive modifications. A
walking or running insect, as a carabid or cicin-
delid beetle (Fig. 62, A) presents an average
condition as regards the legs. In leaping in-
sects (grasshoppers, crickets, Haltica) the hind
femora are enlarged (B) to accommodate the
powerful extensor muscles. In insects that make
little use of their legs, as May flies and Tipulidae,

these appendages are but weakly developed. The spinous legs of dragon
flies form a basket for catching the prey on the wing. Modifications of the
front legs for the purpose of grasping occur in many insects, as the terres-
trial families Mantidae (C) and Reduviidae and the aquatic families Belos-
tomidae and Naucoridas (D}. Swimming species present special adapta-
tions of the legs (Fig. 229), as described in the chapter on aquatic insects.
In digging insects, the fore legs are expanded to form shovel-like organs,

a-

FIG. 59. Transverse sec-
tions of the thoracic segments
of a beetle, Goliathns, to show
the endoskeletal processes.
A, prothorax; B, mesothorax;
C, metathorax; a, a, apoph-
yses; ad, apodeme; p,
phragma. After KOLBE.

1 But on account of the ambiguous use of this last term, the name meron (Fig. 61), pro-
posed by Walton, is to be preferred.

ANATOMY AND PHYSIOLOGY

tb

notably in the mole-cricket (Fig. 62, ), in which the fore tibia has some

resemblance to the human hand, while the tarsus and tibia are remarkably

adapted for cutting roots, after the manner of shears.

The Scarabaeidae have fossorial legs, the anterior tarsi of

which are in some genera reduced (F) or absent ; they

are rudimentary in the female (G) of Phantzus carnifex

and absent in the male (#), and absent in both sexes

of Dcltochilum. Though females of Phanaus lose their

front tarsi by digging, the degenerate condition of

these organs cannot be attributed to the inheritance

of a mutilation, but may have been brought about by

disuse ; though no one has explained why the two sexes

should -differ in this respect. Many insects use the

legs to clean the antennae, head, mouth parts, wings

or legs; the honey bee (with other bees, also ants,

Carabidae, etc.) has a special antenna-cleaner on the

front legs (Fig. 264, D), which is described, with

other interesting modifications of the legs, on page 211.
Indeed, the legs serve many such minor purposes

in addition to locomotion. They are generally used

to hold the female during coition, and in several genera

of Dytiscidae (Dytiscus, Cybister) the male (Fig. 62, /)

has tarsal disks and cupules chiefly on the front tarsi,

for this purpose. Among other secondary sexual pecu-
liarities of the legs may be mentioned the tibial brushes

of the male Catocala concumbens, regarded as scent

organs, and the queer appendages of male Dolicho-

podidae that dangle in the air as these flies perform their dances.

The pulvillus is commonly an adhe-
sive organ. In flies it has glandular hairs
that enable the insects to walk on smooth
surfaces and to walk upside down; so
also in many beetles and notably in the
honey bee (Fig. 63) ; in this insect the
pulvillus is released rapidly from the sur-
face to which it has been applied, by
rolling up from the edges inward.

Sense organs occur on the legs.
Thus tactile hairs are almost always

present on these appendages, while auditory organs occur on the front

FIG. 60. Leg of a
beetle, Cdlosoma cali-
ditm. c, coxa; d,
claws; /, femur; s,
spur; y 1 -/ 5 , tarsal seg-
ments; tb, tibia; Ir,
trochanter.

~ es

em

FIG. 6 1. Left hind leg of B iliac us.
c, coxa genuina; fw,epimeron; es, epi-
sternum; /, femur; m, meron;
chanter.

/, tro-

ENTOMOLOGY

tibiae of Locustidae, Gryllidse and some ants. Finally, the legs may be
used to produce sound, as in Stenobothrus and such other Acridiidae as
stridulate by rubbing the femora against the tegmina.

FIG. 62. Adaptive modifications of the legs. A, Cicindela sexgultata; B, Nemobius
vittatus, hind leg; C, Stagmomantis Carolina, left fore leg; D, Pelocoris femorata, right fore leg;
E, Gryllotalpa borealis, left fore leg; F, Canthon lavis, right fore leg; G, Phanaus carnifex, fore
tibia and tarsus of female; H, P. carnifex, fore tibia of male; I, Dylisciisfasciventris, right fore
leg of male; c, coxa; /, femur; s, spur; /, trochanter; tb, tibia; ts, tarsus.

Legs of Larvae. Thoracic legs, terminating in a single claw, are
present in most larvae. Caterpillars have, in addition, fleshy abdominal
legs (Fig. 64) ending in a circlet of hooks. Most caterpillars have five

ANATOMY AND PHYSIOLOGY

43

pairs of these legs (on abdominal segments 3, 4, 5, 6, and 10), but the rest

vary in this respect. Thus Lagoa has seven pairs (segments 2-7 and 10)

and Geometridae two (segments 6, and 10),

while a few caterpillars (Tischeria, Limacodes)

have none. Larvae of saw flies (Tenthre-

dinidae) have seven or eight pairs of abdominal

legs and larvae of most Panorpidae, eight pairs.

Not a few coleopterous larvae (some Ceram-

bycidae, Phytonomus) also have abdominal legs,

which are incompletely developed, however, as

compared with those of Lepidoptera.

The legless, or apodous, condition occurs
frequently among larvae and always in correla-
tion with a sedentary mode of life; as in
the larvae of many Cerambycidae, almost all
Rhynchophora, a few Lepidoptera, all Dip-
tera, and all Hymenoptera except Tenthre-
dinidae, Siricidae, and other Terebrantia.

Among adult insects, female scale insects are exceptional in being
legless.

Walking. An adult insect, when walking, normally uses its legs in
two sets of three each; thus the front and hind legs of one side and the

FIG. 63. Foot of honey
bee, Apis mellifera. c, c,
claws; p, pulvillus; / 3 -/ 5 , tarsal
segments. After CHESHIRE.

FIG. 64. Caterpillar of Phlegethontius sexta. Natural size.

middle leg of the other move forward almost simultaneously though not
quite, for the front leg moves a little before the middle one, which, in
turn, precedes the hind leg. During these movements the body is sup-

44

ENTOMOLOGY

ported by the other three legs, as on a tripod. The front leg, having been
extended and its claws fixed, pulls the body forward by means of the con-
traction of the tibial flexors; the hind leg, on the contrary, pushes the
body, by the shortening of the tibial extensors, against the resistance
afforded by the tibial spurs; the middle leg acts much like the hind one,
but helps mainly to steady the body. Different species show different
peculiarities of gait. In its analysis, the walking of an insect is rather
intricate, as Graber and Marey have shown.

The mode of action of the principal leg muscles may be gathered from
Fig. 65. Here the flexion of the tibia would cause the tibial spur (s)
to describe the line 5 i; and the backward movement of the leg due to the

. r

FIG. 65. Mechanics of an insect's leg. a, axis of coxa; r, coxa; cl, claw; e, extensor of
tibia; cc, extensor of claw; et, extensor of tarsus; /, flexor of tibia; fc, flexor of claw; //, flexor
of tarsus; r, r, rotators of coxa; s, spur; /, trochanter muscle (elevator of femur) ; //, tibia.
After GRABER.

upper coxal rotator r would cause the spur to follow the arc s 3. As the
resultant of both these movements, the path actually described by the
tibial spur is 5 2; then, as the leg moves forward, the curve is continued
into a loop.

Caterpillars use their legs successively in pairs, and when the pairs of
legs are few and widely separated, as in Geometridae, a curious looping
gait results.

The leg muscles of a cockroach are shown in Fig. 66.

Leaping. The hind legs, inserted nearest the center of gravity, are
the ones employed in leaping, and they act together. A grasshopper
prepares to jump by bending the femur back against the tibia; to make
the jump, the tibia is jerked back against the ground, into which the tibial

AX ATOMY AND PHYSIOLOGY

45

spurs are driven, and the straightening of the leg by means of the power-
ful extensors throws the insect into the air. At the distal end of the femur
are two lobes, one on each side of the tibia, which prevent wobbling move-
ments of the tibia.

Wings. The success of insects as a class is to be attributed largely
to their possession of wings. These and the mouth parts, surpassing all
the other organs as regards range of differentiation, have furnished the
best criteria for the purposes of classification. The
wings of insects present such countless differences
that an expert can usually refer a detached wing
to its proper genus and often to its species, though
no less than three hundred thousand species of in-
sects are already known.

Typically, there are two pairs of wings, attached
respectively to the mesothorax and the metathorax,
the prothorax never bearing wings, as was said.
When only one pair is present it is almost invariably
the anterior pair, as in Diptera and male Coccidae,
though in male Stylopidae it is the posterior pair,
the fore wings being rudimentary.

In bird lice, fleas and most other parasitic in-
sects, the wings have degenerated through disuse.
In Thysanura and Collembola there are no traces
of wings even in the embryo; whence it is inferred
that wings originated later than these orders of
insects.

Miiller and Packard have regarded the wings as
tergal outgrowths; Tower, however, has recently
shown that the wings of Coleoptera, Orthoptera
and Lepidoptera are pleural in origin, arising just
below the line where later the suture between the
pleuron and tergum will originate, though the wings
may subsequently shift to a more dorsal position.

Modifications of Wings. Being commonly more or less triangular,
a w r ing presents three margins: front (costal), outer (apical) and inner
(anal). Various modifications occur in the front wings, which are in
many orders more useful for protection than for flight. Thus, in Orthop-
tera, they are leathery, and are known as tegmina; in Coleoptera they are
usually horny, and are termed elytra; in Heteroptera, the base of the front
wing is thickened and the apex remains membranous, forming a hemely-

FIG. 66. Muscles of
left mid leg of a cock-
roach, posterior aspect.
abc, abductor of coxa;
adc, adductor of coxa;
ef, extensor of femur;
el, extensor of tibia; Jf,
flexor of femur; //,
flexor of tibia; fta,
flexor of tarsus; rt,
retractor of tarsus.
After MIALL and
DENNY.

4 6

ENTOMOLOGY

Scl

Iron. Diptera have, in place of the hind wings, a pair of clubbed threads,
known as balancers, or halteres, and male Coccidae have on each side a
bristle that hooks into a pocket on the wing and serves to support the
latter. In many muscid flies a doubly lobed membranous squama occurs
at the base of the wing.

In Hymenoptera the front and hind wings of the same side are held
together by a row of hooks (hamuli); these are situated on the costal
margin of the hind wing and clutch a rod-like fold of the fore wing. In
very many moths, the two wings are enabled to act as one by means of a
frenulum, consisting of a spine or a bunch of bristles near the base of the
hind wing, which, in some forms, engage a membranous loop on the fore
wing.

Venation, or Neuration. A wing is divided by its veins, or ner-
vures, into spaces, or cells. The distribution of the veins is of great sys-
tematic importance,
Sc2 -- but, unfortunately,

the homologies of the
veins in the different
orders of insects have
not been fixed until
recently, so that no
little confusion has
existed upon the sub-
ject. For example,
the term discal cell,
used in descriptions of Lepidoptera, Diptera, Trichoptera and Psocidae,
has in no two of these groups been applied to the same cell. The
admirable work of Comstock and Needham, however, seems to settle
this disputed subject. By a study of the tracheae which precede and, in a
broad way, determine the positions of the veins, these authors have ar-
rived at a primitive type of tracheation (Fig. 67) to which the more com-
plex types of tracheation and venation may be referred.

In general, the following principal longitudinal veins may be distin-
guished, in the following order: costa, subcosta, radius, media, cubitus, and
anal (Figs. 67-71).

The costa (C) strengthens the front margin of the wing and is essen-
tially unbranched.

The subcosta (Sc) is close behind the costa and is unbranched in the
imagines of many orders in which there are few wing veins, though it is
typically a forked vein.

3d A 2dA

IstA

Cu2

FIG. 67. Hypothetical type of venation. A , anal vein; C,
costa; Cu, cubitus; M, media; R, radius; Sc, subcosta. Figs.
67-71 after COMSTOCK and NEEDHAM.

ANATOMY AND PHYSIOLOGY

47

The radius (R), though subject to much modification, is typically five-
branched, as in Fig. 67. The second principal branch of the radius is
termed the radial sector (Rs).

The media (I/) is often three-branched and is typically four-branched,
according to Comstock and Needham.

The cubitus (Cu) has two branches.

R3

FIG. 68. Wing of a fly, Rhyphus. Lettering as before.

The anal veins (A] are typically three, of which the first is generally
simple, while the second and third are many-branched in wings that have
an expanded anal area.

The Plecoptera, as a whole, show the least departure from the primi-
tive type of venation; which is well preserved, also, in the more general-
ized of the Trichoptera.

Starting from the primitive type, specialization has occurred in two
ways: by reduction and by
addition. Reduction occurs
either by the atrophy of veins
or by the coalescence of two
or more adjacent veins.
Atrophy explains the lack
of all but one anal vein in
Rhyphus (Fig. 68) and other
Diptera, and the absence of
the base of the media in
Anosia (Fig. 69) and many

other Lepidoptera; in the pupa of Anosia, the media may be found com-
plete. Coalescence "takes places in two ways: first, the point at which
two veins separate occurs nearer and nearer the margin of the wing,
until finally, when the margin is reached, a single vein remains where
there were two before; second, the tips of two veins may approach each
other on the margin of the wing until they unite, and then the coales-

2dA

FIG. 69. Wing of a butterfly, Anosia. Lettering as

before.

ENTOMOLOGY

cence proceeds towards the base of the wing. ' ' (Comstock and Needham.)
The former, or outward, kind of coalescence is common in most orders of
insects; the latter, or inward, kind is especially prevalent in Diptera.

Specialization by addition occurs by a multiplication of the branches
of the principal veins.

Comstock and Needham have succeeded in homologizing practically
all the types of neuration, including such perplexing types as those

of Ephemerida (Fig. 70),
Odonata (Fig. 20, B) and
Hymenoptera (Fig. 71),
and their thorough work
affords a sound basis for .a
rational terminology of the
wing veins; there is no
longer any excuse for the
lamentable confusion that
has hitherto attended the
study of venation.

Folding of Wing. In
some beetles (as Chrysobo-

thris) the wings are no larger than the elytra and are not folded; in others,
however, the wings exceed the elytra in size, and when not in use are folded
under the elytra in ways that are simple but efficient, as described by
Kolbe and by Tower. To be understood, the process of folding should
be observed in the living insect. As described by Tower for the Colorado

FIG. 70. Wings of a May fly. Lettering as before.

FIG. 71. A typical hymenopterous wing. Lettering as before.

potato beetle, the folded wing (Fig. 72, B) exhibits a costal joint (a), a
fold parallel to the transverse vein (b), and a complex joint at d. The
wing rotates upon the articular head (ah] and when folded back beneath
the wing-covers the inner end of the cotyla (c] is brought into contact
with a chitinous sclerite of the thorax, which stops the further movement

ANATOMY AND PHYSIOLOGY

49

of the cotyla medianward, and as the wing swings farther back the middle
system of veins (ni) is pushed outward and anteriorly. This motion,
combined with the backward movement of the wing as a whole, produces
the folding of the distal end of the wing. There are no traces of muscles
or elastic ligaments in the wing which could aid in the folding.

Mechanics of Flight. The mechanism of insect flight is much less
complex than one might anticipate. Indeed, owing to the structure of
the wing itself, simple up and down movements are sufficient for the
simplest kind of flight. During oscillation, the plane of the wing changes,
as may be demonstrated by
holding a detached wing by
its base and blowing at right
angles to its surface; the
membrane of the wing then
yields to the pressure of the
air while the rigid anterior
margin does not, to any
great extent. Similarly, as
the wing moves downward
the membrane is inclined
upward by the resistance of
the air, and as the wing c ''
moVes upward the mem-
brane bends downward.
Therefore, by becoming de-
flected, the wing encounters

FIG. 72. Wing of Lcpthiotarsa dcccmlincata. A,

spread; B, folded; a, costal joint; all, articular head;
an, anterior system of veins; b, transverse vein; c,
cotyla; d, joint; m, middle system of veins; /^poste-
rior system of veins. After TOWER.

a certain amount of resist-
ance from behind, which is
sufficient to propel the insect.
The faster the wings vibrate,

the greater the deflection, the greater the resistance from behind, and
the faster the flight of the insect.

The path traced in the air by a rapidly vibrating wing may be deter-
mined by fastening a bit of gold leaf to the tip of the wing and allowing the
insect a wasp, for example to vibrate its wings in the sunlight, against
a dark background. Under these conditions, the trajectory of the wing
appears as a luminous elongate figure 8. During flight, the trajectory
consists of a continuous series of these figures, as in Fig. 73.

Marey, the chief authority on animal locomotion, used chronopho-
tography, among other methods, in studying the process of flight, and
5

ENTOMOLOGY

obtained at first twenty, and later one hundred and ten, successive photo-
graphs per second of a bee in flight. As the wings were vibrating 190
times per second, however, the images evidently represented isolated and
not consecutive phases of wing movement. Nevertheless, the images
could be interpreted without difficulty, in the light of the results ob-
tained by other methods. At length he ob-
tained sharp but isolated images of vibrating
wings with an exposure of only -J^^-Q-Q of a
second.

The frequency of wing vibration may be
ascertained from the note made by the wing
if it vibrates rapidly enough to make one;
and, in any case, may be determined graphic-
ally by means of a kymograph, which, in

one of its forms, consists of a cylinder covered with smoked paper and
revolved by clockwork at a uniform rate. The insect is held in such
a position that each stroke of the wing makes a record on the smoked
paper, as in Fig. 74. Comparing this record with one made on the same
paper by a tuning-fork of known vibration period, the frequency of wing

FIG. 73. Trajectory of the
wing of an insect.

FIG. 74. Records of wing vibration. A , mosquito, A nophclcs. Above is the wing record
and below is the record of a tuning-fork which vibrated 264.6 times per second. B, wasp.
Polistcs. The tuning-fork in this instance had a vibration frequency of 97.6.

vibration can be determined with great accuracy. As the wing moves in
the arc of a circle, the radius of which is the length of the wing, the ex-
treme tip of the wing records only a short mark ; if, however, the wing is
pressed against the smoked cylinder, a large part of the figure-8 trajectory
may be obtained, as in Fig. 74, B. The wings of the two sides move syn-
chronously, as Marey found.

ANATOMY AND PHYSIOLOGY

B

The smaller the wings are, the more rapidly they vibrate. Thus a
butterfly (P. raped) makes 9 strokes per second, a dragon fly 28, a sphin-
gid moth 72, a bee 190 and a house-fly 330.

Wing Muscles. The base of a wing projects into the thoracic cavity
and serves for the insertion of the direct muscles of flight. Regarding the
wing as a lever (Fig. 75, A), with the fulcrum at p, it is easy to understand
how the contraction of muscle e raises the wing and that of muscle d low-
ers it. These muscles are shown
diagrammatically in Fig. 75, B.
Besides these, there are certain
muscles of flight which act in-
directly upon the wings, by
altering the form of the thoracic
wall. Thus the muscle ie (Fig.
75, B) elevates the wing by
pulling the tergum toward the
sternum; and the longitudinal
muscle id depresses the wing
indirectly by arching the ter-
gum of the thorax.

Though up and down move-
ments are all that are necessary
for the simplest kind of insect
flight, the process becomes com-
plex in proportion to the effi-
ciency of the flight. Thus in
dragon flies there are nine
muscles to each wing: five de-
pressors, three elevators and

g

A

FIG. 75. A, diagram to illustrate the action
of the wing muscles of an insect. B, diagram of
wing muscles, a, alimentary canal; en, muscle
for contracting the thorax, to depress the wings;
d, depressor of wing; e, elevator of wing; ex,
muscle for expanding the thorax, to elevate the
wings; id, indirect depressor; ie, indirect ele-
vator; /, leg muscle; p, pivot, or fulcrum; s,
sternum; /, tergum; wg, wing. After GRABER.

one adductor. The earlier ac-
counts ofthe mechanics of flight
by Marey and others have been
recently modified and improved upon by Stellwaag and by Ritter, whose
modern methods of investigation have added considerably to our knowl-
edge of the subject. They show, particularly, the parts played by the
thoracic sclerites during flight.

Abdomen. The chief functions of the abdomen are respiration and
reproduction, to which should be added digestion. The abdomen as a
whole has undergone less differentiation than the thorax and presents a
simpler and more primitive segmentation.

52 ENTOMOLOGY

Segments. A typical abdominal segment bears a dorsal plate, or
tergum, and a ventral plate, or sternum, the two being connected by a pair
of pleural membranes, which facilitate the respiratory movements of the
tergum and sternum. Most of the abdominal segments have spiracles,
one on each side, situated in or near the pleural membranes of the first
seven or eight segments. The total number of pairs of spiracles is as
follows :

Thoracic. Abdominal. Total.

Campodea 3 o 3

Japyx -4 7 Ir

Machilis 2 7 9

Lcpisma 2 8 10

Blattidae, Acridiidae 10

Odonata 2 10

Heteroptera . . 3 ' 7 IO

Lepidoptera 7 9

Diptera .... 79

In most embryo insects there are eleven pairs of spiracles (three thora-
cic and eight abdominal); in adults, however, two pairs are commonly
suppressed the prothoracic and the eighth abdominal.

Number of Abdominal Segments. Though consisting typically
of ten segments the number evident in such generalized insects as Thy-
sanura and Ephemerida eleven occur in various adult Orthoptera, with
traces of a twelfth, while Heymons has detected twelve abdominal seg-
ments in embryos of Orthoptera and Odonata. In the more specialized
orders, ten may usually be distinguished, with more or less difficulty,
though the number is apparently, and in some cases actually, less owing
to modifications of the base of the abdomen in relation to the thorax,
but especially to modifications of the extremity of the abdomen, for
sexual purposes.

Modifications. In aculeate Hymenoptera the first segment of the
abdomen has been transferred to the thorax, where it is known as the
propodeum, or median segment; in other words, what appears to be the
first abdominal segment is actually the second; this, as in bees and wasps,
often forms a petiole, which enables the sting to be applied in almost any
direction. In Cynipidas the tergum of segment two or three occupies
most of the abdominal mass, the remaining segments being reduced and
inconspicuous. The terminal segments of the abdomen often telescope
into one another, as in many Coleoptera and Hymenoptera (Chrysididas),
or undergo other modifications of form and position which obscure the

ANATOMY AND PHYSIOLOGY

53

segmentation. As to the number of evident (not actual) abdominal seg-
ments, Coleoptera show five or six vcntrally and seven or eight dorsally;
Lepidoptera, seven in the female and eight in the male; Diptera, nine
(male Tipulidae) or only four or five; and Hymenoptera, nine (Tenthre-
dinidae) or as few as three (Chrysididae) . In the larvae of these insects,
however, nine or ten abdominal segments are usually distinguishable,
though the tenth is frequently modified,
being in caterpillars united with the ninth.

Appendages. Rudimentary abdom-
inal limbs occur in Thysanura (Mackilis,
Fig. 76). Functional abdominal legs do
not occur in adult insects, but in larvae the
abdominal pro-legs (often called " false legs."
Fig. 64) are homologous with the thoracic
legs and the other paired segmental appen-
dages, as the embryology shows. The em-
bryo of (Ecanthus, according to Ayers, has
ten pairs of abdominal appendages (Fig.
197), equivalent to the thoracic legs. Most
of these embryonic abdominal appendages
are only transitory, but the last three pairs
frequently persist to form the genitalia, as
in Orthoptera (to which order (Ecanthus
belongs). In Collembola, the embryo has
paired abdominal limbs, and those of the
first abdominal segment eventually unite
to form the peculiar ventral tube (Fig. 12)
of these insects, while those of the fourth
segment form the characteristic leaping
organ, orfurcula.

Cerci. In many of the more generalized
insects the abdomen bears at its extremity
two or three appendages termed cerci . These

occur in both sexes and are frequently long and multiarticulate, as in Thy-
sanura (Figs. 76, 9, 10) and Ephemerida (Figs. 19, B.; 84), though shorter
in cockroaches and reduced to a single sclerite in Acridiidae (Fig. 87). The
paired cerci, or cercopoda of Packard, are usually though not always as-
sociated with the tenth abdominal segment and are homologous with
legs, as Ayers has found in (Ecanthus and Wheeler in Xiphidium, As to
their function, the cerci of Thysanura are tactile, and those of the cock-

c

FIG. 76. Ventral aspect of the
abdomen of a female Machilis
iiidi-ilima, to show rudimentary
limbs (a) of segments two to nine.
(The left appendage of the eighth
segment is omitted.) c, c, c, cerci.
After OUDEMANS.

54

ENTOMOLOGY

roach olfactory, while the cerci of male Acridiidse often serve to hold the
female during copulation.

Extremity of Abdomen. Various modifications of the terminal
segments of the abdomen occur for the purposes of defaecation and es-
pecially reproduction. The anus, dorsal in position, opens always through
the last segment and is often shielded above by a suranal plate and on each
side by a lateral plate. The genital orifice is always ventral in position
and occurs commonly on the ninth abdominal segment, though there is
some variation in this respect. The external, or accessory, organs of. re-
production are termed the genitalia.

Female Genitalia. In Neuroptera, Coleoptera, Lepidoptera and
Diptera the vagina simply opens to the exterior or else with the anus into
a common chamber, or cloaca. Often, as in Cerambyx (Fig. 77) and
Cecidomyia (Fig. 78) the attenuated distal segments of the abdomen serve

the purpose of an ovipositor;
6 thus in Cecidomyiidae, the ter-

r minal segments, telescoped into

FIG. 77. Abdomen of female beetle, Cer-
ambyx, in which the last three segments are
used as an ovipositor. After KOLBE.

FIG. 78. Abdomen of a female midge,
Dasyneiira legnminicola, to show the
pseudo-ovipositor.

one another when not in use, form when extruded a lash-like organ ex-
ceeding frequently the remainder of the body in length.

A true ovipositor occurs in Thysanura, Orthoptera, Odonata, Hemip-
tera, Hymenoptera and some other orders of insects. The ovipositor con-
sists essentially of three pairs of valves, or gonapophyses a dorsal, a
ventral and an inner pair. The two inner valves form a channel through
which the eggs are conveyed. In Locustidae (Fig. 79) the three valves of
each side are held together by tongues and grooves, which, however,
permit sliding movements to take place. Most authorities have found
that the gonapophyses belong to the segmental series of paired appen-
dagesare homodynamous with limbs and pertain commonly to ab-
dominal segments seven, eight and nine.

The ovipositor attains its greatest complexity in Hymenoptera, in
which it becomes modified for sawing, boring or stinging. In Sir ex (Fig.
80) the inner valves are united; in Apis the dorsal valves are repre-
sented by a pair of palpi, the inner valves unite to form the sheath

ANATOMY AND PHYSIOLOGY

55

(Fig. 81, B), and the ventral two form the darts, each of which has ten
barbed teeth behind its apex, which tend to prevent the withdrawal of
the sting from a wound. The action of the sting, as described by Che-

FIG. 79. Ovipositor of Locusta. A, lateral aspect; B, ventral aspect; C, transverse sec-
tion; c, cerci; d, dorsal valve; i, inner valve; v, ventral valve. The numbers refer to ab-
dominal segments. After KOLBE and DEWITZ.

shire, is rather complex. Briefly, the sheath serves to open a wound and
to guide the darts; these strike in alternately, interrupted at intervals
by the deeper plunging of the sheath (Fig. Si, A). The poison of the
honey bee is secreted by two glands, one acid and
the other alkaline. The former (Fig. 82) consists
of a glandular region which secretes formic acid, of
a reservoir, and a duct that empties its contents
into the channel of the sheath. The alkaline gland
also opens into the reservoir. It is said that both
fluids are necessary for a deadly effect ; and that in
insects which simply paralyze their prey, as the
solitary wasps, the alkaline glands are functionless.
Male Genitalia. The penis may be hollow or
else solid, and in the latter case the contents of
the ejaculatory duct are spread upon its surface.
Morphologically, the male gonapophyses correspond to those of the
female. The penis (Fig. 83) represents the two inner valves of the ovi-

FIG. 80. Cross-sec-
tion of the ovipositor of
Sirex. c, channel; d,
d, dorsal valves; i,
united inner valves; v,
v, ventral valves. After
TASCHENBERG.

ENTOMOLOGY

positor and is frequently enclosed by one or two pairs of valves. In
Ephemerida the two inner valves are partly or entirely separate from each
other, forming two intromittent organs (Fig. 84).

In male Odonata, the ejaculatory duct opens on the ninth abdominal
segment, but the copulatory organ is placed on the under side of the sec-
ond segment, to wh>ch the spermatozoa are transferred by the bending
of the abdomen. At copulation, the abdominal claspers of the male
grasp the neck of the female, and the latter bends her abdomen forward
until the tip reaches the peculiar copulatory apparatus of the male.

The claspers of the male consist of a single pair, variously formed.
They are present in Ephemerida, Neuroptera, Trichoptera, Lepidoptera
(Fig. 85), Diptera and some Hymenop-
tera, though not in Coleoptera, and
often afford good specific charactefs, as

FIG. Si. Sting of honey bee. A, i, 2, j, posi-
tions in three successive thrusts; s, sheath. B,
cross-section; c, channel; i, united inner valves,
forming the sheath; v, v, ventral valves, or darts.
A, after CHESHIRE; B, after FENCER.

FIG. 82. Sting and poison appara-
tus of honey bee. ag, accessory gland;
p, palpus; pg, poison gland (formic
acid); r, reservoir; s, sting. After
KRAEPELIN.

in Odonata. In butterflies of the genus Thanaos, the claspers are peculiar
in being strongly asymmetrical. In Odonata (Fig. 86, A) and Orthoptera
(Fig. 87, A) the superior appendages of the male often serve as claspers.
In many insects the tergum of the last abdominal segment forms a
small suranal plate (Fig. 87, B, sp) ; this sometimes supplements the clasp-
ers of the male in their function, as in Lepidoptera (Fig. 85, A, s).

2. INTEGUMENT

Insects excel all other animals in respect to adaptive modifications of
the integument. No longer a simple limiting membrane, the integument

ANATOMY AND PHYSIOLOGY

57

has become hardened into an external skeleton, evaginated to form mani-
fold adaptive structures, such as hairs and scales, and invaginated, along
with the underlying cellular layer, to make glands of various kinds.

g

FIG. 83. Extremity of abdomen of a FIG. 84. Extremity of abdomen of a

male beetle, Hydro phi I us, ventral aspect, g, male May fly, Hcxagciiut I'ariabilis, ventral

genitalia; p, penis; v 1 , v' 2 , pairs of valves aspect, c, c', c, cerci; d, d, claspers; i, i,

enclosing the penis; 6-g, sterna of abdom- intromittent organs,
inal segments. After KOLBE.

Chitin. The skin, or cuticula, 1 of an insect differs from that of a
worm, for example, in being thoroughly permeated with a peculiar sub-

8

'P

v

B

FIG. 85. Genitalia of a moth, Saw /a cccropia. A, male; B, female; a, anus; c, c, claspers;
o, opening of common oviduct; p, penis; s, uncus (the doubly hooked organ); i>, vestibule,
into which the vagina opens. The numbers refer to abdominal segments.

stance known as chitin the basis of the arthropod skeleton. This is a
substance of remarkable stability, for it is unaffected by almost all ordi-
nary acids and alkalies, though it is soluble in sodic or potassic hypo-

x The ciiliciila of an insect should be distinguished from the cuticle of a vertebrate, the
former being a hardened fluid, while the latter consists of cells themselves, in a dead and
flattened condition.

ENTOMOLOGY

chlorite (respectively, Eau de Labarraque and Eau de Javelle) and yields
to boiling sulphuric acid. If kept for a year or so under water, however,
chitin undergoes a slow dissolution, possibly a putrefaction, which ac-
counts in a measure for the rapid disappearance of insect skeletons in the
soil (Miall and Denny). By boiling the skin of an insect in potassic hy-
droxide it is possible to dissolve away the cuticular framework, leaving

JO

B

FIG. 86. Terminal abdominal appendages of a dragon fly, Plathemis trimaculata. A,
male; B, female, i, inferior appendage; s, s, superior appendages. The numbers refer to
abdominal segments.

fairly pure chitin, without destroying the organized form of the integu-
ment, though less than half the weight of the integument is due to chitin.
The formula of chitin is given as C 9 Hi 5 NO 6 or Ci 8 Hi 5 NOi2 byKruken-
berg, and Packard adopts the formula CisH^e^Oio; though no two chem-
ists agree &s to the exact proportions of these elements, owing probably
to variations in the substance itself in different insects or even in the same

8g 10

g 10 n

FIG. 87. Extremity of the abdomen of a grasshopper, Melanophis di/erentialis. A, male;
B, female. The terga and sterna are numbered, c, cercus; d, dorsal valves of ovipositor; e,
egg guide; p, podical plate; s, spiracle; sp, suranal plate; D, ventral valves of ovipositor.

species of insect. Iron, manganese and certain pigments also enter into
the composition of the integument.

Chitin is not peculiar to arthropods, for it has been detected in the
setae and pharyngeal teeth of annelid worms, the shell of Lingula and the
pen of the cuttle fish (Krukenberg) .

The chitinous integument (Fig. 88) of most insects consists of two
layers: (i) an outer layer, homogeneous, dense, without lamellae or pore

ANATOMY AND PHYSIOLOGY

59

C'

h

canals, and being the seat of the cuticular colors; (2) an inner layer,
"thickly pierced with pore canals, and always in layers of different re-
fractive indices and different stainability." (Tower.) These two layers,
respectively primary and secondary cuticula, are radically different in
chemical and physical properties. The chitinous cuticula is secreted, as
a fluid, from the hypodermis cells. Each layer arises as a fluid secretion
from the hypodermis cells, the primary cuticula being the first to form and
harden.

The fluid that separates the old from the new cuticula at ecdysisis
poured over the hypodermis by certain large special cells, which, accord-
ing to Tower, "are not true glands, but the setigerous cells which, in early
life, are chiefly concerned with the formation of the hairs upon the body;
but upon the loss of these, the cell takes on the function of secreting the
exuvial fluid, which is most copious at
pupation. These cells degenerate in the
pupa, and take no part in the formation
of the imaginal ornamentation."

Histology. The chitinous cuticula
owes its existence to the activity of the
underlying layer of hypodermis cells (Fig.
88). These cells, distinct in embryonic
and often in early larval life, subse-
quently become confluent by the disap-
pearance of the intervening cell walls,
though each cell is still indicated by its
nucleus. The cells are limited outwardly
by the cuticula and inwardly by a deli-
cate, hyaline basement membrane; they contain pigment granules, fat-
drops, etc.

Externally the cuticula may be smooth, wrinkled, striate, granulate,
tuberculate, or sculptured in numberless other ways; it may be shaped
into all manner of structures, some of which are clearly adaptive, while
others are unintelligible.

Hairs, Setae and Spines. These occur universally, serving a great
variety of purposes; they are not always simple in form, but are often
toothed, branched or otherwise modified (Fig. 89). Hairs and bristles
are frequently tactile in function, over the general integument or else
locally; or olfactory, as on the antennae of moths; or occasionally audi-
tory, as on the antennas of the male mosquito; these and other sensory
modifications are described beyond. The hairy clothing of some hiber-

FIG. 88. Section through integu-
ment of a beetle, Chrysobothris. b,
basement membrane; c 1 , primary
cuticula; c 2 , secondary cuticula; h,
hypodermis cell; n, nucleus. After
TOWER.

6o

ENTOMOLOGY

nating caterpillars (as Isia Isabella) probably protects them from sudden
changes of temperature. Hairs and spines frequently protect an insect

\

B

D

FIG. 89. Modifications of the hairs of bees. A, B, Megachlle; C, E, F, Colldes; D, Chelos-

toma, After SAUNDERS.

from its enemies, especially when these structures are glandular and emit
a malodorous, nauseous or irritant fluid. Glandular hairs on the pulvilli

-_-Jt

tiii

I

FIG. 90. Section of antenna of a moth, FIG. 91. Radial section through the

Satitrnia, to show developing hairs, c, cutic- base of a hair of a caterpillar, Picris rapce.

ula; /, formative cell of hair; h, hypodermis; c, cuticula; /, formative cell; //, hair; hy,

t, trachea. After SEMPER. hypodermis.

of many flies, beetles, etc., enable these insects to walk on slippery sur-
faces. The twisted or branched hairs of bees serve to gather and hold

ANATOMY AND PHYSIOLOGY

6l

pollen grains; in short, these simple structures exhibit a surprising va-
riety of adaptive modifications, many of which will be described in con-
nection with other subjects.

A hair arises from a modified hypodermis cell (Fig. 90), the contents of
which extend through a pore canal into the
interior of the hair (Fig. 91); sometimes, to
be sure, as in glandular or sensory hairs, the
hair cell is multinucleate, representing, there-
fore, as many cells as there are nuclei. The
wall of a hair is continuous with the general
cuticula and at moulting each hair is stripped
off with the rest of the cuticula, leaving in
its place a new hair, which has been forming
inside the old one.

Scales. Besides occurring throughout the
order Lepidoptera and in numerous Trichop-
tera, scales are found in many Thysanura and
Collembola, several families of Coleoptera
(including Dermestidae and Curculionidae), a
few Diptera and a few Psocida?.

Though diverse in form (Fig. 92), scales
are essentially flattened sacs having at one
end a short pedicel for attachment to the
integument. The scales usually bear mark-
ings, which are more or less characteristic of
the species; these markings, always minute,

are in some species so exquisitely fine as to test the highest powers of
the microscope; the scales of certain Collembola (Lepidocyrtus, etc.) have
long been used, under the name of "Podura" scales, to test the resolving
power of objectives, for which purpose they are excelled only by some of

the diatoms. Butterfly scales are
marked with parallel longitudinal
ridges (Fig. 92, C), which are confined
almost entirely to the upper, or ex-
posed, surface of the scale (Fig. 93)
and number from 33 or less (Anosia)
to 1,400 (Morpho) to each scale, the stride being from .002 mm. to .0007
mm. apart (Kellogg) ; between these longitudinal ridges may be dis-
cerned delicate transverse markings. Internally, scales are hollow and
often contain pigments derived from the blood.

FIG. 92. Various forms of
scales. A, R, thysanuran,
Machilis; B, beetle. Anthrenus,
C, butterfly, Picris; D, moth,
Li in a codes.

FIG. 93. Cross-section of scale of
.1 Ho.v/ii. After MAYER.

62

ENTOMOLOGY

On the wing of a butterfly the scales are arranged in regular rows and
overlap one another, as in Fig. 94; in the more primitive moths and in
Trichoptera, however, their distribution is rather irregular.

A scale is the equivalent of a hair, for (i) a complete series of transi-
tions from hairs to scales may be
found on a single individual (Fig. 95) ;
and (2) hairs and scales agree in their
manner of development, as shown by
Semper,' Schaffer, Spuler, Mayer and
others. Both hairs and scales arise
as processes from enlarged hypo-
dermis cells, or formative cells (Fig.
96). The scale at first contains pro-
toplasm, which gradually withdraws,
leaving short chitinous strands to
hold the two membranes of the scale
together.

Uses of Scales. Among Thy-
sanura and Collembola, scales occur
only on such species as live in com-
paratively dry situations, from which

it may be inferred that the scales serve to retard the evaporation of mois-
ture through the delicate integument of these insects. This inference is
supported by the fact that none of the scaleless Collembola can live long
in a dry atmosphere; they soon shrivel and die even under conditions
of dryness which the scaled species are able to withstand. In Lepidop-

FIG. 94. Arrangement of scales on the
wing of a butterfly, Papilio.

\

u/
^

/I

I]

FIG. 95. Hairs and scales of a moth, Samia cecropia.

tera the scales are possibly of some value as a mechanical protection;
they have no influence upon flight, as Mayer has proved, and appear to
be useful chiefly as a basis for the development of color and color patterns
which are not infrequently adaptive.

ANATOMY AND. PHYSIOLOGY 63

Androconia. The males of many butterflies, and the males only,
have peculiarly shaped scales known as androconia (Fig. 97); these are
commonly confined to the upper surfaces of the front wings, where they
are mingled with the ordinary scales or else are disposed in special patches
or under a fold of the costal margin of the wing (Thanaos). The char-
acteristic odors of male butterflies have long been attributed to these
androconia, and M. B. Thomas has found that the scales arise from glan-

B

B

FIG. 96. Development of butterfly scales.
A. Vanessa; B, Anosia. b, basement mem-
brane; /, formative cell; h, hypodermis; s,
scale. After MAYER.

FIG. 97. Androconia of butterflies. A,
Pieris rapcc; B, Everes comyntas.

dular cells, which doubtless secrete a fluid that emanates from the scale
as an odorous vapor, the evaporation of the fluid being facilitated by the
spreading or branching form of the androconium. Similar scales occur
also on the wings of various moths and some Trichoptera (Mystacides).

Glands. A great many glands of various form and function have
been found in insects. Most of these, being formed from the hypoder-
mis, may logically be considered here, excepting some which are inti-
mately concerned with digestion or reproduction.

6 4

ENTOMOLOGY

Glandular Hairs and Spines. The presence of adhesive hairs on
the empodium of the foot of a fly enables the insect to walk on a smooth
surface and to walk upside down; these tenent hairs emit a transparent
sticky fluid through minute pore canals in their apices. The tenent
hairs of Hylobius (Fig. 98) are each supplied with a flask-shaped unicellu-
lar gland, the glutinous secretion of which issues from the bulbous ex-
tremity of the hair. Bulbous tenent hairs occur also on the tarsi of Col-
lembola, Aphididae and other insects.

Nettling hairs or spines clothe the caterpillars of certain Saturniidae
(Automeris), Liparidae, etc. These spines (Fig. 99), which are sharp,
brittle and filled with poison, break to pieces when the insect is handled
and cause a cutaneous irritation much like that made by nettles. In

FIG. 98. Section across tarsus of a beetle,
Hylobius, to show bulbous glandular hairs.

After SlMMERMACHER.

P'iG. 99. Stinging hair of a caterpillar,
Gastro pacha, c, cuticula; g, gland cell; h,
hair; hy, hypodermis. After CLAUS.

Lagoa crispata (Fig. 100) the irritating fluid is secreted, as is usual, by
several large hypodermal cells at the base of each spine. These irritating
hairs protect their possessors from almost all birds except cuckoos.

Repellent Glands. The various offensive fluids emitted by insects
are also a highly effective means of defence against birds and other in-
sectivorous vertebrates as well as against predaceous insects. The blood
itself serves as a repellent fluid in the oil-beetles (Meloidae) and Coccinel-
lidas, issuing as a yellow fluid from a pore at the end of the femur. The
blood of Meloidae (one species of which is still used medicinally under the
name of " Spanish Fly") contains cantharidine, an extremely caustic sub-
stance, which is an almost perfect protection against birds, reptiles and
predaceous insects. Coccinellidae and Lampyridae are similarly exempt

ANATOMY AND PHYSIOLOGY

from attack. Larvae of Cimbex when disturbed squirt jets of a watery
fluid from glands opening above the spiracles. Many Carabidae eject a
pungent and often corrosive fluid from a pair of anal glands (Fig. 146) ;
this fluid in Brachinus, and occasionally in Galerita janus and a few other
carabids, volatilizes explosively upon contact with the
air. When one- of these "bombardier-beetles" is
molested it discharges a puff of vapor, accompanied
by a distinct report, reminding one of a miniature
cannon, and this performance may be repeated several
times in rapid succession; the vapor is acid and corro-
sive, staining the human skin a rust-red color. Indi-
viduals of a large South American Brachinus when
seized "immediately began to play off their artillery,
burning and staining the flesh to such a degree that
only a few specimens could be captured with the naked
hand, leaving a mark which remained for a consider-
able time." (Westwood.)

As malodorous insects, Hemiptera are notorious,

though not a few hemipterous odors are (apart from their associations)
rather agreeable to the human olfactory sense. Commonly the odor is
due to a fluid from a mesothoracic gland or glands, opening between the
hind coxae.

Eversible hypodermal glands of many kinds are common in larvae of

PIG. ioo. Sting-
ing spines of a cater-
pillar, Lagoa crispata.
After PACKARD.

FIG. 10 1. Osmeterium of Papilla
polyxenes.

FIG. 102. Ventral aspect of worker honey bee, show-
ing the four pairs of wax scales. After CHESHIRE.

Coleoptera and Lepidoptera. The larvae of Melasoma lapponica, among
other Chrysomelidae, evert numerous paired vesicles which emit a peculiar
odor. The caterpillars of our Papilio butterflies, upon being irritated,
evert from the prothorax a yellow Y-shaped osmeterium (Fig. 101) which
6

66

ENTOMOLOGY

diffuses a characteristic but indescribable odor that is probably repellent.
The larva of Cerura everts a curious spraying apparatus from the under
side of the neck.

Alluring Glands. Odors are largely used among insects to attract
the opposite sex. The androconia of male butterflies have already been
spoken of. Males of Catocala cjoncumbens disseminate an alluring odor
from scent tufts on the middle legs. Female saturniid moths (as cecro-
pia and promethea) entice the males by means of a characteristic odor
emanating from the extremity of the abdomen. In lycaenid caterpillars,
an eversible sac on the dorsum of the seventh abdominal segment secretes
a sweet fluid, for the sake of which these larvae are sought out by ants.

Wax Glands. Wax is secreted by insects of several orders, but es-
pecially Hymenoptera and Hemiptera. In the worker honey bee the

wax exudes from unicellular hypo-
dermal glands and appears on the
under side of the abdomen as four
pairs of wax scales (Fig. 102).
Plant lice of the genus Schizoneura
owe their woolly appearance to
dense white filaments of wax,
which arise from glandular hypo-
dermal cells. In scale insects,
waxen threads, emerging from
cuticular pores, become matted

FIG. 103. Head of caterpillar of Samia
ccc ropia. a, antenna; c, clypeus; /, labrum;
Ip, labial palpus; m, mandible; -nip, maxillary
palpi; o, ocelli; s, spinneret.

together to form a continuous

shield over and often under the
insect itself, the cast skins often
being incorporated into this waxen

scale. The wax glands in Coccidae are simply enlarged hypodermis
cells.

Silk Glands. Larvae of very diverse orders spin silk, for the purpose
of making cocoons, webs, cases, and supports of one kind or another.
Silk glands, though most characteristic of Lepidoptera and Trichoptera,
occur also in the cocoon-spinning larvae of not a few Hymenoptera (saw
flies, ichneumons, wasps, bees, etc.), in Diptera (Cecidomyiidae), Neurop-
tera (Chrysppidae, Myrmeleonidae) , and in various larvae whose pupae are
suspended from a silken support, as in the coleopterous families Coccinel-
lidae and Chrysomelidae (in part) and the dipterous family Syrphidae, as
well as most diurnal Lepidoptera.

The silk glands of caterpillars are homologous with the true salivary

ANATOMY AND PHYSIOLOGY

glands of other insects, opening as usual through the hypopharynx,
which is modified to form a spinning organ, or spinneret (Fig. 103). The
silk glands of Lepidoptera are a pair of long tubes, one on each side of the
body, but often much longer than the body and consequently convoluted.
Thus in the silk worm (Bombyx mori} they are from four to five times as
long as the body and in Tdca polyphemus, seven times as long. In the

silk worm the convoluted glandular portion
of each tube (Fig. 104) opens into a dilata-
tion, or silk reservoir, which in turn empties
into a slender duct, and the two ducts ioin

FIG. 104. Silk glands of the
silk worm, Bombyx mori. cd,
common duct; d, one of the
paired ducts; g, g, Filippi's After HELM.
glands; g/, gland proper; />,
thread press; r, reservoir.

FIG. 105. Sections of silk gland of the silk worm.
A, radial; B, transverse, b. basement membrane; /,
intima; ,v, glandular cell with branched nucleus.

into a short common duct, which passes through the tubular spinneret.
Two divisions of the spinning tube are distinguished: (i) a posterior
muscular portion, or thread- press and (2) an anterior directing tube. The
thread-press combines the two streams of silk fluid into one, determines
the form of the silken thread and arrests the emission of the thread at
times, besides having other functions. The silk fluid hardens rapidly
upon exposure to the air; about fifty per cent, of the fluid is actual silk

68

ENTOMOLOGY

substance and the remainder consists of protoplasm and gum, with traces
of wax, pigment, fat and resin.

A transverse or radial section of a silk gland shows a layer of glandu-
lar epithelial cells, with the usual intima and basement membrane (Fig.
105); the cells are remarkably large and their nuclei are often branched;
the intima is distinctly striated, from the presence of pore-canals. The
glands arise as evaginations of the pharynx (ectodermal) and the chi-
tinous intima of each gland is cast at each moult, along with the general
integument.

The silk glands of Trichoptera are essentially like those of Lepidop-
tera, but the glands of Chrysopa, Myrmeleon, Coccinellidae, Chrysomeli-
dae and Syrphidae, which open into the rectum, are morphologically quite
different from those of Lepidoptera.

3. MUSCULAR SYSTEM

The number of muscles possessed by an insect is surprisingly large.
A caterpillar, for example, has about two thousand.

The muscles of the trunk are segmentally arranged most evidently

ts-

abc

-is

FIG. 106.

FIG. 107.

FIG. 108.

-Muscles of cockroach; of ventral, dorsal and lateral walls, respectively, a, alary muscle;
abc, abductor of coxa; adc, adductor of coxa; ef, extensor of femur; h, head muscles; Is,
longitudinal sternal; It, longitudinal tergal; ////, lateral thoracic; os, oblique sternal; ot,
oblique tergal; ts, tergo-sternal; ts l , first tergo-sternal. After MIALL and DENNY.

so in the body of a larva or the abdomen of an imago, where the muscu-
lature is essentially the same in several successive segments. In the
thoracic segments of an imago, however, the musculature is, at first sight,

ANATOMY AND PHYSIOLOGY

6 9

al

-
SSP

FIG. 109.
Striated muscle
fiber of an insect.

unlike that of the abdomen, and in the head it is decidedly different;
though future studies will doubtless show that the thoracic and cephalic
kinds of musculature are only modifications of the simpler abdominal
type modifications brought about in relation to the
needs of the legs, wings, mouth parts, antennas and other
movable structures.

The muscular system has been generally neglected
by students of insect anatomy; the only comprehensive
studies upon the subject being those of Straus-Diirckheim
(1828) on the beetle Melolontha; Lyonet (1762), New-
port (1834) and Lubbock (1859) on caterpillars; and
the more recent studies of Lubbock and Janet on Hy-
menoptera.

The more important muscles in the body of a cock-
roach are represented in Figs. 106-108, from Miall and
Denny. The longitudinal sternals with the longitudinal tcrgals act to
telescope the abdominal segments; the oblique sternals bend the abdomen
laterally; the tergo-sternals, or vertical expiratory muscles, draw the

tergum and sternum together. The
muscles of the legs and the wings have
already been referred to.

Structure of Muscles. The mus-
cles of insects differ greatly in form and
are inserted frequently by means of
chitinous tendons. A muscle is a
bundle of long fibers, each of which
has an outer elastic membrane, or
sarcolemma, within which are several
nuclei; thus the fiber represents several
cells, which have become confluent.
With rare exceptions (" alary" muscles
and possibly a few thoracic muscles)
the muscle fibers of an insect present a
striated appearance, owing to alternate

light and dark bands (Fig. 109), the former being singly refracting, or
isotropic, and the latter doubly refracting, or anisotropic.

The minute structure of these fibers, being extremely difficult of
interpretation, has given rise to much difference of opinion. The most
plausible view is that of van Gehuchten, Janet and others, who hold that
both kinds of dark bands (Fig. no) consist of highly elastic threads of

FIG. no. Minute structure of a
striated muscle fiber. A, longitudinal
section; B, transverse section in the
region of /; C, transverse section in the
region of 11. I, longitudinal fibrilke;
11, Krause's membrane; ;//, nucleus;
r, radial fibrilke; s, sarcolemma. After
JANET.

70 ENTOMOLOGY

spongioplasm (anisotropic) embedded in a matrix of clear, semi-fluid,
nutritive hyaloplasm (isotropic). The spongioplasmic threads of the
long bands extend longitudinally and those of the short bands ("Krause's
membrane") radially, in respect to the form of the fiber. Moreover, the
attenuated extremities of the longitudinal fibrillae connect with the radial
fibrillae, the points of connection being marked by slight thickenings, or
nodes, which go to make up Krause's membrane.

Under nervous stimulus a muscle shortens and thickens because its
component fibers do, and this in turn is attributed to the shortening and
thickening of the longitudinal fibrillae. When the stimulus ceases, the
radial fibrillae, by their elasticity, possibly pull the longitudinal ones back
into place. The last word has not been said, however, upon this per-
plexing subject.

Muscular Power. The muscular exploits of insects appear to be
marvellous beside those of larger animals, though they are often exag-
gerated in popular writings. The weakest insects, according to Plateau,
can pull five times their own weight and the average insect, over twenty
times its weight, while Donacia (Chrysomelidae) can pull 42.7 times its
weight. As contrasted with these feats, a man can pull in the same fash-
ion but .86 of his weight and a horse from .5 to .83. How are these dif-
ferences explained?

It is incorrect to say that the muscles of insects are stronger than those
of vertebrates, for, as a matter of fact, the contractile force of a vertebrate
muscle is greater than that of an insect muscle, other things being equal.
The apparently greater strength of an insect in proportion to its weight is
accounted for in several ways. The specific gravity of chitin is less than
that of bone, though it varies greatly in both substances. Furthermore,
the external skeleton permits muscular attachments of the most advan-
tageous kind as compared with the internal skeleton, so that the muscles
of insects surpass those of vertebrates as regards leverage. These reasons
are only of minor importance, however. Small animals in general appear
to be stronger than larger animals (allowing for the differences in weight)
for the same reason that a smaller insect has more conspicuous strength
than a larger one, when the two are similar in everything except weight.
For example: where a bumble bee can pull 16.1 times its own weight, a
honey bee can pull 20.2 ; and w r here the same bumble bee can carry w r hile
flying a load 0.63 of its own weight, the honey bee can carry 0.78. Al-
ways, as Plateau has shown, the lighter of two insects is the stronger in
respect to external manifestations of muscular force in the ratio of this
muscular strength to its own weight.

ANATOMY AND PHYSIOLOGY 71

To understand this, let us assume that a beetle continues to grow (as
never happens, of course). As its weight is increasing so is its strength-
but not in the same proportion. For while the weight say that of a
muscle increases as the cube of a single dimension, the strength of the
muscle (depending solely upon the area of its cross-section) is increasing
only as the square of one dimension its diameter. Therefore the in-
crease in strength lags behind that of weight more and more; consequently
more and more strength is required simply to move the insect itself, and
less and less surplus strength remains for carrying additional weight.
Thus the larger insect is apparently the weaker, though it is actually the
stronger, in that its total muscular force is greater.

The writer uses this explanation to account also for the inability of
certain large beetles and other insects to use their wings, though these
organs are well developed. Increasing weight (due to a larger supply of
reserve food accumulated by the larva) has made such demands upon the
muscular power that insufficient strength remains for the purpose of flight.

Statements such as this are often seen a flea can jump a meter, or
six hundred times its own length. Almost needless to say, the length of
the'body is no criterion of the muscular power of an animal.

4. NERVOUS SYSTEM

The central nervous system extends along the median line of the floor
of the body as a series of ganglia connected by nerve cords. Typically,
there is a ganglion (double in origin) for each primary segment, and the
connecting cords, or commissures, are paired; these conditions are most
nearly realized in embryos and in the most generalized insects Thysa-
nura (Fig. in). In all adult insects, however, the originally separate
ganglia consolidate more or less (Fig. 112) and the commissures frequently
unite to form single cords. Thus in Tabanus (Fig. 112, C) the three
thoracic ganglia have united into a single compound ganglion and the
abdominal ganglia are concentrated in the anterior part of the abdomen;
in the grasshopper, the nerve cord, double in the thorax, is single in the
abdomen. Various other modifications of the same nature occur.

Cephalic Ganglia. In the head the primitive ganglia always unite
to form two compound ganglia, namely, the brain and the suboesophageal
ganglion (disregarding a few anomalous cases in which the latter is said
to be absent).

The brain, or supracesophagcal ganglion (Fig. 113), is formed by the
union of three primitive ganglia, or neuromeres (Fig. 55), namely, (i)
the protocerebrum, which gives off the pair of optic nerves; (2) the deuto-

ENTOMOLOGY

i \VX

t> --*$*-'

ol

II

in

SV--JT;

FIG. in. Central
nervous system of a thy-
sanuran, Macliilis. The
thoracic and abdominal
ganglia are numbered in
succession. a, antennal
nerve; b, brain; e, com-
pound eye; /, labial nerve;
m, mandibular nerve;
;w.v, maxillary nerve; o,
oesophagus; ol, optic lobe;
s, suboesophageal gang-
lion ; sy, sympathetic
nerve. After OUDEMANS.

cerebrum, which innervates the antennae; and (3)
the tritocerebrum, which in Apterygota bears a pair
of rudimentary appendages that are regarded as
traces of a second pair of antennae.

The suboesophageal ganglion (Fig. 113) is
always connected with the brain by a pair of
nerve cords (cesophageal commissures) between
which the oesophagus passes. This compound
ganglion represents at most four neuromeres: (i)
mandibular, innervating the mandibles; (2) super-
lingual, found by the author in Collembola, but
not yet reported in the less generalized insects;
(3) maxillary, innervating the maxillae; (4) labial,
which sends a pair of nerves to the labium.

The minute structure of the brain, though
highly complex, has received considerable study,
but will not be described here for the reason that
the anatomical facts are of no general interest so
long as their physiological interpretation remains
obscure.

Sympathetic System. Lying along the me-
dian dorsal line of the oesophagus is a recurrent,
or stomatogastric, nerve (Fig. 114), which arises
anteriorly in a frontal ganglion and terminates
posteriorly in a stomachic ganglion situated at the
anterior end of the mid intestine. Connected
with the recurrent nerve are two pairs of lateral
ganglia, the anterior of which innervate the dorsal
vessel and the posterior, the tracheae of the head.
The ventral nerve cord may include also a median
nerve thread (Fig. in) which gives off paired
transverse nerves to the muscles of the spiracles.

Structure of Ganglia and Nerves. A gang-
lion consists of (i) a dense cortex, composed of
ganglion cells (Fig. 115), each of which has a
large rounded nucleus and gives off usually a
single nerve fiber; and (2) a clear medullary
portion (Punktsubstanz) derived from the pro-
cesses of the cortical ganglion cells and serving as
the place of origin of nerve fibrillae. There are,

ANATOMY AND PHYSIOLOGY

73

however, ganglion cells from which processes may pass directly into
nerve fibrillae.

A nerve-fiber, in an insect, consists of an axis-cylinder, composed of

A

D

FIG. 112. Successive stages in the concentration of the central nervous system of Diptera.
A, CliiroitoniHs; B, Empis; C, Tabaiius; D, Sarcophaga. After BRANDT.

fibrilla?, and an enveloping membrane, or sheath. The axis-cylinder is
the transmitting portion and the ganglia are the trophic centers, i. e.,
they regulate nutrition. A nerve is always either sensory, transmitting

FlG. 113. Nervous system of the head of a cockroach, a, antennal nerve; ag, anterior
lateral ganglion of sympathetic system; b, brain; d, salivary duct; /, frontal ganglion; //,
hypopharynx; /, labrum; //, labium; ;, mandibular nerve; mx, maxillary nerve; nl, nerve to
labrum; ;///, nerve to labium; o, optic nerve; oc, cesophageal commissure; oc, cesophagus;
pg, posterior lateral ganglion of sympathetic system; r, recurrent nerve of sympathetic system;
s, subcesophageal ganglion. After HO.FER.

impulses inward from a sense organ ; or else motor, conveying stimuli from

the central nervous system outward to muscles, glands, or other organs.

Functions. The brain innervates the chief sensory organs (eyes and

74

ENTOMOLOGY

antennae) and converts the sensory stimuli that it receives into motor

stimuli, which effect co-ordinated muscular
or other movements in response to particular
sensations from the environment. The brain
is the seat of the will, using the term ''will " in
a loose sense; it directs locomotor movements
of the legs and wings. An insect deprived of
its brain cannot go to its food, though it is
able to eat if food be placed in contact with
the end-organs of taste, as those of the palpi;
furthermore, it walks or flies in an erratic
manner, indicating a lack of co-ordination of
muscular action.

The suboesophageal ganglion controls the
mouth parts, co-ordinating their movements
as well as some of the bodily movements.

The thoracic ganglia govern the appen-
dages of their respective segments. These
ganglia and those of the abdomen are to a
great extent independent of brain control,
each of these ganglia being an individual
motor center for its particular segment.
Thus decapitated insects are still able to
breathe, walk or fly, and often retain for
several days some power of movement.

In regard to the sympathetic system, it
has been shown experimentally that the frontal
ganglion controls the swallowing movements
and exerts through the stomatogastric nerve
a regulative action upon digestion. The
dorsal sympathetic system controls the dorsal vessel and the salivary

FIG. 114. Sympathetic
nervous system of an in-
sect, diagrammatically repre-
sented, a, antennal nerve;
b, brain; /, frontal ganglion;
/, /, paired lateral ganglia; m,
nerves to upper mouth parts;
o, optic nerve; r, recurrent
nerve; s, nerve to salivary
glands; st, stomachic gang-
lion. After KOLBE.

n

FIG. 115. Transverse section of an abdominal ganglion of a caterpillar. /, nerve-fibers; g,
ganglion cells; n, nerve-sheath; j>, Punktsubstanz.

ANATOMY AND PHYSIOLOGY 75

glands, while the ventral sympathetic system is concerned with the spi-
racular muscles.

5. SENSE ORGANS

For the reception of sensory impressions from the external world,
the armor-like integument of insects is modified in a great variety of
ways. Though sense organs of one kind or another may occur on almost
any part of an insect, they are most numerous and varied upon the head
and its appendages, particularly the antennae.

Antennal Sensilla. Some idea of the diversity of form in antennal
sense organs may be obtained from Figs. 116-125, taken from a paper
by Schenk, whose useful classification of antennal sensilla, or sense
organs, is here outlined:

1. Sensillum cceloconicum a conical or peg-like projection immersed
in a pit (Figs. 116-117). I n an< probability olfactory.

2. S. basiconicum a cone projecting above the general surface (Fig.
1 1 8). Probably olfactory.

3. S. styloconicum a terminal tooth or peg seated upon a more or less
conical base (Fig. 119). Olfactory.

4. S. chaticum a bristle-like sense organ (Fig. 120). Tactile.

5. S. trichodeum a hair-like sense organ (Figs. 121, 122). Tactile.

6. 6". placodeum a membranous plate, its outer surface continuous
with the general integument (Fig. 123). Function doubtful; not audi-
tory and probably not olfactory, though the function is doubtless a
mechanical one; Schenk suggests that this organ is affected by air
pressure, as when a bee or wasp is moving about in a confined space.

7. 5". ampullaceum a more or less flask-shaped cavity with an axial
rod (Figs. 124, 125). Probably auditory.

These types of sensilla will be referred to in physiological order.

Touch. The tactile sense is highly developed in insects, and end-
organs of touch, unlike those of other senses, are commonly distributed
over the entire integument, though the antennae, palpi and cerci are es-
pecially sensitive to tactile impressions.

The end-organs of touch are bristles (sensilla chaetica). or hairs (sensilla
trichodea), each arising from a special hypodermis cell and having con-
nection with a nerve. Sensilla chaetica doubtless receive impressions
from foreign bodies, while sensilla trichodea, being best developed in the
swiftest flying insects and least so in the sedentary forms, may be affected
by the resistance of the air, when the insect or the air itself is in motion.

ENTOMOLOGY

Not all the hairs of an insect are sensory, however, for many of them
have no nerve connections.

FIGS. 116-125. Types of antennal sensilla, in longitudinal section (excepting Figs. 119
and 120). Fig. 116, sensillum cceloconicum; 117, cceloconicum; 118, basiconicum; 119,
styloconicum; 120, chseticum; 121, trichodeum; 122, trirhodeum; 123, placodeum; 124,
ampullaceum; 125, ampullaceum; c, cuticula; //, hypodermis; ;;, nerve; 5, sensory cell.
Figs. 116, 118, 121, 123, 124, honeybee, Apis mellifera; 117, 119, 122, moth, Fidoiiia piniaria;
1 20, moth, Ino prnni; 125, wasp, Vcspa crabro. After SCHENK.

In blind cave insects the antennae are very long and are exquisitely
sensitive to tactile impressions.

Taste. The gustatory sense is unquestionably present in insects, as
is shown both by common observation and by precise experimentation.

ANATOMY AND PHYSIOLOGY

77

Will fed wasps with sugar and then replaced it with powdered alum,
which the wasps unsuspectingly tried but soon rejected, cleaning the
tongue with the fore feet in a comical manner and manifesting other signs
of what we may call disgust. Forel offered ants honey mixed with mor-
phine or strychnine ; the ants began to feed but at once rejected the mix-
ture. In its range, however, the gustatory sense of insects differs often
from that of man. Thus Will found that Hymenoptera refused honey
with which a very little glycerine had been mixed (though Muscidae did
not object to the glycerine) and Forel found that ants ate unsuspectingly
a mixture of honey and phosphorus until some of them were killed by it.
Under the same circumstances, man would be able to detect the phos-
phorus but not the glycerine.

Location of Gustatory Organs. As would be expected, the end-
organs of taste are situated near the mouth, commonly on the hypo-

tb

sc

g

FIG. 126. Section through tongue of wasp, Vespa i'ul-
garis. c, cuticula; g, gland cell; /?, hypodermis; n, nerve;
ob, gustatory bristle; pli, protecting hair; sc, sensory cell;
tb, tactile bristle. After WILL.

FIG. 127. Tongue of honey
bee, Apis well if era. ^protect-
ing bristles; s, terminal spoon;
/, taste setae. After WILL.

pharynx (Fig. 126), epipharynx and maxillary palpi. On the tongue of
the honey bee the taste organs appear externally as short setae (Fig. 127)
and on the maxillae of a wasp as pits, each with a cone, or peg, projecting
from its base (Figs. 128, 129). Similar taste pits and pegs have been
found by Packard on the epipharynx in most of the mandibulate orders of
insects.

Histology. The end-organs of taste arise from special hypodermis
cells, as minute setae or, more commonly, pegs, each seated in a pit, or
cup, and connected with a nerve fiber (Figs. 129, 130). In some cases,
however, it is difficult to decide whether a given organ is gustatory or
olfactory, owing to the similarity between these two kinds of structures.

ENTOMOLOGY

In aquatic insects, indeed, the senses of taste and smell are not differen-
tiated, these forms having with other of the lower animals simply a
"chemical" sense.

Smell. In most insects the sense of smell is highly efficient and in
many species it is inconceivably acute. Hosts of insects depend chiefly
on their olfactory powers to find food, for example many beetles, the flesh
flies and the flower- visiting moths; or else to discover the opposite sex,
as is notably the case in saturniid moths. In dragon flies, however, this
sense is relied upon far less than that of sight.

Organs of Smell. By means of simple but conclusive experiments.

Hauser and others have shown that
the antennae are frequently olfactory
though not to the exclusion of tac-
tile or auditory functions, of course.
Hauser found that ants, wasps, vari-
ous flies, moths, beetles and larvae,
which react violently toward the vapor
of turpentine, acetic acid and other

fir

tc-..

FIG. 128. Under side of left maxilla
of wasp, Vcspu 1'iilgaris. p, palpus; pr,
protecting hairs; tc, taste cup; ///, tactile
hair. After WILL.

SC

FIG. 129. Longitudinal section of gustatory
end-organ (tc, of Fig. 128). c, cuticula; li, hypo-
dermis; sc, sensory cell; tc, taste cup. After WILL.

pungent fluids, no longer respond to the same stimuli after their antenna?
have been amputated or else covered with paraffine to exclude the air.
His experiments were conducted under conditions such that the results
could not be ascribed to the shock of the operation or to effects upon the
gustatory or respiratory systems; except for having lost the sense of
smell, the insects experimented upon behaved in a normal manner. It
should be said, however, that Carabus, Mclolontha and Silpha still re-
acted to some extent toward strong vapors even after the extirpation of
the antennae; while in Hemiptera the loss of the antennae did not lessen
the response to the odors used. These facts indicate that the sense of

ANATOMY AND PHYSIOLOGY

79

smell is not always confined to the antennae; indeed the maxillary palpi
are frequently olfactory, as in Silf>Iia and Hydaticus; also the cerci, as in
the cockroach and other Orthoptera. Experiments indicate that an
insect perceives some odors by means of the antennae and others by the
palpi or other organs. Hauser found that the flies Sarcophaga and Cal-
liphora, after the amputation of their antennae, became quite indifferent
toward decayed meat, to which they had previously swarmed with great
persistence, though their actions in all other respects remained normal.
Males of many moths and a few beetles are unable to find the females
(see beyond) when the former are deprived of the use of their antennas.

sc

P'iG. 130. Taste cup from maxilla of
Bombits. sc, sensory cell; n, nerve. After
WILL.

FIG. 131. Section of antennal olfactory
organ of grasshopper, Caloptenus. c, cutic-
ula; m, membrane; n, nucleus of sensory
cell; Hi 1 , nerve; p, pit with olfactory peg;
pg, pigment. After HAUSER.

End-Organs. Structures which are regarded as olfactory end-organs
occur commonly on the antennae, often on the maxillary and labial palpi
and sometimes on the cerci. These end-organs are hypodermal in origin
and consist, generally speaking, of a multinucleate cell (Fig. 131) pene-
trated by a nerve and prolonged into a chitinous bristle or peg, which is
more or less enclosed in a pit, as in Tabanus (Fig. 132). In many in-
stances, however, the end-organs take the form of teeth or cones project-
ing from the general surface of the antenna, as in Vespa (Fig. 133).
These cones are usually less numerous than the pits; in Vespa crabro,
for example, the teeth number 700 and the pits from 13,000 to 14,000
on each antenna. The pits are even more numerous in some other in-
sects; thus there are as many as 17,000 on each antenna of a blow fly

8o

ENTOMOLOGY

(Hicks). The male of Melolontha vulgaris, which seeks out the female by
the sense of smell, has according to Hauser 39,000 pits on each antenna,
and the female only 35,000. Pits presumably olfactory in function have
been found by Packard on the maxillary and labial palpi of Perla and
on the cerci of the cockroach, Periplaneta americana. Vom Rath has de-
scribed four kinds of sense hairs from the two larger of the four caudal
appendages of a cricket, Gryllus; some of these (Fig. 134) may be olfac-
tory, though possibly tactile. The same author found on the terminal
palpal segment in various Lepidoptera a large flask-shaped invagina-
tion (Fig. 135) into which project numer-
ous chitinous rods, each a process of a
sensory cell, which is supplied by a
branch of the principal palpal nerve;
these peculiar organs are inferred to be
olfactory.

FIG. 132. Section through antennal olfactory
pit of fly, Tabanns. c, cuticula; p, pit with peg;
pb, protecting bristles; s, sensory cell. After
HAUSER.

FIG. 133. Longitudinal section of
antennal olfactory organ of wasp,
Vespa. c, olfactory cell; en, olfactory
cone; ct, cuticula; It, hypodermis cells;
n, nerve; r, rod. After HAUSER.

The chief reason for regarding these various end-organs as olfactory is
that they appear from their structure to be better adapted to receive that
kind of an impression than any other, so far as we can judge from our own
experience. Though it is easy to demonstrate that the antennas, for
example, are olfactory, it frequently happens that the antennas bear sev-
eral distinct forms of sensory end-organs, so minute and intermingled
that their physiological differences can scarcely be ascertained by experi-
ment but must be inferred from their peculiarities of structure. Schenk,
however, has arrived at precise results by comparing the antennal sen-
silla in the two sexes, selecting species in which the antennae exhibit a pro-
nounced sexual dimorphism, in correlation with sexual differences of be-
havior. Taking Nololophus (Or g via) antiqua, in which the male seeks out

ANATOMY AND PHYSIOLOGY

8l

the female by means of antennal organs of smell, he finds that the male-
has on each antenna about 600 sensilla cceloconica and the female only
75; similarly in the geometrid Fidonia, in which the ratio is 350 to 100.
The sensilla styloconica, also, of these two genera are regarded as olfac-
tory organs. These two kinds of end-organs are not only structurally
adapted for the reception of olfactory stimuli, but their numerical dif-
ferences accord with the observed differences in the olfactory powers of
the two sexes, there being no other antennal end-organs to enter into the
consideration.

Assembling. It is a fact, well known to entomologists, that the
females of many moths and some beetles are able by exhaling an odor to

sc

FIG. 134. Longitudinal section of a por-
tion of a caudal appendage of a cricket,
Gryllus domesticus. b, bladderlike hair; c,
cuticula; It, hypodermis; n, nerve; ns, non-
sensory setae; sc, sense cell; sh, sensory hair.
After VOM RATH.

n

FIG. 135. Longitudinal section of apex
of palpus of Pier is. c, cuticula; //, hypo-
dermis; n, nerve; s, scales; sc, sense cells.
After VOM RATH.

attract the opposite sex, often in considerable numbers. Under favor-
able conditions, a freshly emerged^female of the promethea moth, exposed
out of doors in the latter part of the afternoon, will attract scores of the
males. A breeze is essential and the males come up against the wind ;
if they pass the female, they turn back and try again until she is located,
vibrating the antennas rapidly as they near her. The female, meanwhile,
exhales an appreciable odor, chiefly from the region of the ovipositor, and
males will congregate on the ground at a spot where a female has been.
If one of these males is deprived of the use of his antenna?, however,
7

82 ENTOMOLOGY

he flutters about in an aimless way and is no longer able to find the
female.

Among beetles, males of Polyphylla gather and scratch at places where
females are about to emerge from the ground. Prionus also assembles,
as Mrs. Dimmock observed in Massachusetts. In this instance many
males, with palpitating antennae, ran and flew to the female; moreover, a
number of females were attracted to the scene.

Sounds of Insects. Before considering the sense of hearing, some
account of the sounds of insects is desirable. Most of these are made by
the vibrations of a membrane or by the friction of one part against another.

The wings of many Diptera and Hymenoptera vibrate with sufficient
speed and regularity to give a definite note. The wing tone of a honey
bee is A ' and that of a common house fly is F'. From the pitch the num-
ber of vibrations may be determined; thus A' means 44O 1 vibrations per
second and F f , 352. The numbers thus ascertained may be verified by
Marey's graphic method (Fig. 74) ; he found that the fly referred to ac-
tually made 330 strokes per second against the smoked surface of a re-
volving cylinder.

Flies, bees, dragon flies and some beetles make buzzing or humming
sounds by means of the spiracles, there being behind each spiracle a mem-
brane or chitinous projection which vibrates during respiration. This
"voice 11 should be distinguished from the wing tone when both are pres-
ent, as in bees and flies. A fly will buzz when held by the wings, and some
gnats continue to buzz after losing wings, legs and head. The wing tone
is the more constant of the two; in the honey bee it is A', falling to E'
if the insect is tired, while the spiracular tone of the same insect is at least
an octave higher (A"} and often rises to B" or C", according to the state of
the nervous system; in fact, it is possible and even probable that various
spiracular tones express different emotions, as is indicated by the effects
produced by the voice of the old queen bee upon the young queens and
the males.

The well-known "shrilling" of the male cicada is produced by the
rapid vibration of a pair of membranes, or drums, situated on the basal

abdominal segment, and vibrated each by means of a special muscle.

Frictional sounds are made by beetles in a great variety of ways: by
the rubbing of the pronotum against the mesonotum (many Cerambyci-
dae); or of abdominal ridges against elytral rasps (Elaphrus, Cychrus);
or two dorsal abdominal rasps against speciaJized portions of the wing

1 Upon the basis of C' as 264 vibrations per second. The C' of the physicist has 256 as its
frequency of vibration.

ANATOMY AND PHYSIOLOGY 83

folds (Passalus cornn(ns), not to mention other methods. In most cases
one part forms a rasp and the other a scraper, for the production of sound.

In many of these instances the sound serves to bring the two sexes
together and is not necessarily confined to one sex; thus in Passalus cor-
nutus both sexes stridulate.

A few moths (Sphingidae) and a few butterflies make sounds; the South
American butterfly Ageronia feronia emits a sharp crackling noise as it
flies. A rasp and a scraper have been found in several ants, though ants
very seldom make any sounds that can be distinguished by the human
ear; Mutilla. however, makes a distinct squeaking sound by means of a
stridulating organ similar to those of ants.

Stridulating organs attain their best development in Orthoptera, in
which group the ability to stridulate is often restricted to the male, though
not so often as is commonly supposed. Among Acridiida?, Stenobothrus
rubs the hind femora against the tegmina to make a sound, the femur
bearing a series of teeth, which scrape across the elevated veins of the
wing-cover; while the male of Dissosteira makes a crackling sound during
flight or while poising, by means of friction between the front and hind
wings, where the two overlap.

Locustidae and Gryllidas stridulate by rubbing the bases of the teg-
mina against each other. Thus in the male Microcentrum laurifolium the
left tegmen, which overlaps the right, bears a file-like organ of about fifty-
five teeth (Fig. 136), while the opposite tegmen bears a scraper, at right
angles to the file. The tegmina are first spread a little; then, as they close
gradually, the scraper clicks across the teeth, making from twenty to
thirty sharp ''tic 1 ' -like sounds in rapid succession. This call guides the
female to the male and when they are a few inches apart she makes now
and then a short, soft chirp, to which he responds with a similar chirp,
which is quite unlike the first call and, moreover, is made by the opening
of the tegmina. These and other details of the courtship may readily be
observed in twilight and even under artificial light, as the latter, if not
too strong, does not disturb the pair. Something similar may be ob-
served in the daytime in Orchelimum, Xiphidium and the tree crickets,
(Ecanthus. The stridulating areas are usually membranous and the rasp-
ing organs are modified veins. Frequently the wing-covers bulge out to
form a resonant chamber that reinforces the sound.

The naturalist can recognize many a species of grasshopper by its
song; Scudder has expressed some of these songs in musical notation.
The usual song of the common meadow-grasshopper, Orchelimum vul-

8 4

ENTOMOLOGY

gare, may be represented by a prolonged zr . . . sound, followed by
a staccato jip-jip-jip-jip. . . .

In Orthoptera, the frequency of stridulation increases with the tem-
perature; and the correlation between the two is so close that it is easy to

s 'l\pw

B

FIG. 136. Stridulating organs of Microcentrum laurifolium. A, dorsal aspect of file (st)
when the tegmina are closed; B, ventral aspect of left legmen to show file; C, dorsal aspect
of right legmen to show scraper (s).

compute the temperature from the number of calls per minute, by means
of formulae. .The formula for a common cricket [probably a tree-cricket,
(Ecanthus niveus], as given by Professor Dolbear, is

A 7 40 i i ive j T N

T= 50 + ~, which simplified is 1 =40 + -.

Here T stands for temperature and A r , the rate per minute.

A similar formula for the katydid (Cyrtophyllus perspicillatus}, based
upon observations made by R. Hayward, would be

N-ig

Here, in computing A T , either the "katy-did" or the " she-did" is taken
as a single call.

AXATOMY AND PHYSIOLOGY 85

A. F. Shull, who has made precise observations on the stridulation of
GEcanthus, finds that there are numerous variations of rate that cannot be
accounted for by differences of temperature; that Dolbear's formula
cannot be applied without a possible error of 6.65 F.; that humidity
seems to affect the rate of chirping and that crickets show a certain in-
dividuality in their manner of chirping under the same external condi-
tions.

Hearing. There is no doubt that insects can hear. The presence
of sound-making organs is strong presumptive evidence that the sense of
hearing is present. Female grasshoppers and beetles make locomotor
and other responses to the sounds of the males, and male grasshoppers will
answer the counterfeit chirping made with a quill and a file.

Auditory organs are not restricted to any one region of an insect, but
occur, according to the species, on antennae, abdomen, legs, or elsewhere.

The antennae of some insects are evidently stimulated by certain notes,
particularly those made by their own kind. Thus the antennae of the
male mosquito are auditory, as proved by the well-known experiments of
Mayer. He fastened a male Culex to a microscope slide and sounded
various tuning forks. Certain tones caused certain of the antennal hairs
to vibrate sympathetically, and the greatest amount of vibration oc-
curred in response to 512 vibrations per second, or the note C" , which is
approximately the note upon which the female hums. The male prob-
ably turns his head until the two antennae are equally affected by the note
of the female, when, by going straight ahead, he is able to locate her with
great precision.

In the lack' of experimental evidence, other organs are inferred to be
auditory on account of their structure. Acridiidae bear on each side of
the first abdominal segment a tympanal sense organ the subject of
Graber's well-known figure (Fig. 137). This organ is admirably adapted
to receive and transmit sound-waves. The tympanum, or membrane,
is tense, and can vibrate freely, as the air pressure against the two sur-
faces of the membrane is equalized by means of an adjacent spiracle,
which admits air to the inner surface. Resting against the inner face of
the tympanum are two processes (Fig. 137, p, />), which serve probably
to transfer the vibrations, and there is also a delicate vesicle connected
by means of an intervening ganglion with the auditory nerve, which in
this case comes from the metathoracic ganglion. The nerve terminations
consist of delicate bristle-like processes which are probably affected by the
oscillations of the fluid contained in the vesicle just referred to.

Other tympanal organs, doubtless auditory, are found on the fore

86

ENTOMOLOGY

tibiae of Locustidae, ants, termites and Perlidae, on the femora of Pedicu-
lidae and the tarsi of some Coleoptera.

Several types of chordotonal organs have been described, of which
those of the transparent Corethra larva may serve as an example. These
organs, situated on each side of abdominal segments 4-10, inclusive, con-
sist each (Fig. 138) of a tense cord, probably capable of vibration, which
is attached at its posterior end to the integument and at its anterior end
to a ligament. Between the cord and the supporting ligament is a small

n

-tm

FIG. 137. Inner aspect of right tympanal sense
organ of a grasshopper, Caloptenus italicus. b, chitin-
ous border; c, closing muscle of spiracle; gn, gan-
glion; m, tympanum; n, nerve; o, opening muscle
of spiracle; p, />, processes resting against tympanum;
.v, spiracle; tin, tensor muscle of tympanum; v, vesicle.
After GRABER.

FIG. 138. Chordotonal sense
organ of aquatic dipterous larva,
L'onthru plitmicornis. cd, cord; c.i;.
chordotonal ganglion; /, fibers of an
inlegumental nerve; g, ganglion of
ventral chain; /, ligament; m, lon-
gitudinal muscles; n, chordotonal
nerve; r, rods (nerve terminations);
/, tactile setae. After GRABER.

ganglion, which receives a nerve from the principal ganglion of the seg-
ment.

Vision. The external characters of the two kinds of eyes ocelli and
compound eyes have already been described. While the lateral ocelli
are comparatively simple in structure, consisting of a small number of
cells, the dorsal ocelli almost rival the compound eyes in complexity.

Dorsal Ocelli. These consist (Fig. 139) of (i) lens, (2) vitreous

ANATOMY AND PHYSIOLOGY

body, (3) retina, (4) nerve fibers, (5) pigmcnted hypodermis cells, and (6)
accessory cells, between the retinal cells and the nerve fibers. The lens,
usually biconvex in form, is a local thickening of the general cuticula;
it is supplemented in its function by the vitreous body, consisting of a
layer of transparent hypodermis cells; these in many insects are elongate,
constituting a vitreous layer of rather more impor-
tance than the one represented in Fig. 139. The
retina consists of cells more or less spindle-shaped
and associated in pairs or in groups of two or three,
each group being termed a retinula. The basal end
of each retinal cell is continuous with a nerve fiber

r

n

FIG. 139. Median ocellus of honey bee, Apis mcllifera, in
sagittal section. //, hypodermis; /, lens; n, nerve; p, iris pigment;
r, retinal cells; v, vitreous body. After REDIKORZEW.

n

n

FIG. 140. An ocel-
lar retinula of the honey
bee, composed of two
retinal cells. A , longi-
tudinal section; B,
transverse section; u,
n, nerves; p, pigment;
r, rhabdom. After RED-
IKORZEW.

(Fig. 140), according to Redikorzew and others,
and in some instances (Calopteryx) a nerve fiber
enters the cell. Each retinula contains a longi-
tudinal rod, or rhabdom, in the secretion of which
all the cells of the retinula are concerned. Between the retinal cells and
nerve fibers are indifferent, or accessory cells. Pigment granules, usually
black, are contained in these cells, also in the retinal cells and around the
lens, in the last instance forming the iris.

Vision by Ocelli. Though the ocellus is constructed on somewhat

88

ENTOMOLOGY

the same plan as the human eye, its capacity for forming images must be
extremely limited; for since the form of the lens is fixed and also the dis-
tance between the lens and the retina, there is no power of accommodation,
and most external objects are out of focus; to make an image, then, the
object must be at one definite distance from the lens, and as the lens is
usually strongly convex, this distance must be small; in other words,
insects, like spiders, are very near-sighted, so far as the ocelli are con-
cerned; furthermore, the small number of retinal rods implies an image
of only the coarsest kind.

If the compound eyes of a grasshopper are covered with an opaque
varnish and the insect is placed in a box with only a single opening, it

readily finds its way out by means of its
ocelli; if all three ocelli are also covered,
however, it no longer does so, except by
accident, though it can make its escape
when only one of the ocelli is left uncovered.
The ocelli, then, can distinguish light from
darkness and they are probably more
serviceable to the insect in this way than
in forming images.

Compound Eyes. As regards delicacy
and intricacy of structure, the compound
eye of an insect is scarcely if at all inferior
to the eye of a vertebrate. In radial sec-
tion (Fig. 141), a compound eye appears
as an aggregation of similar elongate ele-
ments, or ommatidia, each of which ends
externally in a facet. The following struc-
tures compose, or are concerned with, each

ommatidium: (i) cornea, (2) crystalline lens, or cone, (3) rhabdom and
retinula, (4) pigment (iris and retinal), ($) fenestrate membrane, (6) fibers
of the optic nerve, (7) trachea.

The cornea (Fig. 142) is a biconvex transparent portion of the exter-
nal chitinous cuticula. Immediately beneath it are the cone cells, which
may contain a clear fluid or else, as in most insects, solid transparent
cones. The rhabdom is a transparent chitinous rod or a group of rods
(rhabdomeres) situated in the long axis of the ommatidium and surrounded
by greatly elongated cells, which constitute the retinula. Two zones of
pigment are present: an outer zone, of iris pigment, in which the pig-
ment in the form of fine black granules is contained chiefly in short cells

'i>nc

FIG. 141. Portion of compound
eye of fly, Calliphora vomitoria,
radial section, c, cornea; i, iris
pigment; , nerve fibers; nc, nerve
cells; r, retinal pigment; /, trachea.
After HICKSON.

ANATOMY AND PHYSIOLOGY

8 9

pc-

n

iv

that surround the retinula distally; and an
inner zone of retinal pigment, in which the
pigment cells are long and slender, and en-
close the retinula proximally. All these parts
are hypodermal in origin, as is also the fenes-
trate basement membrane, through which pass
tracheae and nerve fibers. The nerve fibriHae,
which are ultimate branches of the optic nerve,
pass into the retinal cells the end-organs of
vision. Under the basement membrane is a
fibrous optic tract of complex structure.

Physiology. After much experimenta-
tion and discussion upon the physiology of the
compound eye the subject of the monumental
works of Grenacher and Exner Miiller's
"mosaic" theory is still generally accepted,
though it was proposed early in the last cen-
tury. It is thought that an image is formed
by thousands of separate points of light, each
of which corresponds to a distinct field of vision
in the external world. Each ommatidium is
adapted to transmit light along its axis only
(Fig. 143), as oblique rays are lost by absorp-
tion in the black pigment which surrounds the
crystalline cone and the axial rhabdom. Along
the rhabdom, then, light can reach and affect
the terminations of the optic nerve. Each
ommatidium does not itself form a picture ; it
simply preserves the intensity and color of the
light from one particular portion of the field of
vision; and when this is done by hundreds or
thousands of contiguous ommatidia, an image
results. All that the painter does, who copies
an object, is to put together patches of light
in the same relations of quality and position
that he finds in the object itself and this is
essentially what the compound eye does, so far
as can be inferred from its structure.

Exner, removing the cones with the corneal
cuticula (in Lampyris), looked through them from behind with the aid of

rh

FIG. 142. Structure of an
ommatidium of Callipliora
I'on/iton'd. A, radial section
(chiefly); B, transverse sec-
tion through middle region;
C, transverse section through
basal region; bm, basement
membrane; c, cornea; n,
nucleus; nv, nerve nbrilke;
pc, pseudocone; pg 1 , pg 2 , cells
containing iris pigment; pg*,
cell containing retinal pigment ;
r, one of the six retinal cells
which compose the retinula; rh,
rhabdom, composed of six rhab-
domeres; /, trachea; Iv, tracheal
vesicle. After HICKSON.

9 o

ENTOMOLOGY

a microscope and found that the images made by the separate ommatidia
were either very close together or else overlapped one another, and that
in the latter case the details corresponded; in other words, as many as
twenty or thirty ommatidia may co-operate to form an image of the same
portion of the field of vision; this "superposition" image being corres-
pondingly bright an advantage, probably, in the case of nocturnal insects.
Large convex eyes indicate a wide field of vision, while small numerous
facets mean distinctness of vision, as Lubbock has pointed out. The

closer the object the better the sight, for the
greater will be the number of lenses employed to
produce the impression, as Mollock says. If
Mtiller's theory is true, an image may be formed
of an object at any reasonable distance, no power
of accommodation being necessary; while if, on
the other hand, each cornea with its crystalline
cones had to form an image after the manner of
an ordinary hand-lens, .only objects at a definite
distance could be imaged.

The limit of the perception of form by insects
is placed at about two meters for Lampyris, 1.50
meters for Lepidoptera, 68 cm. for Diptera and
58 cm. for Hymenoptera.

It is generally agreed, however, that the com-
pound eyes are specially adapted to perceive
movements of objects. The sensitiveness of in-
sects to even slight movements is a matter of
common observation; often, however, these in-
sects can be picked up with the fingers, if the
operation is performed slowly until the insect is
within the grasp. A moving object affects different facets in succession,
without necessitating any turning of the eyes or the head, as in verte-
brates. Furthermore, on the same principle, the compound eyes are
serviceable for the perception of form when the insect itself is moving
rapidly.

The arrangement of the pigment depends adaptively upon the quality
of the light, as Stefanowska and Exner have shown; thus, when the light
is too strong, the iris and retinal pigment cells elongate around the om-
matidium and their pigment granules absorb from the cone cells and rhab-
dom the excess of light. If the light is weak, they shorten, and absorb
but a minimum amount of light.

FIG. 143. Diagram of
outer transparent portion
of an ommatidium to illus-
trate the transmission of an
axial ray 0*4) and the re-
peated reflection and ab-
sorption of an oblique ray
(B), which at length emerges
at C. p, iris pigment.

ANATOMY AND PHYSIOLOGY 9 1

Origin of Compound Eye. The compound eye is often said to
represent a group of ocelli, chiefly for the reason that externally there
appears to be a transition from simple eyes, through agglomerate eyes,
to the facetted type. This plausible view, however, is probably incor-
rect, for these reasons among others. In the ocellus, a single lens serves
for all the retinulae, while in the compound eye there are as many lenses
as there are retinulae. Moreover, ocelli do not pass directly into com-
pound eyes, but disappear, and the latter arise independently of the for-
mer.

Probably, as Grenacher holds, both the ocellus and the compound
eye are derived from a common and simpler type of eye are "sisters,"
so to speak, derived from the same parentage.

Perception of Light through the Integument. In various in-
sects, as also in earthworms, blind chilopods and some other animals,
light affects the nervous system through the general integument. Thus
eyeless dipterous larvae avoid the light, or, more precisely, they retreat
from the rays of shorter wave-length (as the blue), but come to rest in
the rays of longer wave-length (red), as if they were in darkness (see page
286). The blind cave-beetles of the genus Anophthalmus react to the
light of a candle (Packard). Graber found that a cockroach deprived
of its eyesight could still perceive light, but Lubbock found that an ant
whose eyes had been covered with an opaque varnish became indifferent
to light.

Color Sense. Insects undoubtedly distinguish certain colors, though
their color sense differs in range from our own. Thus ants avoid violet
light as they do sunlight, but probably cannot distinguish red or orange
light from darkness; on the other hand, they are extremely sensitive to
the ultra-violet rays. Honey bees frequently select blue flowers: white
butterflies (Pieris) prefer white flowers, and yellow butterflies (Colias)
appear to alight on yellow flowers in preference to white ones (Packard).
In fact, the color sense is largely relied upon by insects to find particular
flowers and by butterflies to a large extent to find their mates. To be
sure, insects will visit flowers after the brightly colored petals have been
removed or concealed, as Plateau found, but this does not prove that
the colors are of no assistance to the insect, though it does show that
they are not the sole attraction the odor also being an important
guide'. The honey bee is able to distinguish color patterns, according
to the experiments of C. H. Turner.

Problematical Sense Organs. As all our ideas in regard to the

92 ENTOMOLOGY

sensations of insects are necessarily inferences from our own sensory ex-
periences, they are inevitably inadequate. Wliile it is certain that in-
sects have at least the senses of touch, taste, smell, hearing and sight, it is
also certain that these senses of theirs differ remarkably in range from
our own, as we have shown. W T e can form no accurate conception of
these ordinary senses in insects, to say nothing of others that insects have,
some of which are probably peculiar to insects. Thus they have many
curious integumentary organs which from their structure and nerve con-
nections are probably sensory end-organs, though their functions are
either doubtful or unknown. Such an organ is the sensillum placodeum
(p. 76), the use of which is very doubtful, though the organ is possibly
affected by air pressure. Insects are extremely sensitive to variations of
wind, temperature, moisture and atmospheric pressure, and very likely
have special end-organs for the perception of these variations; indeed,
the sensilla trichodea are probably affected by the wind, as we have said.
The halteres of Diptera, representing the hind wings, contain sensory
organs of some sort. They have been variously regarded as olfactory
(Lee), auditory (Graber), and as organs of equilibration. When one or
both halteres are removed, the fly can no longer maintain its equilibrium
in the air, and Weinland holds that the direction of flight is affected by
the movements of these " balancers."

6. DIGESTIVE SYSTEM

The alimentary tract in its simplest form is to be seen in Thysanura,
Collembola and most larvae, in which (Fig. 144) it is a simple tube ex-
tending along the axis of the body and consisting of three regions, namely,

FIG. 144. Alimentary tract of a collembolan, Orchesdla. F, fore gut; H, hind gut; .17,
mid gut; c, cardiac valve; cm, circular muscle; Im, longitudinal muscle; p, pharynx; py,
pyloric valve.

ore, mid and hind gut. These regional distinctions are fundamental, as
the embryology shows, for the middle region is entodermal in origin and
the two others are ectodermal, as appears beyond.

There are many departures from this primitive condition,- and the

ANATOMY AND PHYSIOLOGY

93

most specialized insects exhibit the following modifications (Figs. 145,
146) of the three primary regions:

Fore intestine (stomodtzmn) : mouth, pharynx, oesophagus, crop, pro-
ventriculus (gizzard), cardiac valve.

Mid intestine (mesenteron) : ventriculus (stomach).

Hind intestine (proctodceum) : pyloric valve, ileum, colon, rectum,
anus.

Stomodaeum. The mouth, the anterior opening of the food canal, is
to be distinguished from the pharynx, a dilatation for reception of food.
In the pharynx of mandibulate insects the food is acted upon by the
saliva; in suctorial forms the pharynx acts as a pumping organ, in the
manner already described.

The (esophagus is commonly a simple tube of small and uniform caliber,
varying greatly in length according to the kind of insect. Passing be-
tween the commissures that connect the brain with the subcesophageal

cr

0-.

FIG. 145. Alimentary tract of a grasshopper, Mclanoplus d(ffercntialis. c, colon; cr,
crop; gc, gc, gastric caeca; /', iieum; m, mid intestine, or stomach; mt, Malpighian, or kidne\ .
tubes; o, oesophagus; p, pharynx; r, rectum; s, salivary gland of left side.

ganglion (Fig. 113), the oesophagus leads gradually or else abruptly into
the crop or gizzard, or when these are absent, di,rectly into the stomach.
In addition to its function of conducting food, the oesophagus is sometimes
glandular, as in the grasshopper, in which it is said to secrete the "mo-
lasses" which these insects emit.

The crop is conspicuous in most Orthoptera (Fig. 145) and Cole-
optera (Fig. 146) as a simple dilatation. In Neuroptera (Fig. 147) its
capacity is increased by means of a lateral pocket the food reservoir;
this in Lepidoptera, Hymenoptera and Diptera is a sac (Fig. 148, c)
communicating with the oesophagus by means of a short neck or a long
tube, and serving as a temporary receptacle for food. In herbivorous
insects the crop contains glucose formed from starch by the action of
saliva or by the secretion of the crop itself; in carnivorous insects this se-
cretion converts albuminoids into assimilable peptone-like substances.

Next comes the enlargement known as the proventriculus, or gizzard.

94

ENTOMOLOGY

which is present in many insects, especially Orthoptera and Coleoptera
(Fig. 146), and is usually found in such mandibulate insects as feed upon
hard substances. The proventriculus is lined with chitinous teeth or
ridges for straining the food, and has powerful circular muscles to squeeze
the food back into the stomach, as well as longitudinal muscles for re-
laxing, or opening, the gizzard.
Some authors maintain that the
proventriculus not only serves as
a strainer, but also helps to com-
minute the food, like the gizzard
of a bird.

FIG. 146. Digestive system of a beetle,
Carabns. a, anal gland; c (of fore gut), crop;
c (of hind gut), colon, merging into rectum;
d, evacuating duct of anal gland; g, gastric
caeca; /, ileum; in, mid intestine; mt, Mal-
pighian tubes; o, oesophagus; p, proventricu-
lus; r, reservoir. After KOLBE.

FIG. 147. Digestive system of Mynnc-

Icon larva, c, caecum; cr, crop; m, mid

intestine; ;;//, Malpighian tubes; s, spin-
neret. After MEINERT.

In most insects a cardiac valve guards the entrance to the stomach,
preventing the return of food to the gullet. This valve (Figs. 144, 149)
is an intrusion of the stomodaeum into the mesenteron, forming a circular
lip which permits food to pass backward, but closes upon pressure from
behind.

ANATOMY AND PHYSIOLOGY

95

Mesenteron. The ventriculus, otherwise known as the ;;//</ intes-
tine, or stomach, is usually a simple tube of large caliber, as compared with

mt

cl

o

cm

FIG. 148. Alimentary tract of a moth, Sphinx, c, food reservoir; cl, colon; cm, caecum;
/, ileum; m, mid intestine; nit, Malpighian tubes; o, cesophagus; r, rectum; s, salivary
gland. After WAGNER.

the oesophagus or intestine, and into the ventriculus may open glandular
blind tubes, or gastric ctzca (Figs. 145, 146); these, though numerous in
some insects, are commonly few in number and restricted to the anterior
region of the stomach. The gastric caeca of Orthop-
tera secrete a weak acid which emulsifies fats, or
one which passes forward into the crop, there to
act upon albuminoid substances. In the stomach
.the food may be acted upon by a fluid secreted by
specialized cells of the epithelial wall. In various
insects, certain cells project periodically into the
lumen of the stomach as papilla?, which by a process
of constriction become separated from the parent
cells and mix bodily with the food. This phenom-
enon takes place in the larva of Ptychoptera (van
Gehuchten), also in nymphs of Odonata (Needham),
and is probably of widespread occurrence among
insects. The chief function of the stomach, how-
ever, is absorption, which is effected by the general
epithelium. Physiologically, the so-called stomach
of an insect is quite unlike the stomach of a verte-
brate, being more like an intestine.

Proctodaeum. At the anterior end of the hind
intestine there is usually a pyloric valve, which pre-
vents the contents of the intestine from returning into the stomach.
This valve may operate by means of a sphincter, or constricting,
muscle, or may, as in Collembola (Fig. 144), consist of a back-

FIG. i4Q. Cardiac
valve of young muscid
larva, o, oesophagus;
p, proven triculus; v,
valve. In an older
larva the valve pro-
jects into the mid in-
testine. After KOWA-
LEVSKY.

9 6

ENTOMOLOGY

ward-projecting circular ridge, or lip, which closes upon pressure from
behind.

In its primitive condition the hind intestine is a simple tube (Fig. 144).
Usually, however, it presents two or even three specialized regions,
namely and in order, ileum, colon and rectum (Fig. 145). The hind in-
testine varies greatly in length and is frequently so long as to be thrown
into convolutions (Fig. 150). The ileum is short and stout in grass-
hoppers (Fig. 145); long, slender and convoluted in many carnivorous
beetles; and quite short in caterpillars and most other larvae; its func-
tion is absorption. The colon, often absent, is evident in Orthoptera and

r-...

FIG. 150. Digestive system of Bclos-

FIG. 151. Wall of mid intestine of silk

toma. c, caecum; i, ileum; m, mid intestine; worm, transverse section. &, basement mem-
mt, Malpighian tubes; r, salivary reservoir; brane; c, circular muscle; i, intima; I, longi-
s, salivary gland. After LOCY, from the tudinal muscle; n, n, nuclei of epithelial

A merican Naturalist.

cells; s, secretory cell.

Lepidoptera and may bear (Benacus, Dytiscus, Silphidae, Lepidoptera) a
conspicuous csecal appendage (Figs. 148, 150) of doubtful function,
though possibly a reservoir for excretions. The colon contains indigest-
ible matter and the waste products of digestion, including the excretions
of the Malpighian tubes. The rectum (Fig. 145) is thick-walled, strongly
muscular and often folded internally. Its office is to expel excrementi-
tious matter, consisting largely of the indigestible substances chitin,
cellulose and chlorophyll. The rectum terminates in the anus, which
opens through the last segment of the abdomen, always above the genital
aperture.

ANATOMY AND PHYSIOLOGY

97

g

C

Histology. The epithelial wall of the alimentary tract is a single
layer of cells (Fig. 151), which secretes the intima. or lining layer, and the
basement membrane a delicate, structureless enveloping layer. The
intima, which is continuous with the external cuticula, is chitinous in the
fore and hind gut (which are ectodermal in origin), but not in the mid gut
(entodermal), and usually exhibits extremely tine transverse striae, which
are due probably to minute pore canals. Surrounding the basement
membrane is a series of circular muscles and outside these is a layer of
longitudinal muscles. The circular muscles serve to
constrict the pharynx in sucking insects and, in general,
to squeeze backward the contents of the alimentary
canal by successively reducing its caliber. The longi-
tudinal muscles, restricted almost entirely to the mid
intestine, act in opposition to the constricting muscles
to enlarge the lumen of the food canal and in addition
to effect peristaltic movements of the stomach.

The intima of the crop is sometimes shaped into
teeth, and that of the proventriculus is heavily chiti-
nized and variously modified to form spines, teeth or
ridges.

Salivary Glands. In their simplest condition,
the salivary glands are a pair of blind tubes (Fig. 152),
one on each side of the oesophagus and opening separ-
ately at the base of the hypopharynx. Commonly,
however, the glands open through two salivary ducts
into a common, or evacuating, duct; a pair of salivary
reservoirs (Fig. 153) may be present and the glands
are frequently branched or lobed, and, though usually
confined to the head, may extend into the thorax or
even into the abdomen.

Many insects have more than one pair of glands
opening into the pharynx or oesophagus; thus the
honey bee has six pairs and Hymenoptera as a whole have as many as
ten different pairs. Though all these are loosely spoken of as salivary
glands, it is better to restrict that term to the pair of glands that open
at the hypopharynx.

All these cephalic glands are evaginations of the stomodaeum (ecto-
dermal in origin) and consist of an epithelial layer with the customary
intima and basement membrane (Fig. 154). The nuclei are large, as is

usually the case in glandular cells, and the cytoplasm consists of a dense

8

g

FIG. 152. A sim-
ple salivary gland of
CccriliKs. c, canal; d,
duct; g, g. glandular
cells. After KOLBE.

9 8

ENTOMOLOGY

framework (appearing in sections as a network) enclosing vacuoles of a
clear substance the secretion; the chitinous intima is penetrated by fine
pore canals through which the secretion passes. In many insects, no-
tably the cockroach, the common duct is held distended by spiral threads
which give the duct much the appearance of a trachea.

S

FIG. 153. Right salivary gland of cockroach, ventral FIG. 154. Histology of

aspect, c, common duct; g, gland; h, hypopharynx; r, salivary gland of Cizcilius,
reservoir. After MIALL and DENNY. radial section. b, basement

membrane; c, canal; ^glandu-
lar cell; /, intima; n, nucleus.
After KOLBE.

In herbivorous insects the saliva changes starch into glucose, as in
vertebrates; in carnivorous forms it acts on proteids and is often used
to poison the prey, as in the larva of Dyti&us. In the mosquito each
gland is three-lobed (Fig. 155) ; the middle lobe is different in appearance
from the two others and secretes a poisonous fluid which is carried out

along the hypopharynx. Though this
poison is said to facilitate the process
of blood-sucking by preventing the
coagulation of the blood, its primary
use was perhaps to act upon proteids
in the juices of plants.

Malpighian Tubes. The kid-
ney, or Malpighian, tubes, present in
nearly all insects, are long, slender, blind tubes opening into the intestine
immediately behind the stomach as a rule (Figs. 145, 146), but always
into the intestine. The number of kidney tubes is very different in dif-
ferent insects; Collembola have none, while Odonata have fifty or more
and Acridiidas as many as one hundred and fifty; commonly, however,
there are four or six, as in Coleoptera, Lepidoptera and many other orders.

FIG. 155. One of the three-lobed
salivary glands of a mosquito. The
middle lobe secretes the poison. After
MACLOSKIE, from the American Nat-
uralist.

ANATOMY AND PHYSIOLOGY

99

Not more than six and frequently only four occur in the embryo (Wheeler ) ,
though these few embryonic tubes may subsequently branch into many.
The Malpighian tubes (Fig. 156) are evaginations of the proctodaeum
and are consequently ectodermal. A cross-section of a tube shows a
ring of from one to six or more large polygonal cells (Fig. 157), which
often project into the lumen of the tube; the nuclei are usually large and

may be branched, as in Lepidoptera. A chitinous
intima, traversed by pore-canals, lines the tube,
and a delicate basement membrane is present, sur-
rounded by a peritoneal layer of connective tissue.
Furthermore, the urinary tubes are richly supplied
with tracheae. In function, the Malpighian tubes
are analogous to the vertebrate kidneys and con-

Ojt

n

FIG. 156. Portion FIG. 157. Cross-section of Malpighian tube of silkworm,

of Malpighian tube of Bombyx mori. b, basement membrane; c, crystals; I, intima;

caterpillar, Samia ce- 1, lumen; n, nucleus; p, peritoneal layer. Greatly magnified.
cropia, surface view.

tain a great variety of substances, chief among which are uric acid and its
derivatives (such as urate of sodium and of ammonium), calcium oxalate
and calcium carbonate.

Parts of the fat-body may also be concerned in excretion; thus the
fat-body in Collembola and Orthoptera serves for the permanent storage
of urates.

7. CIRCULATORY SYSTEM

Insects, unlike vertebrates, have no system of closed blood-vessels,
but the blood wanders freely through the body cavity to enter eventually
the dorsal vessel, which resembles a heart merely in being a propulsatory
organ.

Dorsal Vessel. The dorsal vessel (Figs. 158, 162) is a delicate tube
extending along the median dorsal line immediately under the integu-

IOO

ENTOMOLOGY

ment.

al

A simple tube in some larva?, it consists in most adults chiefly
of a series of chambers, each of which has on each
side a valvular opening, or ostium (Fig. 159),
which permits the ingress of blood but opposes
its egress; within the chambers occur other val-
vular folds that allow the blood to move forward
only. With few exceptions (Ephemerida?) the
dorsal vessel is blind behind and the blood can
enter it only through the lateral ostia.

Aorta. The posterior, or pulsating portion
(heart) of the dorsal vessel is confined for the
most part to the abdomen ; the anterior portion,
or aorta, extends as a simple attenuated tube
through the thorax and into the head, where it
passes under the brain and usually divides into
two branches (Fig. 162), each of which may
again branch. In the head the blood leaves the
aorta abruptly and enters the general body cavity.

Alary Muscles. Extending outward from
the "heart," or propulsatory portion, and making
with the dorsal wall of the body a pericardial
chamber, is a loose diaphragm, formed largely by
paired fan-like muscles the alary muscles (Figs.
158, 1 60). These are thought to assist the heart
in its propulsatory action.

Structure of the Heart. The dorsal vessel
has a delicate lining-membrane, or intima, and a thin enveloping mem-

FIG. 158. Dorsal ves-
sel of beetle, Lucaims. a,
aorta; al, alary muscle; o,
ostium. After STRAUS-

DiJRCKHEIM.

FIG. 159. Diagram of a portion of the
heart of a dragon fly nymph, Epillnra.
o, ostium; v, valve; the arrows indicate
the course of the blood. After KOLBE.

FIG. 160. Diagrammatic cross-section of
pericardial region of a grasshopper, (Edipoda.
a, alary muscle; d, dorsal vessel; s, suspensory
muscles; sp, septum. After GRABER.

ANATOMY AND PHYSIOLOGY

101

brane; between these, in the heart, is a layer of fine muscle fibers, circular
or spiral in direction, which effect the contractions of the organ.

Ventral Sinus. In many if not most insects a pulsatory septum
(Fig. 178, v) extends across the floor of the body cavity to form a sinus,

d

g

FIG. 161. Blood corpuscles of a grasshopper, Slcnobolhnis. a-f, corpuscles covered with fat-
globules; g, corpuscle after treatment with glycerine, showing nucleus. After GRABER.

in which the blood flows backward, bathing, the ventral nerve cord as it
goes. This ventral sinus supplements the heart in a minor way, as do
also the local pulsatory sacs which have been discovered in the legs
of aquatic Hemiptera and the head of Orthoptera.

Blood. The blood, or hcemolymph,
of an insect consists chiefly of a watery
fluid, or plasma, which contains cor-
puscles, or leucocytes. Though usually
colorless, the plasma is sometimes yellow
(Coccinellida?, Meloidse), often greenish
in herbivorous insects from the pres-
ence of chlorophyll, and sometimes of
other colors; often the blood owes its
hue to yellow or red drops of fat on the
surface of the blood corpuscles (Fig.
161).

Leucocytes. The corpuscles, or
leucocytes, are minute nucleated cells,
6 to 30 jj. in diameter, variable in form
even in the same species, but commonly
(Fig. 161) round, oval or ovate in pro-
file, though often disk-shaped, elongate
or amoeboid in form.

Function of theBlood. Theblood
of insects contains many substances,
including egg albumin, globulin, fibrin, iron, potassium and sodium
(Mayer), and especially such a large amount of fatty material that its
principal function is probably one of nutrition; the blood of an insect
contains no red corpuscles and has little or nothing to do with the

FIG. 162. Diagram to indicate the
course of the blood in the nymph of a
dragon fly, Epithccu. a, aorta; //,
heart; the arrows show directions taken
by currents of blood. After KOLBE.

102 ENTOMOLOGY

aeration of tissues, that function being relegated to the tracheal sys-
tem.

Circulation. The course of the circulation is evident in trans-
parent aquatic nymphs or larvae. In odonate or ephemerid nymphs,
currents of blood may be seen (Fig. 162) flowing through the spaces
between muscles, tracheae, nerves, etc., and bathing all the tissues;
separate outgoing and incoming streams may be distinguished in the
antennas and legs; the returning blood flows along the sides of the body
and through the ventral sinus and the pericardial chamber, eventually to
enter the lateral ostia of the dorsal vessel. A circulation of blood occurs
in the wings of freshly emerged Odonata, Ephemerida, Coleoptera,
Lepidoptera, etc., the currents trending along the tracheae; this circula-
tion ceases, Ijowever, with the drying of the wings.

The chambers of the dorsal vessel expand and contract successively
from behind forward. At the expansion (diastole} of a chamber its ostia
open and admit blood; at contraction (systole) the ostia close, as well as
the valve of the chamber next behind, while the chamber next in front
expands, affording the only exit for the blood. The valves close partly
through blood-pressure and partly by muscular action.

The rate of pulsation depends to a great extent upon the activity of
the insect and upon the temperature and the amount of oxygen or car-
bonic acid gas in the surrounding atmosphere. Oxygen accelerates the
action of the heart and carbonic acid gas retards it. A decrease of 8 or
10 C. in the case of the silkworm lowers the number of beats from 30 or 40
to 6 or 8 per minute. The more active an insect, the faster its heart beats.

The rate of pulsation is very different in the different stages of the
same insect. Thus in Sphinx ligustri, according to Newport, the mean
number of pulsations in a moderately active larva before the first moult
is about 82 or 83 per minute; before the second moult, 89, sinking to 63
before the third moult, to 45 before the fourth, and to 39 in the final larval
stage; the force of the circulation, however, increases as the pulsations
decre'ase in number. During the quiescent period immediately preceding
each moult, the number of beats is about 30. In the pupal stage the
number sinks to 22, and then lowers until, during winter, the pulsations
almost cease. The moth in repose shows 41 to 50 per minute, and after
flight as many as 139.

*

8. FAT-BODY

The fat-body appears (Fig. 163) as many-lobed masses of tissue filling
in spaces between to her organs and occupying a large part of the body

ANATOMY AND PHYSIOLOGY

103

cavity. The distribution of the fat-body is to a certain extent definite,
however, for the fat-tissue conforms to the general segmentation and is
arranged in each segment with an approach to symmetry. Much of this
tissue forms a distinct peripheral layer in each segment, and masses of
fat-body occur constantly on each side of the alimentary tract and also at

FIG. 163. Transverse section of the abdomen of a caterpillar, Pieris raps, b, blood cor-
puscles; c, cuticula; d, dorsal vessel; /, fat-body; g, ganglion; //, hypodermis; /, leg; m,
muscle; ml, mid intestine, containing fragments of cabbage leaves; mt, Malpighian tube; s,
silk gland; sp, spiracle; tr, trachea.

the sides of the dorsal vessel, in the latter case forming the pericardia!
fat-body.

Fat-Cells. The fat-cells (Fig. 164) are large and at first more or less
spherical, with a single nucleus (though there are said to be two in Apis
and several in Musca), but the cellular structure of the fat-tissue is often
difficult to make out because the cells are usually filled with globules of
fat (Fig. 165), while old cells break down, leaving only a disorderly net-
work. The fat-cells sometimes contain an albuminoid substance, and

ENTOMOLOGY

B

FIG. 164. Fat-cells of a cater-
pillar, Picris. A, cells filled with
drops of fat; B, cell freed of fat-drops,
showing nucleus. After KOLBE.

usually the fat-body includes considerable quantities of uric acid or its
derivatives, frequently in the form of conspicuous concretions.

Functions. The physiology of the fat-system is still obscure. Prob-
ably the fat-body combines several functions. In caterpillars and other
larvae it furnishes a reserve supply of nutriment, at the expense of which

the metamorphosis takes place; the
amount of fat increases as the larva
grows, and diminishes in the pupal stage,
though some of it lasts over to furnish
nourishment for the imago and its germ
cells. The gradual accumulation of uric
acid and urates in the fat-body indicates
an excretory function, particularly in
Collembola, which have no Malpighian
tubes. The intimate association between the ultimate tracheal branches
and the fat-body has led some authorities to ascribe a respiratory func-
tion to the latter. A close relation of some sort exists also between the
fat-system and the blood-system; fat-cells are found free in the blood,
and the blood corpuscles originate in the thorax and abdomen from tis-
sues that can scarcely be distinguished
from fat-tissues. The corpuscles (leu-
cocytes, or phagocytes] which in some
insects absorb effete larval tissues
during metamorphosis have been by
some authors regarded as wandering
fat-cells. Cells constituting the peri-
cardial fat-body are attached to the
lateral muscles (alary muscles) of the
dorsal vessel, but almost nothing is
known as to their function.

(Enocytes. Associated with the
fat-body proper and with tracheae as
well are the peculiar yellow cells
known as (Enocytes (Fig. 166), that
occur in abdominal segments of larvae,
as compared with all other insect-cells excepting ova, and are essentially
isolated from one another, though grouped among tracheal branches into
loose clusters, one on each side of a spiracle-bearing segment.

After arising from the primitive ectoderm the cenocytes never divide,

FIG. 165. Section through fat-body
of a silkworm, showing nucleated cells,
loaded with drops of fat.

These cells are enormous in size

ANATOMY AND PHYSIOLOGY

105

FIG. 166. CEno-
cytes and accom-
panying trachea,
from abdomen of a
silkworm.

but gradually increase in size (Wheeler), and their size is in a general way
proportional to that of the fat-body.

Their function has been problematical until recently. Many ob-
servers have regarded them as ductless glands, having seen "microscop-
ical exudations around the periphery of the cytoplasm,
especially at times when the nucleus is greatly ramified,
and therefore manifesting its great activity" (Glaser).
R. W. Glaser has recently thrown light upon the
nature of the cenocytic fluid. By using three-year-old
caterpillars of Zeuzera pyrina, which have a great
amount of fatty tissue and correspondingly large
cenocytes, he was able to extract enough of the fluid
for chemical experiments. He found by carefully con-
ducted tests that the fluid had the power of oxidizing
fats, by means of enzymes known as oxidases (though
no fat-splitting enzyme, or lipase, was present), and
concluded that the secretion of the cenocytes is used
to oxidize the reserve food stored up by the larva in the form of fat.
Photogeny. This phenomenon appears sporadically and by various
means in protozoans, worms, insects, fishes and other animals. Lumi-
nosity in insects, though sometimes merely an incidental and pathological

effect of bacteria, is usually pro-
duced by special organs in which
light is generated probably by
the oxidation of a fatty substance.
There are not many luminous
insects. Those best known are
the Mexican and West Indian
beetles of the genus Pyrophorus
(Elateridae), in which the pro-
notum bears a pair of luminous
spots, and the common fire-flies
(Lampyridas). In Lampyridas
the light is emitted from the
ventral side of the posterior ab-
dominal segments, and the struc-
ture of the photogenic organ is essentially the same throughout the family.
InPhotimis this organ (Fig. 167) consists of two layers: a ventral photogenic
layer and a dorsal reflecting layer. The latter, white and opaque, consists
of polygonal cells containing large quantities of crystals of urates; the

FIG. 167. Transverse section of portion of
photogenic organ of Photum*. c, cylinder; p,
photogenic layer; r, reflecting layer; /, trachea.
After TOWNSEXD.

IO6 ENTOMOLOGY

former layer is composed of tracheal structures and intervening paren-
chyma cells. The tracheae branch profusely in the photogenic layer,
where, the larger air-tubes are each surrounded by a more or less cylin-
drical mass of cells; tracheal branches penetrate between the cells of each
cylinder, at the edge of which they pass into tracheoles which penetrate
the photogenic tissue and anastomose with those of adjacent cylinders;
in the meshes of the tracheolar network is a granular substance of fatty
nature ("differentiated fat-body"), the oxidation of which is the source
of the luminosity, it is inferred. The photogenic tissues of Photinus,
after being dried and kept in sealed tubes, have retained their photogenic
power for more than eighteen months, glowing after this interval upon
the "application of water in the presence of air or oxygen" (McDermott).
Three factors are involved in the production of the light : a substance to
be oxidized, oxygen and water.

The rays emitted by the common fire-flies are remarkable in being
almost entirely light rays. According to Young and Langley, the radia-
tions of an ordinary gas-flame contain less than three per cent, of visible
rays, the remainder being heat or chemical rays, of no value for illumina-
ting purposes; while the light-giving efficiency of the electric arc is only
ten per cent, and that of sunlight only thirty-five per cent. The luminous
efficiency of the fire-fly, however, is not much under one hundred per
cent. ; in Photuris pennsylvanica it is about ninety-two per cent., according
to Coblentz an efficiency as yet unapproached by artificial means. The
actinic power of the light is so slight that it affects a photographic plate
only after a long exposure. Coblentz, who has recently applied most re-
fined rriethods of measurement to the radiation of fire-flies, found that
exposures of one to five hours were necessary with the spectograph. He
was unable to detect any infra-red radiation; the thermal radiation, if
present, being immeasurably small as yet. The intensity of the glow
averages ^-jj.irro candle power in our common fire-flies, according to
Coblentz.

This luminosity serves to bring the sexes together. "The male flies
over the tops of the grasses, weeds, etc., dropping down between them
and flashing; any females that come within the range of his flash, answer
by their slower flash; if the male sees this answering flash from one, he
approaches her, flashes again, to which she answers, and he then finally
locates her definitely by means of subsequent flashes," as McDermott
says. He found that he could get responses from the females by imi-
tating the flash of the male with a small electric bulb or even with a com-
mon safety match, and that he could deceive the males also by flashing the
tiny electric light after the manner of the female.

ANATOMY AND PHYSIOLOGY

107

9. RESPIRATORY SYSTEM

In insects, as contrasted with vertebrates, the air itself is conveyed to
the remotest tissues by means of an elaborate system of branching air-
tubes, or trachea, which receive air through paired segmentally-arranged
spiracles. Each spiracle is commonly the mouth of a short tube which
opens into a main tracheal trunk (Fig.
1 68) extending along the side of the
body. From the two main trunks
branches are sent which divide and
subdivide until they become extremely
delicate tubes, which penetrate even
between muscle fibers, between the
ommatidia of the compound eyes and
possibly enter cells. In most cases
each main longitudinal trunk gives oft*
in each segment (Fig. 169) three large
branches: (i) an upper, or dorsal,
branch, which goes to the dorsal mus-
cles; (2) a middle, or visceral, branch,
which supplies the alimentary tract and
the reproductive organs; (3) a lower, or
ventral, branch, which pertains to the
ventral ganglia and muscles.

In many swiftly flying insects (dra-
gon flies, beetles, moths, flies and bees)
there occur tracheal, pockets, or air-sacs,
which were formerly and erroneously
supposed to diminish the weight of the
insect, but are now regarded as simply
air-reservoirs.

Types of Tracheation.--T wo
types of tracheal system are distin-
guished for convenience: (i) The pri-
mary, open, or holopneustic type described above, in which the spiracles
are functional; (2) the secondary, closed, or apneustic type, in which the
spiracles are either functionless or absent. This type is illustrated in
Collembola and such aquatic nymphs and larvae as breathe either directly
through the skin or else by means of gills. The two types, however, are
connected by all sorts of intermediate stages.

vs

FIG. 1 68. Tracheal system of an
insect. a, antenna; b, brain; /, leg;
n, nerve cord; p, palpus; s, spiracle;
st, spiracular, or stigmatal, branch; /,
main tracheal trunk; v, ventral branch;
vs, visceral branch. After KOLBE.

io8

ENTOMOLOGY

Tracheal Gills. In many aquatic nymphs and larvae the spiracles
are suppressed (though they become functional in the imago) and res-

FIG. 169. Diagrammatic cross-section of the thorax of an insect, a. alimentary canal-,
d, dorsal vessel; g, ganglion; s, spiracle; w, wing; /, dorsal tracheal branch; 2, visceral
branch; j, ventral branch.

piration is effected by means of gills; these are cuticular outgrowths

which usually, though not invariably, contain tracheae, and are commonly

lateral or caudal in position. Lateral
tracheal gills are highly developed in
ephemerid nymphs (Fig. 170), in which
a pair occurs on some or all of the first
seven segments of the abdomen ; a few gen-
era, however, have cephalic or thoracic
gills. Larvae of Trichoptera have paired
abdominal gills varying greatly in form
and position, and Perlidae often have
paired thoracic gills. Caudal tracheal gills
are conspicuous in nymphs of Agrionidae
(Fig. 171) as three foliaceous appendages.
A few coleopterous larvae of aquatic habit,
as Gyrinus and Cnemidotus, possess tracheal
gills, as do also caterpillars of the genus
Para pony x (Fig. 172), which feed on the

leaves of several kinds of water plants.

Though manifold in form, tracheal gills are generally more or less

foliaceous or filamentous, presenting always an extensive respiratory

FIG. 170. Lateral gill from ab-
domen of a May fly nymph, Hcxa-
genia variabilis. Enlarged.

ANATOMY AND PHYSIOLOGY

FIG. 171. Caudal
gills of an agrionid
nymph, enlarged.

surface; their integument is thin and the trachea? spread closely beneath
it. These adaptations are often supplemented by waving movements of
the gills, as in May fly nymphs, and by frequent movements of the insect
from one place to another.

Especially noteworthy are the rectal tracked gills of odonate nymphs.
In these insects the lining of the rectum forms nu-
merous papillae or lamellae, which contain a profusion
of delicate tracheal branches; these are bathed by
water drawn into the rectum and then expelled, at
rather irregular intervals. A similar rectal respira-
tion occurs also in ephemerid nymphs and mosquito
larvae.

A few forms, chiefly Perlidae, are exceptional in

retaining tracheal gills in the adult stage; in some imagines they are merely
vestiges of the nymphal gills, but in others, such as Pteronarcys (Fig. 18),
which habitually dips into the water and rests in moist situations, the

gills probably supplement the spiracles. Further
details on the respiration of aquatic insects are
given in Chapter IV.

Spiracles. The paired external openings of the
tracheae occur on the sides of the thorax and ab-
domen', there being never more than one pair to a
segment. Though the thysanuran Japyx has n
pairs, no winged insect has more than 10; although
there are in all 1 2 segments which may bear spira-
cles the three thoracic and the first nine abdominal
segments. (Additional details are given on page p. 52.)
The spiracles, or stigmata, are usually provided
with bristles, hairs or other processes to exclude
dust ; or the hairs of the body may serve the same
purpose, as in Lepidoptera and Diptera; in many
beetles the spiracles are protected by the elytra; in
other beetles, however, and in many Hemiptera and
Diptera the spiracles are unprotected externally.
Larvae that live in water or mud may have spiracles
at the end of a long tube, which can be thrust up
into the pure air; this is true of the dipterous larvae of Erislalis, Bitta-
comorpha (Fig. 173) and Culex (Fig. 230).

Closure of Spiracles. As a rule, a spiracle is opened and closed
periodically by means of a valve, operated by a special ocdusor muscle.

FIG. 172. Cater-
pillar of Par a pony x
obscuralis, to show tra-
cheal gills. Length, 15
mm. After HART.

no

ENTOMOLOGY

In dipterous larvae the closure is effected by the contraction of a circular
muscle, but Coleoptera and Lepidoptera, among other insects, have a
somewhat complex apparatus for closing the trachea immediately behind

the spiracle. Thus, in the stag-beetle, a crescentic
bow (Fig. 174, b) extends half around the trachea, and
the rest of the circumference is spanned by a lever (I)
and a band (bd) ; these three chitinous parts, articulated
together, form a ring around the trachea. Further-
more, a muscle (m) connects the lever and the band.
As the muscle shortens, the lever turning upon the
end of the band as a fulcrum, pulls the bow toward
the lever and band until the enclosed trachea is pinched
together. When the muscle relaxes, the trachea opens
by its own elasticity.

Structure of Tracheae. The tracheae originate
in the embryo as simple in-pocketings of the outer
germ layer, or ectoderm, and from these the countless
tracheal branches are derived by the same process of
invagination. The lining membrane of a trachea is,
then, continuous with the external cuticula, and the
cellular wall of a trachea is continuous with the rest
of the hypodermis. This wall consists of a layer of polygonal cells (Fig.
175) fitting closely together as a pavement epithelium. The chitinous

FIG. 173. Larva
of Blttacomor pha
clavipes showing res-
piratory tube.- Nat-
ural size. After
HART.

- m

FIG. 174. Apparatus for closing the spiracular trachea in a beetle, Lucanus. A, trachea
opened; B, closed; b, bow; bd, band; c, external cuticula; /, lever; m, muscle; s, spiracle;
/, trachea. After JUDEICH and NITSCHE.

lining, or intima, is thickened at regular intervals to form thread-like
ridges, which course around the inner circumference in essentially a
spiral manner, though the continuity of the so-called spiral thread is

ANATOMY AND PHYSIOLOGY

III

frequently interrupted. These elastic threads, or tccnidia, serve to keep
the trachea open without affecting its flexibility.

The ultimate tracheal branches (Fig. 176) are extremely delicate
tubes, which do not end blindly, but anastomose with one another, form-

1 .- i <r /.*

' K$

,-ft

FIG. 175. Structure of a
trachea. //, tracheal hypodermis;
i, intima; t, taenidium.

FIG. 176. Tracheal capillary end-network from
silk gland of Porthctria dispar. p, peritracheal mem-
brane; t, tracheal capillary. After WISTINGHAUSEN.

ing a capillary network of confluent tubes. Some authors have held that
the finest tracheal filaments penetrate epithelial or other cells.

t

A

B

FIG. 177. Transverse sections of abdominal segments, to illustrate respiratory movements.
A, cockroach (Blatta); B, bee (Bombus); s, sternum; /, tergum. The dotted lines indicate
positions of terga and sterna after expiration; the continuous lines, after inspiration. After
PLATEAU.

Respiration. The external signs of respiration are the regular open-
ing and closing movements of some of the spiracles and the rhythmic con-
traction and expansion of the abdomen. During contraction, the dorsal
and ventral walls approach each other (Fig. 177) and during expansion
they separate. The tergum moves more than the sternum in Cole-

112

ENTOMOLOGY

optera and Heteroptera, and vice versa in Acridiidae, Odonata, Diptera
and aculeate Hymenoptera. The width of the abdomen usually changes
but little during respiration, for the tergal and sternal movements are
taken up by the pleural membranes which, as in the grasshopper, infold at
contraction and straighten out at expansion. Other respiratory move-
ments occur, but they are of minor importance.

The rate of respiration increases or diminishes with the activity of the
insect and with temperature and other conditions. In six specimens of
Melanoplus differentialis , held between the fingers, the thoracic spiracles

opened and closed respectively 34, 43,
45, 54, 60 and 61 times per minute.
Four individuals of M. femur-rubrum
under the same circumstances gave
70, 78, 90 and 92.

At expansion inspiration takes
place, and at contraction expiration
occurs. In the grasshopper, the tho-
racic spiracles open almost simulta-
neously with the expansion of the ab-
domen. Contraction is effected by
special vertical expiratory muscles
(Fig. 178), but expansion is due to
the elasticity of the abdominal wall,
as a rule; this is the reverse of what
occurs in mammals, where expiration
is passive and inspiration active.
Inspiratory muscles are found, how-
ever, in Acridiidae, Trichoptera and
Hymenoptera.

Though the respiratory move-
ments of an insect may be studied with a hand-lens, a more precise
method is that of Plateau the chief authority on insect physiology
who made use of the stereopticon to project an enlarged profile of the
insect upon a screen, on which could be marked the different contours
of the abdomen at its phases of inspiration and expiration.

The way in which the air reaches the finest tracheal branches is not
clearly ascertained, but it is thought that air is forced into these tubes
by pressure from the abdominal muscles, while its escape through the
spiracles is being prevented by the compression of the stigmatal tracheae.
The respiratory movements are entirely reflex and are independent of
the brain or suboesophageal ganglion, for they continue after decapitation

FIG. 178. Diagrammatic cross-sec-
tion of abdomen of a grasshopper, Ac rid -
ium. d, dorsal septum, or diaphragm;
ex, expiratory muscle; /, fat-body; g,
ganglion; h, heart; in, inspiratory mus-
cle; v, ventral septum, below which is
the ventral sinus. The dorsal and
ventral septa rise and fall periodically.
After GRABER.

ANATOMY AND PHYSIOLOGY

and even in the detached abdomen of a grasshopper or dragon fly. Each
ventral ganglion of the body is an independent respiratory center for its
particular segment.

10. REPRODUCTIVE SYSTEM

The sexes are always separate in insects, hermaphroditism occurring
only as an abnormal condition. The sexual organs, situated in the ab-
domen, consist essentially of a pair of ovaries or testes and a pair of ducts
(oviducts or seminal ducts, respectively). Primitively, the ducts open
separately, as they still do in Ephemeridae, but in almost all other insects
the two ducts enter a common evacuating duct (vagina or ejaculatory

,'TI

FIG. 179. Reproductive -system of male
beetle, Melolontha. a, accessory gland; c, copu-
latory organ; d, ejaculatory duct; s, seminal
vesicle; t, testis; v, vas deferens. After KOLBE.

FIG. i So. Reproductive system of
male Lepidoptera. a, accessory gland;
d, ejaculatory duct; /, united testes; v,
vas deferens. After KOLBE.

duct); this opens ordinarily between the penultimate and antepenulti-
mate segments of the abdomen, i. e., usually the ninth and eighth, at any
rate never through the last abdominal segment.

Homologies. As in other animals, the reproductive organs are
homologous in the two sexes. Thus:

MALE. FEMALE

Testes = Ovaries

Seminal ducts = Oviducts
Ejaculatory dnct = \\igina

Seminal vesicle = Seminal receptacle
Accessory glands = Accessory glands
Penis and accessor ies = Ovipositor

Male Organs. Each testis, though sometimes a single blind tube, is
usually a group of tubes or sacs (Fig. 179), testicular follicles, which open
9

114

ENTOMOLOGY

into a seminal duct (vas defer ens). In most Lepidoptera the testes are
secondarily united into a single mass (Fig. 180) as also in Acridiidae.
The two seminal ducts enter the common ejaculatory duct, which termin-
ates in the intromittent organ, or penis. Often each vas deferens is di-
lated near its mouth into a seminal vesicle, or reservoir; or there may be
only a single seminal vesicle, arising from the common duct. One or
more pairs of glands opening into the vasa deferentia or the ductus ejac-
ulatorius secrete a fluid which mixes with the spermatozoa and oftentimes

unites them into packets, known as spermato-
phores.

All these parts are subservient to the forma-
tion, preservation and emission of the spermatozoa.
These minute, thread-like bodies (Fig. 181) arise
in the testicular follicles from a germinal epithe-
lium, and consist, as in vertebrates, of a head,
middle- piece and a vibratile tail without enter-
ing into the finer structure.

Female Organs. Each^ary (Fig. 182) con-
sists of one or more tubes opening into an oviduct.
The two oviducts enter a common duct, the
vagina, which opens to the exterior, often through
an ovipositor. Frequently the vagina is expanded
as a pouch, or bursa copulatrix, though in Lepidop-
tera the bursa and the vagina are distinct from
each other and open separately (Fig. 183). In
most insects a dorsal evagination of the vagina
forms a seminal receptacle, or spermatheca, from
which spermatozoa emerge to fertilize the eggs.
The accessory glands, either paired or single,
provide a secretion for attaching the eggs to

foreign objects, cementing the eggs together, forming an egg-capsule, etc.
In each ovarian tube, or ovariole, are found ova in successive stages of
growth, the largest and oldest ovum being nearest the oviduct. In the
primitive type of egg-tube, as in Thysanura and Orthoptera (Fig. 184,
A) every chamber contains an ovum; in more specialized types, every
other chamber contains a nutritive cell instead of a germ cell, the nutri-
tive cells serving as food for the adjacent ova (B] ; or the nutritive cells,
instead of alternating with the ova, may be collected in a special chamber,
beyond the ovarian chambers (C). An egg-tube is usually prolonged dis-
tally as a terminal filament or suspensor, the free end of which is attached
near the dorsal vessel.

FIG. 181. Sperma-
tozoa. A, locustid grass-
hopper; B, cockroach,
Blatta; C, beetle, Copris.
After BUTSCHLI and BAL-
LOWITZ.

ANATOMY AND PHYSIOLOGY

II

Ovaries and testes arise from indifferent cells, or primitive germ cells,
which are at first exactly alike in the two sexes. In the female, certain
of these cells form ova and others form a follicle around each ovum (Fig.
185). In the male, the primary germ cells form cells termed spcrmaio-
gonia; each of these forms a spermalocyte, and this gives rise to four sper-
matozoa.

Hermaphroditism. The phenomenon of hermaphroditism, or the
combination of male and female characters in the same individual, occurs

FIG. 182. Reproductive system of queen
honey bee. a, accessory sac of vagina; b,
bulb of stinging apparatus; c, colleterial, or
cement, gland; o, ovary; od, oviduct; p,

vs

FIG. 183. Reproductive system of 'fe-
male Lepidoptera. b, bursa copulatrix; /,
terminal filament; g, cement glands; o, o,
ovaries; od, oviduct; r, receptaculum seminis;

poison glands; pr, poison reservoir; r, re- , vagina; vs, vestibule, or entrance to bursa.
ceptaculum seminis; re, rectum; v, vagina. After KOLBE.
After LEUCKART.

only as an extremely rare abnormality among insects. Speyer estimated
that in Lepidoptera only one individual in thirty thousand is hermaphro-
ditic. Bertkau (1889) listed 335 hermaphroditic arthropods, of which 8
were crustaceans, 2 spiders, 2 Orthoptera, 8 Diptera, 9 Coleoptera, 51
Hymenoptera and 255 Lepidoptera. The large proportion of Lepidoptera
is due in great measure to the fact that they are collected oftener than
other insects (excepting possibly Coleoptera) and that sexual dimorph-
ism is so prevalent in the order that hermaphrodites are easily recognized.
The most common kind of hermaphroditfsm is that in which one side
is male and the other female, as in Fig. 186. Bertkau found this right-

n6

ENTOMOLOGY

and-left hermaphroditism in 153 individuals. In other instances the
antero-posterior kind may occur, as when the fore wings are of one sex
and the hind wings of the other; rarely, the characters of the two sexes
are intermingled.

Hermaphroditic insects are such rarities that very few of them have
been sacrificed to the dissecting needle in order to determine whether the

phenomenon involves the primary organs
as well as the secondary sexual characters.
Where dissections have been made it has
been found usually that hermaphroditism
does extend to the reproductive organs them-

B

FIG. 184. Types of ovarian
tubes. ^4, without nutritive cells;
B, with alternating nutritive and
egg-cells; C, with terminal nutri-
tive chamber; c, terminal cham-
ber; e, egg-cell; ep, follicle epi-
thelium; /, terminal filament; s,
strands connecting ova with nutri-
tive chamber; y, yolk, or nutritive
cells. From Lang's Lclirbucli.

FIG. 185. Ovum of a butterfly, ]'(inesxa, in its
follicle, e, follicle epithelium; g, germinal vesicle;
n, branching nucleus of nutritive cell; o, ovum.
After WOODWORTH.

selves. Thus a butterfly with male wings
on the right side and female wings on the
left would have a testis on the right side of
the abdomen and an ovary on the left side.

Parthenogenesis. Reproduction without fertilization is a normal
phenomenon in not a few insects. This parthenogenesis may easily be
observed in plant lice. In these insects there are many successive broods
consisting of females only, which bring forth living young; at definite
intervals, however, and usually in autumn, males appear also, and ferti-
lized eggs are laid which last over winter. This cyclic reproduction, by
the way, is known as heterogeny. Among Hymenoptera, parthenogenesis
is prevalent, usually alternating with sexual reproduction, as in many

AX ATOMY AND PHYSIOLOGY

117

FIG. 186. Hermaphrodite gypsy moth,
Port heir ia d is par; right side, male; left,
female. Natural size.^After TASCHENBERG
from Hertwig's Lclirhiu'lt.

Cynipidae. In some Cynipidae, however, males are unknown; such is
the case also in some Tenthredinidae. The statement has long been made
that the unfertilized eggs of worker ants, bees and wasps produce invari-
ably males; it has been found, however, that the parthenogenetic
worker eggs of the ant La-
sins nigcr may produce normal
workers (Reichenbach, Mrs. A.
B. Comstock). Males may, of
course, result from fertilized eggs,
as in the honey bee, according to
Dickel, who maintains, indeed,
that all the eggs laid by the queen
bee are fertilized. Partheno-
genesis has been recorded as oc-
curring also in a few moths, some
Coccidae and many Thysanoptera.
Paedogenesis. In Miastor

and a few other genera of CccidomyiidcB young are produced by the larva.
This extraordinary form of parthenogenesis is termed pcedogenesis, and is
limited apparently to the family Cecidomyiidae. The paedogenetic larvae
of Miastor (Fig. 187) develop before the oviducts have appeared and

escape by the rupture of the
mother. After several suc-
cessive generations of this
kind the resulting larvae pu-

FIG. 187. Young psedogenetic larvae of Miastor pa t e anc [ f orm normal male
in the body of the mother larva. Greatly enlarged.
-After PAGENSTECHER. and female flies.

An excellent account of

Miastor has been given by Dr. Felt, who has discovered this remarkable
genus in New York State.

The pupa of a species of Chironomus occasionally deposits unfertilized
eggs, which develop, however, in the same manner as the fertilized eggs
of the species.

CHAPTER III

DEVELOPMENT

i. EMBRYOLOGY

Ovum. The ovum of an insect, as of any other animal, is a single
cell (Fig. 188), with a large nucleus (germinal vesicle*), & large nucleolus,
nutritive matter, or yolk (deuto plasm), contained in the cytoplasm, and a

cell wall (vitelline membrane} secreted by the ovum;
the egg-shell, or chorion, is secreted around the
ovum by surrounding ovarian cells.

Maturation. As a preparation for fertilization
the germinal vesicle divides twice, forming two polar
bodies, and as the first of these bodies may itself
divide, there result four cells; three of these, how-
ever the polar bodies are minute and rudimen-
tary.

These phenomena of orogenesis are paralleled in
the development of the spermatozoa, or spermato-
genesis; for the primary spcrmatocyte gives rise to
two secondary spcrmatocytes, and these to four sper-
matids, each of which forms a spermatozoon.

By means of this maturation process the number
of chromosomes in the egg-nucleus is reduced to half
the number normal for somatic cells (body cells as
distinguished from germ cells) . A similar reduction
occurs also during the development of the sperma-
tozoon, and when sperm-nucleus and egg-nucleus
unite, the resulting nucleus contains the normal
number of chromosomes. The meaning of these
reduction phenomena highly important from the
standpoint of heredity is a much debated subject.
Fertilization. As the eggs pass through the
vagina, they are capable of being fertilized by sper-
matozoa, previously stored in the seminal recep-
tacle. Through the micro pyle of the chorion one or more spermatozoa
enter and a sperm nucleus unites with the egg-nucleus to form what is

nS

FIG. 188 Sagittal
section of egg of fly,
Aliisca, in process of
fertilization. c, cho-
rion; d, dorsal; ;, mi-
cropyle, with gelatinous
exudation; p, male and
female pronuclei, before
union; pb, polar bodies;
pr, peripheral proto-
plasm; v, ventral; vt,
vitelline membrane; y,
yolk. After HENKING
and BLOCHMANN.

DEVELOPMENT

y

FIG. 189. Equatorial section of egg of a
beetle, Clytra'laviiiscula. b, blastoderm; g, germ
band; y, yolk granule; yc,, yolk cell. After
LECAILLON.

known as the segmentation nucleus. Through this union of nuclear sub-
stances the qualities of the two parents are combined in the offspring.
Needless to say, the minute details of the process of fertilization are of
the highest biological impor-
tance.

Blastoderm. In an arthro-
pod ovum the yolk occupies a
central position (centrolecithal
type), being enclosed in a thin
layer of protoplasm. From the
segmentation nucleus just men-
tioned are derived many nuclei,
some of which migrate outward
with their attendant protoplasm
to form with the original peri-
pheral protoplasm a continuous
cellular layer, the blastoderm (Fig.
189).

Germ Band. The blasto-
derm, at first of uniform thick-
ness, becomes thicker in one re-
gion, by cell multiplication, forming the germ band (primitive streak, etc.) ;
this appears in surface view as an oval or elongate area, denser than the
remaining blastoderm, with which it is, of course, continuous.

Gastrulation. The germ band next infolds along the median line,

appearing in cross-
section as in Fig. 190;
the two lips of the
median groove close
together over the in-
vaginated portion and
form an outer layer,
or ectoderm (Fig. 191),
while the invaginated
portion spreads out

as an inner layer, which is destined to form two layers, known respectively
as entoderm and mesoderm. This formation of two primary germ layers
by invagination or otherwise is termed gastrulation; it is an important
stage in the development of all eggs, and among insects several variations
of the process occur.

FIG. 190. Transverse section of germ band of Clytra at gas-
trulation. g, germ band; /, inner layer. After LECAILLON.

I2O

ENTOMOLOGY

Amnion and Serosa. Meanwhile, the blastoderm has been folding
over the germ band from either side, as shown in Fig. 190, and at length
the two folds meet and unite to form two membranes (Fig. 192), namely,
an inner one, or amnion, and an outer one, or serosa.

ac

FIG. 191. Transverse section of germ
layers and amnion folds of Clytra. a, am-
nion; e, ectoderm; i, inner layer (meso-ento-
derm); 5, serosa. Original, based on Lec-
aillon's figures.

FIG. 192. Transverse section of germ
layers and embryonal membranes of Clytra.
a, amnion; ac, amnion cavity; e, ectoderm;
/, inner layer (meso-entoderm) ; s, serosa.
After LECAILLON.

Segmentation and Appendages. On the germ band, which repre-
sents the ventral part of the future insect, the body segments are marked
off by transverse grooves (Figs. 193, 195); this segmentation beginning

ifci/fvs.
ttWfti:j.-.

-> .-- *?.!?"

^^fex

TV*' '''; '^v'' '* ''

t/&'&;-l:

\n^<

*&&

w im

ft

&* ?.*

Iff

iifi^ v*v

iff

s* &

w.= 5J&

n "a

is ^

A

^ft^i

>* &L V* X*

^?W C

FIG. 193. Germ band of a beetle, Mclasoina, in three successive stages. .4, unsegmented;
B, with oral segments demarkated; C, with three oral, three thoracic and two abdominal
segments. After GRABER.

usually at the anterior end of the germ band and progressing backward.
Furthermore, an anterior infolding occurs (Fig. 194), forming the stomo-
dceum, from which the mouth, pharynx, oesophagus and other parts of the
fore gut are to arise; a similar but posterior invagination, or proctodcsum
(Fig. 194), is the beginning, or fundament, of the hind gut.

DEVELOPMENT

121

At the anterior end of the germ band is a pair of large procephalic
lobes (Figs. 193, 195), which eventually bear the lateral eyes, and im-
mediately behind these are the funda-
ments of the antennae. The funda-
ments of the primary paired append-
ages are out-pocketings of the ecto-
dermal germ band, and at first an-
tennae, mouth parts and legs are all
alike, except in their relative positions.
Behind the antennae (in Thysanura
and Collembola at least) appears a
pair of rudimentary appendages (Fig.
195, /) which are thought to represent

the second antennae of Crustacea; instead of developing, they disappear in
the embryo or else persist in the adult as mere rudiments. In front of
these transitory intercalary appendages is the mouth-opening, above which

s ' . < g

FIG. 194. Diagrammatic sagittal
section of hymenopterous egg to show
stomadaeal (5) and proctodaeal (/>) in-
vaginations of the germ band (g). After
GRABER.

~~pr

FIG. 195. Ventral aspect of germ
band of a collembolan, A unrida maril-
ima. a, antenna; a 1 - a 5 , abdominal ap-
pendages; /, intercalary appendage;
/, labrum; //, left labial appendage;
m, mandible; mx, maxilla; p, pro-
cephalic lobe; pr, proctodseum; t l -t 3 ,
thoracic legs.

si I

V.

g

FIG. 196. Anterior aspect of embryonal mouth
parts of a collembolan, A unrida niaritima. a, an-
tenna; /, labrum; /g, prothoracic leg; li, left funda-
ment of labium; /, lingua; m, mandible; ;K.V, max-
illa; p, maxillary palpus; si, superlingua. After
FOLSOM.

the labrum and clypeus are already indicated by a single, median evagina-
tion. Behind the mouth the mandibles, maxillae and labium are repre-

122

ENTOMOLOGY

sented by three pairs of fundaments, and in Thysanura and Collembola a
fourth pair is present to form the superlinguae (Fig. 196, 5/), already re-
ferred to. Next in order are the three pairs of thoracic legs (Fig. 195)
and then, in many cases, paired abdominal appendages (Figs. 195, 197),
indicating an ancestral myriopod-like condition; some of these abdom-
inal limbs disappear in the embryo but others develop into abdominal

prolegs (Lepidoptera and Tenthredinidae) ,
external genital organs (Orthoptera, Hy-
menoptera, etc.) or other structures. The
study of these embryonic fundaments sheds
much light upon the morphology of the ap-
pendages and the subject of segmentation.

Two Types of Germ Bands. The germ
band described above belongs to the simple
overgrown type, exemplified in Clytra, in which
the germ band retains its original position
and the amnion and serosa arise by a pro-
cess of overgrowth (Figs. 191, 192), as dis-
tinguished from the invaginated type, illus-
trated in Odonata, in which the germ band
invaginates into the egg, as in Fig. 198, until
the ventral surface of the embryo becomes
turned around and faces the dorsal side of
the egg. In this event, a subsequent process
of revolution occurs, by means of which the
ventral surface of the embryo resumes its
original position (Fig. 199).

Dorsal Closure. As was said, the germ
band forms the ventral part of the insect.
To complete the general form of the body the
margins of the germ band extend outward
and upward (Fig. 200) until they finally close
over to form the dorsal wall of the insect.
Besides this simple method, however, there
are several other ways in which the dorsal closure may be effected.

Nervous System. Soon after gastrulation, the ventral nervous
system arises as a pair of parallel cords from cells (Fig. 201, n) which have
been derived by direct proliferation from those of the germ band, and are
therefore ectodermal in origin. This primitive double nerve cord be-
comes constricted at intervals into segments, or neuromeres, which cor-

FIG. 197. Embryo of (Ecan-
ihus, ventral aspect, a , antenna ;
a 1 -^ 5 , abdominal appendages; e,
end of abdomen; /, labrum; li,
left fundament of labium; Ip,
labial palpus; l l -P, thoracic legs;
m, mandible; mp, maxillary
palpus; mx, maxilla; p, pro-
cephalic lobe; pr, proctodasum.
After AYERS.

DEVELOPMENT

123

respond to the segments of the germ band. Each neuromere consists of
a pair of primitive ganglia, and these are connected together by paired

FIG. 198. Diagrammatic sagittal sections to illustrate invagination of germ band in
Caloptcryx. a, anterior pole; ac, amnion cavity; am, amnion; b, blastoderm; d, dorsal;
g, germ band; h, head end of germ band; p, posterior pole; s, serosa; v, ventral; y, yolk.-
After BRANDT.

a

a

FIG. 199. Diagrammatic sagittal sections to illustrate revolution of Caloplcryx embryo,
a, antenna; aw, amnion; /, labium; l l -P, thoracic legs; m, mandible; ix, maxilla; s, serosa.
After BRANDT.

nerve cords, which later may or may not unite into single cords; more-
over, some of the ganglia finally unite to form compound ganglia, such

124

ENTOMOLOGY

as the brain and the subcesophageal ganglion. In front of the oesophagus
(Fig. 55) are three neuromeres : (i) proto cerebrum, which is to bear the
compound eyes; (2) deuto cerebrum, or antennal neuromere; (3) tritocere-
brum, which belongs to the segment which bears the rudimentary inter-
calary appendages spoken of above. Behind the oesophagus are, at
most, four neuromeres, namely and in order, mandibular, superlingual

n

B

FIG. 200. Diagrammatic transverse sections to illustrate formation of dorsal wall in
the beetle, Lcptinotarsa. a. amnion (breaking up in C); g, germ band; s, serosa. After
WHEELER, from the Journal of Morphology.

(found only in Collembola as yet), maxillary and labial. Then follow the
three thoracic ganglia and ten (usually) abdominal ganglia. The first
three neuromeres always unite to form the brain, and the next four
(always three; but four in Collembola and perhaps other insects), to
form the subcesophageal ganglion. Compound ganglia are frequently
formed also in the thorax and abdomen by the union of primitive ganglia.
Tracheae. The tracheae begin as paired invaginations of the ecto-
derm (Fig. 202, t); these simple pockets elongate and unite to form the

main lateral trunks, from which
arise the countless branches of
the tracheal system.

M e soderm . From the inner
layer which was derived from the
germ band by gastrulation (Figs.
190-192) are formed the impor-
tant germ layers known as meso-
derm and entoderm. Most of the
layer becomes mesoderm, and this splits on either side into chambers,
or ccelom sacs (Fig. 201, c), a pair to each segment. In Orthoptera these
ccelom sacs are large and extend into the embryonic appendages, but in
Coleoptera, Lepidoptera and Hymenoptera they are small. These sacs
may share in the formation of the definite body-cavity, though the last
arises independently, from spaces that form between the yolk and the

FIG. 201. Transverse section of germ layers
of Clytra. c, ccelom sac; ;/, neuroblasts (primi-
tive nervous cells). After LECAILLOX.

DEVELOPMENT 125

mesodermal tissues. From the ccelom sacs develop the muscles, fat-
body, dorsal vessel, blood corpuscles, ovaries and testes; the external
sexual organs, however, as well as the vagina and ejaculatory duct, are
ectodermal in origin.

Entoderm. At its anterior and posterior ends, the inner layer just
referred to gives rise to a mass of cells which are destined to form the
mesenteron, from which the mid intestine develops. One mass is ad-
jacent to the blind end of the stomodasal invagination and the other to
that of the proctodaeal in-folding. The two masses become U-shaped
(Fig. 203), and the lateral arms of the two elongate and join so that the
entodermal masses become connected by two lateral strands of cells;

-7- -5

-/--my

m s

FIG. 202. Transverse section of abdomen of Clytra embryo at
an advanced stage of development, a, appendage; e, epithelium of
mid intestine; g, ganglion; m, Malpighian tube; mi, muscular layer
of mid intestine; ms, muscle elements; my, mesenchyme (source of
fat-body); s, sexual organ; /, tracheal invagination. After LECAIL-
LON.

FIG. 203. Dia-
gram of formation
of entoderm in Lep-
tinotarsa. e, e, en-
todermal masses; m.
mesoderm. After
WHEELER.

by overgrowth and undergrowth from these lateral strands a tube is
formed which is destined to become the stomach, and by the disappear-
ance of the partitions that separate the mesenteron from the stomodaeum
at one end and from the proctodaeum at the other end, the continuity of
the alimentary canal is established. The fore and the hind gut, then,
are ectodermal in origin, and the mid gut entodermal.

Polyembryony. In certain Hymenoptera a single egg may give
rise to many individuals. Thus in some Chalcididae and Proctotrypidae,
according to Marchal, the fertilized ovum segments into many (12-100)
embryos, which develop into as many adults, all the individuals from the
same ovum being of the same sex.

126

ENTOMOLOGY

2. EXTERNAL METAMORPHOSIS

Metamorphosis. One of the most striking phenomena of insect
life is expressed by the term metamorphosis, which means conspicuous
change of form after birth. The egg of a butterfly produces a larva;
this eats and grows and at length becomes a pupa; which, in turn, de-
velops into an imago. These stages are so different (Fig. 27) that with-
out experience one could not know that they pertained to the same in-
dividual.

Holometabola. The more specialized insects, namely, Neuroptera,
Mecoptera, Trichoptera, Lepidoptera, Coleoptera (Fig. 204), Diptera

FIG. 204. Cyllene carycE. A, larva; B, pupa; C, imago. X 3.

(Figs. 205, 29), Siphonaptera (Fig. 30) and Hymenoptera (Fig. 284),
undergo this indirect, or complete, 1 metamorphosis, involving profound
changes of form and distinguished by an inactive pupal stage. These
insects are grouped together as Holometabola.

Larvas receive such popular names as "caterpillar" (Lepidoptera),
"grub" (Coleoptera), and "maggot" (Diptera), while the pupa of a
moth or butterfly (especially the latter) is called a "chrysalis."

Heterometabola. In a grasshopper, as contrasted with a butterfly,
the imago, or adult, is essentially like the young at birth, except in hav-
ing wings and mature reproductive organs, and the insect is active
throughout life; hence the metamorphosis is termed direct, or incomplete.

1 These terms, though somewhat misleading in implication, are currently used.

DEVELOPMENT

127

This type of transformation, without a true pupal period, is character-
istic of the more generalized of the metamorphic insects, namely, Orthop-
tera, Platyptera, Plecoptera, Ephemerida (Fig. 19), Odonata (Fig. 20),
Thysanoptera and Hemiptera (Fig. 206). These orders constitute the
group Heterometabola. Within the limits of the group, however, various
degrees of metamorphosis occur;
thus Plecoptera, Ephemerida and
Odonata undergo considerable
change of form; a resting, or
quiescent, period may precede the
imaginal stage, as in Cicada (Fig.
207); while male Coccidae have
what is essentially a complete
metamorphosis. In fact, the vari-
ous kinds of metamorphosis grade
into one another in such a way as
to make their classification to some

extent arbitrary and inadequate.

iic. 205. Pnormia rcgi.na. A, larva; B,

As there is no distinction be- puparium; C, imago, x 5.

tween larva and pupa in most

heterometabolous insects, it is customary to use the term nymph during
the interval betw r een egg and imago.

FIG. 206. Six successive instars of the squash bug, Aiiasa tristis. X 2.

Ametabola. The most generalized insects, Thysanura and Collem-
bola, develop to sexual maturity without a metamorphosis; the form

128

ENTOMOLOGY

at hatching is retained essentially throughout life, there are no traces of
wings even in the embryo, and there is no change of habit. These two
orders form the group Ametabola. All other insects have a metamor-

B

C

FIG. 207. Cicada tibiccn. A, imago emerging from nymphal skin; B, the cast skin; C,

imago. Natural size.

phosis in the broad sense of the term, and are therefore spoken of as
Metabola. In this w r e follow Packard, rather than Brauer, who uses a
somewhat different set of terms to express the same ideas.

B

V E

FIG. 208. Eggs of various insects. A, butterfly, Polygonia interrogationis; B, house fly,
Miisca domestica; C, chalcid, Brucliophagus funcbris; D, butterfly, Papilio troilus; E, midge,
Diisynnira Irifolii; F, hemipteron, Triphlcps insidiasiis; G, hemipteron, Podisiis spinosus;
H, fly, Drosopliiln ampelophila. Greatly magnified.

Stadium and Instar. During the growth of every insect, the skin
is shed periodically, and with each moult, or ecdysis, the appearance of
the insect changes more or less. The intervals between the moults are
termed stages, or stadia. To designate the insect at any particular stage,
the term instar has been proposed and is growing in favor; thus the

DEVELOPMENT

129

FIG. 209. Three eggs of the cabbage butter-
fly. Pier is rapes. Greatly magnified, but all
drawn to same scale.

insect at hatching is the first instar, after the first moult the second in star,
and so on.

Egg. The eggs of insects are exceedingly diverse in form. Com-
monly they are more or less spherical, oval, or elongate, but there are
innumerable special forms, some
of which are quite fantastic.
Something of the variety of form
is shown in Fig. 208. As regards
size, most insect eggs can be dis-
tinguished by the naked eye;
many of them tax the vision,
however, for example, the ellip-
tical eggs of Dasyneura legu-
minicola, which are but .300 mm.
in length and .075 mm. in width;

the oval eggs of the cecropia moth, on the other hand, are as long as 3
mm.

The egg-shell, or chorion, secreted around the ovum by cells of the
ovarian follicle, may be smooth but is usually sculptured, frequently with

ridges which, as in lepidopterous eggs, may serve,
to strengthen the shell. The ornamentation of
the egg-shell is often exquisitely beautiful, though
the particular patterns displayed are probably of
no use, being incidentally produced as impressions
from the cells which secrete the chorion. Varia-
tions of form, size and pattern are frequent in eggs
of the same species, as appears in Fig. 209.

Always the chorion is penetrated by one or
more openings, constituting the micro pyle, for the
entrance of spermatozoa.

As a rule, the eggs when laid are accompanied
by a fluid of some sort, which is secreted usually
by a cement gland or glands, opening into the
vagina. This fluid commonly serves to fasten
the eggs to appropriate objects, such as food
plants, the skin of other insects, the hairs of
mammals, etc.; it may form a pedicel, or stalk,
for the egg, as in Ckrysopa (Fig. 210); may surround the eggs as a gelat-
inous envelope, as in caddis flies, dragon flies, etc. ; or may form a cap-
sule enclosing the eggs, as in the cockroach.

FIG. 210. Chrysopa, laying
eggs. Slightly enlarged.

10

130 ENTOMOLOGY

The number of eggs laid by one female differs greatly in different
species and varies considerably in different individuals of the same species.
Some of the fossorial wasps and bees lay only a dozen or so and some
grasshoppers two or three dozen, while a queen honey bee may lay a
million. Two females of the beetle Prionus laticollis had, respectively,
332 and 597 eggs in the abdomen (Mann). A. A. Girault gives the fol-
lowing numbers of eggs per female, from an examination of twenty egg-
masses of each species:

Maximum. Minimum. Average.

Thyridopteryx ephemereeformis 1076 753 941

Clisiocampa americana 466 313 375-5

Chionas pis furfur a 84 33 66.5

Hatching. Many larvae, caterpillars for example, simply eat their
way out of the egg-shell. Some maggots rupture the shell by contortions
of the body. Some larvae have special organs for opening the shell;
thus the grub of the Colorado potato beetle has three pairs of hatching
spines on its body (Wheeler) and the larval flea has on its head a tempo-
rary knife-like egg-opener (Packard). The process of hatching varies
greatly according to the species, but has received very little attention.

Larva. Although larvae, generally speaking, differ from one another
much less than their imagines do, they are easily referable to their orders
and usually present specific differences. Larvae that display individual
adaptive characters of a positive kind (Lepidoptera, for example) are
easy to place, but larvae with negative adaptive characters (many Dip-
tera and Hymenoptera) are often hard to identify.

Thysanuriform Larvae. Two types of larvae have been recognized
by Brauer, Packard and other authorities: thysanuriform and cruciform;
respectively generalized and specialized in their organization. The
former term is applied to many larvas and nymphs (Fig. 211, C, D} on
account of their resemblance to Thysanura, of which Campodea and
Lepisma are types. The resemblance lies chiefly in the flattened form,
hard plates, long legs and antennae, caudal cerci, well-developed mandib-
ulate mouth parts, and active habits, with the accompanying sensory
specializations. These characteristics are permanent in Thysanura, but
only temporary in metamorphic insects, and their occurrence in the latter
forms may properly be taken to indicate that these insects have been
derived from ancestors which were much like Thysanura.

Thysanuriform characters are most pronounced in nymphs of Blat-
tidae, Forficulidae, Perlidae, Ephemeridae and Odonata, but occur also in
the larvae of some Neuroptera (Mantis pa) and Coleoptera (Carabidae

DEVELOPMENT

and Meloidae). These primitive characters are gradually overpowered,
in the course of larval evolution, by secondary, or adaptive, features.

FIG. 211. Types of larvae. A, B, Thysanura; C, D, thysanuriform nymphs; E-I, cruci-
form larvae. A, Campodea; B, Lepisma; C, perlid nymph (Plecoptera) ; D, Libcllula (Odo-
nata); E, Tenthredopsis (Hymenoptera) ; F, Laclinostcrna (Coleoptera); G, Melanotus
(Coleoptera) ; H, Bombus (Hymenoptera); 7, Hypoderma (Diptera).

Eruciform Larvae. The prevalent type of larva among holometab-
olous insects is the eruciform (Fig. 211, E-I), illustrated by a caterpillar

A

D

FIG. 212. Mantispa. A, larva at hatching thysatiwriform; B, same larva just before
first moult now becoming eruciform. C, imago, the wings omitted; D, winged imago,
slightly enlarged. A and B after BRAUER; C and D after EMERTON, from Packard's Text-
Book of Entomology, by permission of the Macmillan Co.

or a maggot. Here the body is cylindrical and often fleshy; the integ-
ument weak; the legs, antennae, cerci, and mouth parts reduced, often
to disappearance; the habits sedentary and the sense organs correspond-

132 ENTOMOLOGY

ingly reduced. These characteristics are interpreted as being results of
partial or entire disuse, the amount of reduction being proportional to
the degree of inactivity. Extreme reduction is seen in the maggots of
parasitic and such other Diptera as, securing their food with almost no
exertion, are simple in form, thin-skinned, legless, with only a mere
vestige of a head and with sensory powers of but the simplest kind.

Transitional Forms. The eruciform is clearly derived from the
thysanuriform type, as Brauer and Packard have shown, the continuity
between the two types being established by means of a complete series
of intermediate stages. The beginning of the eruciform type is found
in Neuroptera, where the campodeoid sialid larva assumes a quiescent
pupal condition. The key to the origin of the complete metamorphosis,
involving the eruciform condition, Packard finds in the neuropterous
genus Mantis pa (Fig. 212), the first larva of which is truly campodea-
form and active. Beginning a sedentary life, however, in the egg-sac of
a spider, it loses the use of its legs and the antennae become partly aborted,
before the first moult. In Packard's words, "Owing to this change of
habits and surroundings from those of its active ancestors, it changes its
form, and the fully grown larva becomes cylindrical, with small slender
legs, and, owing to the partial disuse of its jaws, acquires a small, round
head." Meloidae (Fig. 218) afford other excellent examples of the tran-
sition from the thysanuriform to the eruciform condition during the life
of the individual.

Thysanuriform characters become gradually suppressed in favor of
the eruciform, until, in most of the highly developed orders (Mecoptera,
Trichoptera, Lepidoptera, Diptera, Siphonaptera and Hymenoptera) ,
they cease to appear, except for a few embryonic traces an illustration
of the principle of "acceleration in development."

Growth. The larval period is pre-eminently one of growth. In
Heterometabola, growth is continuous during the nymphal stage, but
in Holometabola this important function becomes relegated to the larval
stage, and pupal development takes place at the expense of a reserve sup-
ply of food accumulated by the larva.

The rapidity of larval growth is remarkable. Trouvelot found that
the caterpillar of Telea polyphemus attains in 56 days 4,140 times its
original weight (^V grain), and has eaten an amount of food 86,000
times its primitive weight. Other larvae exceed even these figures; thus
the maggot of a common flesh fly attains 200 times its original weight in
24 hours.

Ecdysis. The exoskeleton, unfitted for accommodating itself to the

DEVELOPMENT I 33

growth of the insect, is periodically shed, and along with it go not only
such integumentary structures as hairs and scales, but also the chitinous
lining, or intima, of the stomoda^um, proctodaeum, tracheae, integumen-
tary glands, etc. The process of moulting, or ecdysis, in caterpillars is
briefly as follows. The old skin becomes detached from the body by an
intervening fluid of hypodermal origin; the skin dries, shrinks, is pushed
backward by the contractions of the larva, and at length splits near the
head, frequently under the neck; through this split appear the new head
and thorax, and the old skin is worked back toward the tail until the
larva is freed of its exuvice. The details of the process, however, are by
no means simple. Ecdysis is probably something besides a provision
for growth, for Collembola continue to moult long after growth has
ceased, and the winged May fly sheds its skin once after emergence.
The meaning of this is not known, though perhaps ecdysis has an excre-
tory importance in the case of Collembola, which are exceptional among
insects in having no Malpighian tubes.

Number of Moults. The frequency of moulting differs greatly in
different orders of insects. Acridiidae have five moults; Lepidoptera
usually four or five, but often more, as in Isia (Pyrrharctia) Isabella,
which moults as many as ten times (Dyar) ; Musca domestica has three
(Packard) ; the honey bee probably six (Cheshire) ; and the seventeen-
year locust about twenty-five or thirty (Riley). Packard suggests that
cold and lack of food during hibernation in arctians (as /. Isabella) and
partial starvation in the case of some beetles, cause a great number of
moults by preventing growth, the hypodermis cells meanwhile retaining
their activity.

The appearance of the insect often changes greatly with each moult,
particularly in caterpillars, in which the changes of coloration and
armature may have some phylogenetic significance, as Weismann has
attempted to show in the case of sphingid larvae.

Adaptations of Larvae. Larvae exhibit innumerable conformities
of structure to environment. The greatest variety of adaptive structures
occurs in the most active larvae, such as predaceous forms, terrestrial, or
aquatic. These have well-developed sense organs, excellent powers of
locomotion, special protective and aggressive devices, etc. In insects
as a whole, the environment of the larva or nymph and that of the adult
are very different, as in the dragon fly or the butterfly, and the larvae are
modified in a thousand ways for their own immediate advantage, with-
out any direct reference to the needs of the imago.

The chief purpose, so to speak, of the larva is to feed and grow, and

134 ENTOMOLOGY

the largest modifications of the larva depend upon nutrition. Take as
one extreme, the legless, headless, fleshy and sluggish maggot, embedded
in an abundance of food, and as the other extreme the active and "wide-
awake" larva of a carabid beetle, dependent for food upon its own powers
of sensation, locomotion, prehension, etc., and obliged meanwhile to
protect or defend itself. Between these extremes 'come such forms as
caterpillars, active to a moderate degree. The great majority of larval
characters, indeed, are correlated with food habits, directly or indirectly;
directly in the case of the mouth parts, sensory and locomotor organs,
and special structures for obtaining special food; indirectly, as in re-
spiratory adaptations and protective structures, these latter being numer-
ous and varied.

Larvae that live in concealment, as those that burrow in the ground
or in plants, have few if any special protective structures; active larvae,
as those of Carabidae, have an armor-like integument, but owe their pro-
tection from enemies chiefly to their powers of locomotion and their
aversion to light (negative phototropisni) ; various aquatic nymphs (Zaitha,
Odonata) are often coated with mud and therefore difficult to distin-
guish so long as they do not move; caddis worms are concealed in their
cases, and caterpillars are often sheltered in a leafy nest. There is no
reason to suppose that insects conceal themselves consciously, however,
and one is not warranted in speaking of an instinct for concealment in the
case of insects since everything goes to show that the propensity to
hide, though advantageous indeed, is simply a reflex, inevitable, negative
reaction to light (negative phototropism) or a positive reaction to contact
(positive thigmotropism) .

Exposed, sedentary larvae, as those of many Lepidoptera and Cole-
optera, often exhibit highly developed protective adaptations. Cater-
pillars may be colored to match their surroundings and may
resemble twigs, bird-dung, etc.; or larvae may possess a disagreeable
taste or repellent fluids or spines, these odious qualities being frequently
associated with warning colors.

Larvae need protection also against adverse climatal conditions,
especially low temperature and excessive moisture. The thick hairy
clothing of some hibernating caterpillars, as Isia (Pyrrharctia) Isabella,
doubtless serves to mollify sudden changes of temperature. Naked
cutworms hibernate in well-sheltered situations, and the grubs of the
common "May beetles," or "June bugs," burrow down into the ground
below the reach of frost. Ordinary high temperatures have little effect
upon larvae, except to accelerate their growth. Excessive moisture is
fatal to immature insects in general conspicuously fatal to the chinch

DEVELOPMENT

FIG. 213. Obtect
pupa of milkweed but-
terfly, Anosia plexippus,
natural size.

bug, Rocky Mountain locust, aphids and sawfly larvae. The effect of
moisture may be an indirect one, however; thus moisture may favor the
development of bacteria and fungi, or a heavy rain may be disastrous
not only by drowning larvae, but also by washing them off their food
plants.

As a result of secondary adaptive modifications, larvae may differ
far more than their imagines. Thus Platygaster in
its extraordinary first larval form (Fig. 219) is en-
tirely unlike the larvae of other parasitic Hymenop-
tera, reminding one, indeed, of the crustacean
Cyclops rather than the larva of an insect. As
Lubbock has said, the characters of a larva depend
(i) upon the group of insects to which the larva
belongs and (2) upon the special environment of
the larva.

Pupa. The term pupa is strictly applicable to
holometabolous insects only. Most Lepidoptera
and many Diptera have an obtect pupa (Fig. 213),
or one in which the appendages and body are com-
pactly united; as distinguished from the free pupa of Neuroptera, Trichop-
tera, Coleoptera and others, in which the appendages are free (Fig. 204).
This distinction, however, cannot always be drawn sharply. Diptera
present also the coarctate type of pupa (Fig. 205), in which the pupa re-
mains enclosed in the old larval skin, or puparium.

Pupal characters, though doubtless of great adaptive and phylogenetic
significance, have received but little attention. Lepi-
dopterous pupae present many puzzling characters, for
example, an eye-like structure (Fig. 214) suggesting
an ancestral active condition, such as still occurs
among heterometabolous insects.

Pupation of a Caterpillar. The process of pupa-
tion in a caterpillar has been carefully observed by
Riley. The caterpillar of the milkweed butterfly
(PI. I, A) spins a mass of silk in which it entangles
its suranal plate and anal prolegs and then hangs down-
ward, bending up the anterior part of the body (B), which gradually
becomes swollen. The skin of the caterpillar splits dorsally from the
head backward, and is worked back toward the tail (C and D) by the
contortions of the larva.

The way in which the pupa becomes attached to its silken support

FIG. 214. Head
of chrysalis of Pa-
pilio polyxenes, to
show eye-like struc-
ture. Enlarged.

i 3 6

ENTOMOLOGY

is rather complex. Briefly, while the larval skin still retains its hold on
the support, the posterior end of the pupa is withdrawn from the old
integument and by the vigorous whirling and twisting of the body the
hooks of the terminal cremaster of the pupa are entangled in the silken
support. At first the pupa is elongate (E) and soft, but in an hour or so
it has contracted, hardened, and assumed its characteristic form and
coloration (F).

Pupal Respiration. Except under special conditions, pupae breathe
by means of ordinary abdominal spiracles. Aquatic pupae have special
respiratory organs, such as the tracheal filaments of Simulium (Fig. 231),
and the respiratory tubes of Culex (Fig. 230).

Pupal Protection. Inactive and helpless, most pupae are concealed
in one way or another from the observation of enemies
and are protected from moisture, sudden changes of
temperature, mechanical shock and other adverse in-
fluences. The larvae of many moths burrow into the
ground and make an earthen cell in which to pupate;
a large number of coleopterous larvae (Lachnosterna,
Osmoderma, Passalus, Lucanus, etc.) make a chamber
in earth or wood, the walls of the cell being strength-
ened with a cementing fluid or more or less silk, form-
ing a rude cocoon. Silken cocoons are spun by some
Neuroptera (Chrysopidae, Fig. 215), by Trichoptera
(whose cases are essentially cocoons), Lepidoptera,
a few Coleoptera (as Curculionidae, Donacia), some
Diptera (as Cecidomyiidae), Siphonaptera, and many
Hymenoptera (for example, Tenthredinidas, Ichneu-
monidae, wasps, bees and some ants).

The cocoon-making instinct is most highly developed in Lepidoptera
and the most elaborate cocoons are those of Saturniidae. The cocoon
of Samia cecropia is a tough, water-proof structure and is double (Fig.
216), there being two air spaces around the pupa; thus the pupa is pro-
tected against moisture and sudden changes of temperature and from
most birds as well, though the downy woodpecker not infrequently punc-
tures the cocoon. S. cecropia binds its cocoon firmly to a twig; Tropcea
luna and Tdea polyphemus spin among leaves, and their cocoons (with
some exceptions) fall to the ground; Callosamia promethea, whose cocoon
is covered with a curved leaf, fastens the leaf to the twig with a wrapping
of silk, so that the leaf with its burden hangs to the twig throughout the
winter. The leaves surrounding cocoons may render them inconspicuous

FIG. 215. Co-
coon of Chrysopa ,
after emergence of
imago. Slightly en-
larged.

PLATE I.

D

F

Successive stages in the pupation of the milkweed caterpillar. Anoxia plcxippns.

Natural size.

DEVELOPMENT

139

or may serve merely as a foundation for the cocoon. While silk and
often a water-proof gum or cement form the basis of a cocoon, much
foreign material, such as bits of soil or wood, is often mixed in ; the cocoons
of many common Arctiidas, as Diacrisia virginica and Isia Isabella, con-
sist principally of hairs, stripped from the body of the larva.

Butterflies have discarded the cocoon, the last traces of which occur
in Hesperiidas, which draw together a few leaves with a scanty supply of
silk to make a flimsy substitute for a cocoon. Papilionid and pierid pupae
are supported by a silken girdle (Fig. 27), and nymphalid chrysalides
hang freely suspended by the tail (Fig. 213).

Cocoon-Spinning. The caterpillar of Telea polyphemus "feels
with its head in all directions, to discover any leaves to which to attach

FIG. 216. Cocoon of Samia cccropia, cut open to show the two silken layers and the enclosed

pupa. Natural size.

the fibres that are to give form to the cocoon. If it finds the place suit-
able, it begins to wind a layer of silk around a twig, then a fibre is attached
to a leaf near by, and by many times doubling this fibre and making it
shorter every time, the leaf is made to approach the twig at the distance
necessary to build the cocoon; two or three leaves are disposed like this
one, and then fibres are spread between them in all directions, and soon
the ovoid form of the cocoon distinctly appears. This seems to be the
most difficult feat for the worm to accomplish, as after this the work is
simply mechanical, the cocoon being made of regular layers of silk united
by a gummy substance. The silk is distributed in zigzag lines about
one-eighth of an inch long. When the cocoon is made, the worm will
have moved his head to and fro, in order to distribute the silk, about two
hundred and fifty-four thousand times. After about half a day's work,
the cocoon is so far completed that the worm can hardly be distinguished

140 ENTOMOLOGY

through the line texture of the wall; then a gummy resinous substance,
sometimes of a light brown color, is spread over all the inside of the cocoon.
The larva continues to work for four or five days, hardly taking a few
minutes of rest, and finally another coating is spun in the interior, when
the cocoon is all finished and completely air tight. The fibre diminishes
in thickness as the completion of the cocoon advances, so that the last
internal coating is not half so thick and so strong as the outside ones."
(Trouvelot.)

Emergence of Pupa. Subterranean pupae wriggle their way to the
surface of the ground, often by the aid of spines (Fig. 217), that catch
successively into the surrounding soil. These locomotor spines may occur
on almost any part of the pupa, but occur commonly on the abdominal
segments, as in lepidopterous pupae; the extremity of the abdomen, also,
bears frequently one or more spinous projections, as in Tipulidas, Cara-

bidae and Lepidoptera, to assist the escape of the pupa.
These structures are found also in pupae, as those of
Sesiidae, that force their way out of the stems of plants
in which the larvae have lived. The emergence from
the cocoon is accomplished in some cases by the pupa,
in others by the imago. Hemerobiidas, Trichoptera
and the primitive lepidopteron Eriocephala use the
pupal mandibles to cut an opening in the cocoon;
while many lepidopterous pupae have on the head a
beak for piercing the cocoon, or teeth for rending or
FIG. 2i 7 .-Sub- cutting the silk.

terranean pupa of

Anisota. Enlarged. Eclosion. During the last few hours before the

emergence of a butterfly the colors of the imago develop
and may be seen through the transparent skin of the chrysalis (PL II A}.
No movement occurs, however, until several seconds before emergence;
then, after a few convulsive movements of the legs and thorax of the
imprisoned insect, the pupa skin breaks in the region of the tongue and
legs (B), a secondary split often occurs at the back of the thorax, and
the butterfly emerges (C-E) with moist body, elongated abdomen and
miniature wings. Hanging to the empty pupa case (F), or to some other
available support, the insect dries and its wings gradually expand (G, H)
through the pressure of the blood. At regular intervals the abdomen
contracts and the wings fan the air, and sooner or later a drop or two of
a dull greenish fluid (the meconium) is emitted from the alimentary canal.
The expansion of the wings takes place rapidly, and in less than an hour,
as a rule, they have attained their full size (/).

Pl.ATl [I.

Successive stages in the emergence of the milkweed butterfly, Anoxia plexippus,

from the chrysalis. Natural size.

DEVELOPMENT 143

T. polyphemus is "provided with two glands opening into the mouth,
which secrete during the last few days of the pupa state, a fluid which is
a dissolvent for the gum so firmly uniting the fibres of the cocoon. This
liquid is composed in great part of bombycic acid. When the insect has
accomplished the work of transformation which is going on under the
pupa skin, it manifests a great activity, and soon the chrysalis covering
bursts open longitudinally upon the thorax ; the head and legs are soon
disengaged, and the acid fluid flows from its mouth, wetting the inside of
the cocoon. The process of exclusion from the cocoon lasts for as much
as half an hour. The insect seems to be instinctively aware [?] that
some time is required to dissolve the gum, as it does not make any at-
tempt to open the fibres, and seems to wait with patience this event.
When the liquid has fully penetrated the cocoon, the pupa contracts its
body, and pressing the hinder end, which is furnished with little hooks,
against the inside of the cocoon, forcibly extends its body; at the same
time the head pushes hard upon the fibres and a little swelling is observed
on the outside. These contractions and extensions of the body are re-
peated many times, and more fluid is added to soften the gum, until
under these efforts the cocoon swells, and finally the fibres separate, and
out comes the head of the moth. In an instant the legs are thrust out,
and then the whole body appears; not a fibre has been broken, they have
only been separated.

"To observe these phenomena, I had cut open with a razor a small
portion of a cocoon in which was a living chrysalis nearly ready to trans-
form. The opening made was covered with a piece of mica, of the same
shape as the aperture, and fixed to the cocoon with mastic so as to make
it solid and air-tight ; through the transparent mica I could see the move-
ments of the chrysalis perfectly well.

"When the insect is out of the cocoon, it immediately seeks for a
suitable place to attach its claws, so that the wings may hang down, and
by their own weight aid the action of the fluids in developing and un-
folding the very short and small pad-like wings. Every part of the in-
sect on leaving the cocoon, is perfect and with the form and size of ma-
turity, except the pad-like wings and swollen and elongated abdomen,
which still gives the insect a worm-like appearance; the abdomen con-
tains the fluids which flow to the wings.

"When the still immature moth has found a suitable place, it re-
mains quiet for a few minutes, and then the wings are seen to grow very
rapidly by the afflux of the fluid from the abdomen. In about twenty
minutes the wings attain their full size, but they are still like a piece of

144 ENTOMOLOGY

wet cloth, without consistency and firmness, and as yet entirely unfit for
flight, but after one or two hours they become sufficiently stiff, assuming
the beautiful form characteristic of the species." (Trouvelot.) The
expansion of the wing is due to blood-pressure brought about chiefly by
the abdominal muscles. In the freshly-emerged insect, the two mem-
branes of the wing are corrugated, and expansion consists in the flattening
out of these folds. The wing is a sac, which would tend to enlarge into
a balloon-shaped bag, were it not for hypodermal fibers which hold the
wing-membranes closely together (Mayer). Samia cecropia also uses
a dissolvent fluid ; Tropcea luna, Philosamia cynthia and others cut and
force an opening through the cocoon by means of a pair of saw-like organs,
one at the base of each front wing.

Hypermetamorphosis. In a few remarkable instances, metamor-
phosis involves more than three stages, owing to the existence of super-
numerary larval forms. This phenomenon of hypermetamorphosis occurs
notably in the coleopterous genera Meloe, Epicauta, Sitaris, Rhipiphorus
and Stylops, in male Coccidas and several parasitic Hymenoptera.

In Meloe, as described by Riley, the newly-hatched larva (triungulin}
form) is active and campodea-form. It climbs upon a flower and thence
upon the body of a bee (Anthophord), which carries it to the nest, where
it eats the egg of the bee. After a moult, the larva though still six-
legged, has become cylindrical, fleshy and less active, resembling a lamel-
licorn larva; it now appropriates the honey of the bee. With plenty
of rich food at hand the larva becomes sluggish, and after another moult
appears as a pseudo-pupa, with functionless mouth parts and atrophied
legs. From this pseudo-pupa emerges a third larval form, of the pure
cruciform type, fat and apodous like the bee-larvae themselves. After
these four distinct stages the larva becomes a pupa and then a beetle.

Epicauta, another meloid, has a similar history. The triungulin
(Fig. 218, A) of E. vittata burrows into an egg-pod of Mclanoplus differen-
tiates and eats the eggs of that grasshopper. After a moult the second
larva (carabidoid form) appears; this (B) is soft, with reduced legs and
mouth parts and less active than the triungulin. A second moult and
the scarabceidoid form of the second larva is assumed; the legs and
mouth parts are now rudimentary and the body more compact than
before. A third and a fourth moult occur with little change in the form
of the second larva, which is now in its ultimate stage (C). After the
fifth moult, however, the coat elate larva, or pseudo-pupa, appears; this
(D) hibernates and in spring sheds its skin and becomes the third larva,
which soon transforms to a true pupa (E), from which the beetle (F)

DEVELOPMENT

shortly emerges. Thus the pupal stage is preceded by at least three
distinct larval stages.

In the anomalous beetle Stylo ps, the males are winged, but the fe-
males are maggot-like and sedentary, living in the bodies of bees and
wasps. Packard found as many as three hundred triungulin larva.-
issuing from a female Stylo ps in the body of an Andrcna. The further
life history of Stylo ps is but imperfectly known; probably the triungulin
climbs upon a bee or a wasp and enters its body, after the manner of the
European Rhipiphorus paradox us, whose life history is much better
understood.

FIG. 218. Stages in the hypermetamorphosis of Epicauta. A, triungulin; B, carabidoid
stage of second larva; C, ultimate stage of second larva; D, coarctate larva; E, pupa; F,
imago. E is species cinerea; the others are viltata. All enlarged except F. After RILEY,
from Trans. St. Louis Acad. Science.

The most extraordinary metamorphoses have been found among
parasitic Hymenoptera, as in Platygaster, a proctotrypid which infests
the larva of Cecidomyia. The egg of Platygaster, according to Ganin,
hatches into a larva of bizarre form (Fig. 219, A), suggesting the crusta-
cean Cyclops, rather than an insect. This first larva has a blind food
canal and no nervous, circulatory or respiratory systems. After a moult
the outline is oval (B}, and there are no appendages as yet, though the
nervous system is partially developed. Another moult, and the third
larva appears (C), elliptical in contour, externally segmented, with
tracheae and a pair of mandibles. From now on, the development is
essentially like that of other parasitic Hymenoptera.

ii

146

ENTOMOLOGY

Equally anomalous are the changes undergone by Polynema, a proc-
totrypid parasite in the eggs of dragon flies, and by the proctotrypid
Teleas, which affects the eggs of the tree cricket (CEcanthus). In all
these cases the larvae go through changes which in most other insects are
confined to the egg stage. In other words, the larva hatches before its
embryonic development is completed, so to speak.

Significance of Metamorphosis. "The essential features of meta-
morphosis," says Sharp, "appear to be the separation in time of growth
and development and the limitation of the reproductive processes to a
short period at the end of the individual life."

mo

- m

..-m

FIG. 219. Stages in the hypermetamorphosis of Platygaster, A, first larva; B, second
larva; C, third larva; a, antenna; b, brain; /, fat-tissue; h, hind intestine; m, mandible;
mo, mouth; MS, muscle; n, nerve cord; r, reproductive organ of one side; s, salivary gland;
I, trachea. After GANIN.

The simplest insects, Thysanura, have no metamorphosis, and show
no traces of ever having had one. Hence it is inferred that the first
insects had none; in other words, the phenomenon of metamorphosis
originated later than insects themselves. Successive stages in the evolu-
tion of metamorphosis are illustrated in the various orders of insects.

The distinctive mark of the simplest metamorphosis, as in Orthop-
tera and Hemiptera, is the acquisition of wings; growth and sexual
development proceeding essentially as in the non-metamorphic insects
(Thysanura and Collembola). Here the development of wings does not
interfere with the activity of the insect; its food habits remain unaltered;
throughout life the environment of the individual is practically the same.

DEVELOPMENT 147

Even when considerable difference exists between the nymphal and
imaginal environments, as in Ephemerida and Odonata, the activity of
the individual may still be continuous, even if somewhat lessened as the
period of transformation approaches.

With Neuroptera, the pupal stage appears. In these and all other
holometabolous insects the larva accumulates a surplus of nutriment
sufficient for the further development, which becomes condensed into
a single pupal stage, during which external activity ceases temporarily.

With the increasing contrast between the organization of the larva
and that of the imago, the pupal stage gradually becomes a necessity.
Metamorphosis now means more than the mere acquisition of wings, for
the larva and the imago have become adapted to widely different en-
vironments, chiefly as regards food. The caterpillar has biting mouth
parts for eating leaves, while the adult has sucking organs for obtaining
liquid nourishment; the maggot, surrounded by food that may be ob-
tained almost without exertion, has but minimum sensory and locomotor
powers and for mouth parts only a pair of simple jaws; as contrasted
with the fly, which has wings, highly developed mouth parts and sense
organs, and many other adaptations for an environment which is strik-
ingly unlike that of the larva; so also in the case of the higher Hymen-
optera, where maternal or family care is responsible for the helpless con-
dition of the larva.

Thus it is evident that the change from larval to imaginal adapta-
tions is no longer congruous with continuous external activity; a quies-
cent period of reconstruction becomes inevitable.

As was said, the cruciform type of larva has been derived from the
thysanuriform type, the strongest evidence of this being the fact that
among hypermetamorphic insects, the change from the one to the other
takes place during the lifetime of the individual. Furthermore, the
cruciform condition is plainly an adaptive one, brought about by an
abundant and easily obtainable supply of food. The lack of a thysanuri-
form stage in the development of the most specialized cruciform larvae,
as those of flies and bees, is regarded by Hyatt and Arms as an illustra-
tion of the general principle known as "acceleration of cfevelopment,''
according to which newer and useful adaptive characters tend to appear
earlier and earlier in the development, gradually crowding upon and
forcing out older and useless characters. In connection with this sub-
ject, the appearance of temporary abdominal legs in embryo bees is
significant, as indicating an ancestral active condition. In accounting for

148

ENTOMOLOGY

the evolution of metamorphosis, the theory of natural selection finds
one of its most important applications.

3. INTERNAL METAMORPHOSES

In Heterometabola, the internal post-embryonic changes are as di-
rect as the external changes of form; in Holometabola, on the contrary,
not all the larval organs pass directly into imaginal organs, for certain
larval tissues are demolished and their substance reconstructed into
imaginal tissues. When indirect, however, the internal metamorphosis

FIG. 220.-- Diagram-
matic transverse section of
Corethra larva, to show
imaginal buds of wings (w)
and legs (/); //, hypoder-
mis; 7, integument. Modi-
fied from Lang's Lehrbitcli.

IV:

.W

W:

.W

I

D

FIG. 221. Diagrammatic transverse sections of muscid
larvae, to show imaginal buds. //, larval hypodermis; /, larval
integument; ih, imaginal hypodermis; /, imaginal bud of leg;
w, imaginal bud of wing. Modified from Lang's Lehrbuch.

is nevertheless continuous and gradual, without the abruptness that
characterizes the external transformation. In the larval stage imaginal
organs arise and grow; in the pupa stage the purely larval organs grad-
ually disappear while the imaginal organs are continuing their develop-
ment.

Phagocytes. The destruction of larval tissues, or histolysis, is due
often to the amoeboid blood corpuscles, known as leucocytes or p/iago-

DEVELOPMENT

149

cytcs, which attack some tissues and absorb their material, but later are
themselves food for the developing imaginal tissues. The construction
of tissues is termed histogencsis.

In Coleoptera, however, the degeneration of the larval muscles is entire-
ly chemical, there being no evidence of phagocytosis, according to Dr. R.
S. Breed. Berlese, indeed, goes so far as to deny in general the destruc-
tive action of leucocytes on larval tissues.

Imaginal Buds. The wings and legs
of a fly originate in the larva in the form
of cellular masses, or imaginal buds, as
Weismann discovered. Thus in the larva
of Corethra, there are in each thoracic seg-
ment a pair of dorsal buds and a pair of
ventral buds (Fig. 220), each bud being
clearly an evagination of the hypodermis
at the bottom of a previous imagination.
The six ventral buds form the legs eventu-
ally; of the dorsal buds, the middle and
posterior pairs form, respectively, the
wings and the halteres, and the anterior
pair form the pupal respiratory processes.
Each imaginal bud is situated in a peri-
podal cavity, the wall of which (peripodal
membrane) is continuous with the general
hypodermis; as the legs and wings develop,
they emerge from their peripodal sacs and
become free.

In Corethra but little histolysis occurs,
most of the larval structures passing direct-
ly into the corresponding structures of the
adult. Corethra, indeed, is in many re-
spects intermediate between heterome-
tabolous and holometabolous insects as
regards its internal changes.

Muscidae. In Muscidae, as compared with Corethra. the imaginal
buds are more deeply situated, the peripodal membrane forming a stalk
(Fig. 221), and the processes of histolysis and histogenesis become ex-
tremely complicated. The hypodermis, muscles, alimentary canal and
fat-body are gradually broken down and remodeled, and part of the
respiratory system is reorganized, though the dorsal vessel and the central

FIG. 222. Imaginal buds of full
grown larva of Picris, dorsal aspect.
b, brain; ?, mid' intestine; s 1 , pro-
thoracic spiracle; s 4 , first abdominal
spiracle; sg, silk gland; /, pro-
thoracic bud; II, bud of fore wing;
///, bud of hind wing. After
GONIN.

ENTOMOLOGY

FIG. 223. Section through left hind wing in larva of Picris rapce, the section being a frontal
one of the caterpillar; the base of the wing is anterior in position, and the apex posterior, c,
cuticula; h, hypodermis; /, trachea; w, developing wing. After MAYER.

v-..

B

FIG. 224. Internal transformations of Sphinx ligustri. A, larva; B, pupa; C, moth;
a, aorta; an, antenna; b, brain; /, fore intestine; fr, food reservoir; h, hind intestine; ht,
heart; m, mid intestine; mt, Malpighian tubes; p, proboscis; s, subcesophageal ganglion;
/, testis; lg, thoracic ganglia; v, ventral nerve cord. After NEWPORT.

DEVELOPMENT 151

nervous system, uninterrupted in their functions, undergo comparatively
little alteration.

The imaginal hypodermis of the thorax arises from thickenings of
the peripodal membrane which spread over the larval hypodermis, while
the latter is gradually being broken down by the leucocytes; in the head
and abdomen the process is essentially the same as in the thorax, the
new hypodermis arising from imaginal buds.

Most of the larval muscles, excepting the three pairs of respiratory
muscles, undergo dissolution. The imaginal muscles have been traced
back to mesodermal cells such as are always associated with imaginal
buds.

Hymenoptera and Lepidoptera. The internal transformation in
Hymenoptera, according to Bugnion, is less profound than in Muscidas
and more extensive than in Coleoptera and Lepidoptera. The internal
metamorphosis in Lepidoptera resembles in many respects that of Core-
thra. In both these orders the dorsal pair of prothoracic buds is absent.
In a full-grown caterpillar the fundaments of the imaginal legs and wings
(Fig. 222) may be seen, the wings in a frontal section of the larva ap-
pearing as in Fig. 223. Many of the details of the internal metamorphosis
in Lepidoptera have been described by Newport and Gonin. Figure 224,
after Newport, shows some of the more evident internal differences in
the larva, pupa and imago of a lepidopterous insect.

Significance of Pupal Stage. To repeat among holometabolous
insects the function of nutrition becomes relegated to the larval stage
and that of reproduction to the imaginal stage. Larva and imago be-
come adapted to widely different environments. So dissimilar are the
two environments that a gradual change from the one to the other is no
longer possible; the revolutionary changes in structure necessitate a
temporary cessation of external activity.

CHAPTER IV

ADAPTATIONS OF AQUATIC INSECTS

Ease, versatility and perfection of adaptation are beautifully exempli-
fied in aquatic insects.

Systematic Position. Aquatic insects do not form a separate group
in the system of classification, but are distributed among many orders, of
which Plecoptera, Ephemerida, Odonata and Trichoptera are pre-emi-
nently aquatic. One third of the families of Heteroptera and less than
one fourth those of Diptera are more or less aquatic. One tenth of the
families of Coleoptera frequent the water at one stage or another, but
only half a dozen genera of Lepidoptera. A few Collembola live upon
the surface of water; and several Hymenoptera, though not strictly
aquatic, are known to parasitize the eggs and larvae of aquatic insects.

The change from the terrestrial to the aquatic habit has been a gradual
change of adaptation, not an abrupt one. Thus at present there are some
tipulid larvae that inhabit comparatively dry soil; others live in earth
that is moist; many require a saturated soil near a body of water and
many, at length, are strictly aquatic. Among beetles, also, similar transi-
tional stages are to be found.

Food. Insects have become adapted to utilize with remarkable
success the immense and varied supply of food that the water affords.
Hosts of them attack such parts of plants as project above the surface of
the water, and the caterpillar of Paraponyx (Fig. 172) feeds on submerged
leaves, especially of VaUisneria, being in this respect almost unique
among Lepidoptera. Hydrophilid beetles and many other aquatic in-
sects devour submerged vegetation. The larvae of the chrysomelid genus
Donacia find both nourishment and air in the roots of aquatic plants.
Various Collembola subsist on floating algae, and larvae of mosquitoes and
black-flies on microscopic organisms near the surface, while larvae of
Chironomus find food in the sediment that accumulates at the bottom of a
body of water.

Predaceous species abound in the water. Notonecta (Fig. 225) ap-
proaches its prey from beneath, clasps it with the front pair of legs and
pierces it. Nepa and Ranatra likewise have prehensile front legs along

ADAPTATIONS OF AQUATIC INSECTS

153

with powerful piercing organs. Belostoma and Benacus (Fig. 22) even
kill small fishes by their poisonous punctures. Some other kinds, as the
water-skaters (Gerridae, Fig. 226), depend on dead or disabled insects.
The species of Hydro philus (Fig. 227) are to some extent carnivorous as
larvae but phytophagous as imagines, while Dytiscidae are carnivorous

FIG. 225. Backswimmer, Xotonccta insiilala,
natural size.

FIG. 226. Water-skater, Gcrris remigis,
natural size.

throughout life. Aquatic insects eat not only other insects, but also
worms, crustaceans, mollusks or any other available animal matter.

Even aquatic insects are not exempt from the attacks of parasitic
species. A few Hymenoptera actually enter the water to find their
victims, for example, the ichneumon Agriotypus,
which lays its eggs on the larvae of caddis flies.

Locomotion. Excellent adaptations for aquat-
ic locomotion are found in the common Hydro phi-
lus triangular is (Fig. 227). Its general form re-
minds one of a boat, and its long legs resemble
oars. The smoothly elliptical contour and the
polished surface serve to lessen resistance. Owing
to the form of the body (Fig. 228, A] and the pres-
ence of a dorsal air-chamber under the elytra, the
back of the insect tends to remain uppermost, while
in Notonecta (Fig. 228, B), on the other hand, the
conditions are reversed, and the insect swims with
its back downward. The legs of Hydro philus, ex-
cepting the first pair, are broad and thin (Fig. 229,^!)

and the tarsi are fringed with long hairs. When swimming, the "stroke "
is made by the flat surface, aided by the spreading hairs; but on the
"recover," the leg is turned so as to cut the water, while the hairs fall
back against the tarsus from the resistance of the water, as the leg is
being drawn forward. The hind legs, being nearest the center of gravity.

FIG. 227. Hydroph-
triaii^idiris, nat-
ural size.

154

ENTOMOLOGY

are of most use in swimming, though the second pair also are used for
this purpose; indeed, a terrestrial insect, finding itself in the water,

FIG. 228. Transverse sections of (A) Hydrophilus and (B) Notonecta. e, elytron; h, hemely-

tron; /, meta thoracic leg.

instinctively relies upon the third pair of legs for locomotion. Hydroph-
ilus uses its oar-like legs alternately, in much the same sequence

as land insects, but Cybister
and other Dytiscidae, which
are even better adapted
than Hydrophilus for aquatic
locomotion, move the hind
legs simultaneously, and
therefore can swim in a
straight line, without the
wobbling and less econom-
ical movements that charac-
terize Hydrophilus.

Larvae of mosquitoes pro-
pel themselves by means of
lashing, or undulatory, move-
ments of the abdomen. A
peculiar mode of locomo-
tion is found in dragon fly
nymphs, which project them-
selves by forcibly ejecting a
stream of water from the
anus.

On account of the large
amount of air that they carry

about, most aquatic imagines are lighter than the water in which they live,
and therefore can rise without effort, but can descend only by exertion, and
can remain below only by clinging to chance stationary objects. The mos-

FIG. 229. Left hind legs of aquatic beetles. A,
Hydrophilus triangularis; B, Cybister fimbriolatus; c,
coxa; /, femur; s, spur; t, tarsus; ti, tibia; tr, tro-
chanter.

ADAPTATIONS OF AQUATIC INSECTS

155

quito larva (Fig. 230, A} is often heavier than water, but the pupa (Fig.
230, B) is lighter, and remains clinging to the surface film.

The tension of this surface film is sufficient, to support the weight of
an insect up to a certain limit, provided
the insect has some means of keeping its
body dry. This is accomplished usually by
hairs, set together so thickly that water
cannot penetrate between them. As the
legs and body of Gerris are rendered water-
proof by a velvety clothing of hairs, the in-
sect, though heavier than water, is able to
skate about on the surface. Gyrinus, by
means of a similar adaptation, can circle
about on the surface film, and minute col-
lembolans leap about on the surface as
readily as on land.

The modifications of the legs for swim-
ming have often impaired their usefulness
for walking, so that many aquatic Coleop-
tera and Hemiptera can move but awk-
wardly on land. When walking, it is inter-
esting to note, Cybister and some other
aquatic forms no longer move their hind
legs simultaneously as they do in swimming,
but use them alternately, like terrestrial
species.

The adaptations for swimming do not
necessarily affect the power of flight. Dy-
tiscus, Hydrophilus, Gyrinus, Notonecta,
Benacus and many other Coleoptera and
Hemiptera leave the water at night and fly
around, often being found about electric
lights.

Respiration. Aquatic insects have not
only retained the primitive, or open (holo-

pneustic), type of respiration, characterized by the presence of spiracles,
but have also developed an adaptive, or closed (apneustic) , type, for
utilizing air that is mixed with water.

Through minor modifications of structure and habit, many holo-
pneustic insects have become fitted for an aquatic life. In these in-

FIG. 230. Larva (.4) and pupa
(B) of mosquito, Culcx pipiens. r,
respiratory tube; /, tracheal gills.

156 ENTOMOLOGY

stances the insects have some means of carrying down a supply of air
from the surface of the water. Thus Xotonecta bears on its body a
silvery film of air entangled in closely set hairs, which exclude the water.
Gyrinus descends with a bubble of air at the end of the abdomen. Dy-
tiscus and Hydro philus have each a capacious air-space between the
elytra and the abdomen, into which space the spiracles open. Nepa
and Ranatra have each a long respiratory organ composed of two valves,
which lock together to form a tube that communicates with the single
pair of spiracles situated near the end of the abdomen. The mosquito
larva, hanging from the surface film, breathes through a cylindrical tube
(Fig. 230, A, r) projecting from the penultimate abdominal segment;
the pupa, however, bears a pair of respiratory tubes on the back of the
thorax (Fig. 230, B, r, r), which is now upward, probably in order to
facilitate the escape of the fly. The rat-tailed maggot (Eristalis), three
quarters of an inch long, has an extensile caudal tube seven times that
length, containing two tracheae terminating in spiracles, through which
air is brought down from above the mud in which the larva lives. Sim-
ilarly, in the dipterous larva, Bittacomor pha clampes (Fig. 173), the
posterior segments of the abdomen are attenuated to form a long re-
spiratory tube. The larva of Donacia appears to have no special ad-
aptations for aquatic respiration except a pair of spines near the end of
the body, for piercing air chambers in the roots of the aquatic plants
in which it dwells.

The simplest kind of apneustic respiration occurs in aquatic nymphs
such as those of Ephemerida and Agrionidae, whose skin at first is thin
enough to allow a direct aeration of the blood. This cutaneous res-
piration is possible during the early life of many aquatic species.

Branchial respiration, however, is the prevalent type among aquatic
nymphs and is perhaps the most important of their adaptive character-
istics. Thin-walled and extensive outgrowths of the integument, con-
taining tracheal branches or, rarely, only blood, enable these forms to
obtain air from the water. May fly nymphs (Figs. 19, A; 170), with
their ample waving gills, offer familiar examples of branchial respiration.
Tracheal gills are very diverse in form and situation, occurring in a few
species of May fly nymphs on the thorax or head, though commonly re-
stricted to the sides of the abdomen, where they occur in pairs or in
paired clusters (Fig. 19, A}. Caudal gills are found in agrionid nymphs
(Fig. 171). The aquatic caterpillars of Paraponyx (Fig. 172) are unique
among Lepidoptera in having gills, which are filamentous in this instance.

Caddis worms, enclosed in their cases, maintain a current of water by

ADAPTATIONS OF AQUATIC INSECTS

157

means of undulatory movements of the body, and the larva? and pupa-
of most black flies (Simuliidae, Fig. 231) secure a continuous supply of
fresh air simply by fastening themselves to rocks in swiftly flowing
streams.

Rectal respiration is highly developed in odonate nymphs. In these,
the rectum is lined with thousands of tracheal branches, which are bathed
by water drawn in from behind, and then expelled.

All these kinds of respiration cutaneous, branchial and rectal
occur in young ephemerid nymphs; while mosquito larvae have in ad-
dition spiracular respiration.

With the arrival of imaginal life, tracheal gills disappear, except in
Perlidae, and even in these insects the gills are of
little, if any, use.

Marine Insects. Except along the shore, the
sea is almost devoid of insect life, the exceptions
being a few chironomid larvae which have been
dredged in deep water, and fifteen species of Halo-
bates (belonging to the same family as our familiar
pond-skaters), which are found on warm smooth
seas, where they subsist on floating animal re-
mains.

Between tide-marks maybe found various beetles
and collembolans, which feed upon organic debris;
as the tide rises, the former retreat, but the latter
commonly burrow in the sand or under stones and
become submerged, for example the common A nurida
maritima.

Insect Drift. Seaweed or other refuse cast upon the shore harbors
a great variety of insects, especially dipterous larvae, staphylinid scaven-
gers and predaceous Carabidae. On the shores of inland ponds and lakes
a similar assemblage of insects may be found feeding for the most part
on the remains of plants or animals, or else on one another. During a
strong wind, the leeward shore of a lake is an excellent collecting ground,
as many insects are driven against it. On the shores of the Great Lakes
insects are occasionally cast up in immense numbers, forming a broad
windrow, fifty or perhaps a hundred miles long. Needham has described
such an occurrence on the west shore of Lake Michigan, following a gale
from the northeast. In this instance, a liter of the drift contained nearly
four thousand insects, of which 66 per cent, were crickets (Nemobius),
20 per cent. Acridiidae, and the remainder mostly beetles (Carabidae,

FIG.

limn; A, larva; B,
pupa, showing respira-
tory filaments.

158 ENTOMOLOGY

Scarabaeidae, Chrysomelidae, Coccinellidae, etc.), dragon flies, moths,
butterflies (Anosia, Pieris, etc.) and various Hemiptera, Hymenoptera
and Diptera. A large proportion of the insects were aquatic forms, such
as Hydro philus, Cybister, Zaitha, and a species of caddis fly; these had
doubtless been carried out by freshets, while the butterflies and dragon
flies had been borne out by a strong wind from the northwest, after which
all were driven back to the coast by a northeast wind. While some of
these insects survived, notably Coccinellidae, Trichoptera, Asilidae,
Acridiidee and Gryllidae, nearly all the rest were dead or dying, in-
cluding the dragon flies, flies, bumble bees and wasps. Foraging Cara-
bidae were observed in large numbers, also scavengers of the families
Staphylinidae, Silphidae and Dermestidas.

On the seashore and on the shores of the Great Lakes, the salient
features of insect life are essentially the same. Similar species occur in
the two places with similar biological relations, on account of the general
similarity of environment.

Origin of the Aquatic Habit. The theory that terrestrial insects
have arisen from aquatic species is no longer tenable, for the evidence
shows that the terrestrial type is the more primitive. Aquatic insects
still retain the terrestrial type of organization, which remains unob-
scured by the temporary and comparatively slight adaptations for an
aquatic life. Thus, the development of tracheal gills has involved no
important modification of the fundamental plan of tracheal respiration.
It is significant, moreover, that the most generalized, or most primitive,
insects Thysanura are without exception terrestrial. Aquatic in-
sects do not constitute a phylogenetic unit, but represent various orders,
which are for the most part undoubtedly terrestrial, notwithstanding the
fact ^that a few of these orders (Plecoptera, Ephemerida, Odonata, Tri-
choptera) are now wholly aquatic in habit. Adaptations for an aquatic
existence have arisen independently and often in the most diverse
orders of insects.

CHAPTER V

COLOR AND COLORATION

The naturalist distinguishes between the terms color and coloration.
A color is a single hue, while coloration refers to the arrangement of colors.

Sources of Color. The colors of insects are classed as (i) pigmental
(chemical}, those due to internal pigments; (2) structural (physical},
those due to structures that cause interference or reflection of light;
and (3) combination colors (chemico-physical), which are produced in both
ways at once.

Structural Colors. The iridescence of a fly's wing and that of a
soap bubble are produced in essentially the same way. The wing, how-
ever, consists of two thin, transparent, slightly separated lamellae, which
diffract white light into prismatic rays, the color differences depending
upon differences in the distance between the two membranes.

The brilliant iridescent hues of many butterfly scales are due to the
diffraction of light by fine, closely parallel striae (Fig. 92) just as in the
case of the " diffraction gratings" used by the physicist, which consist of
a glass or metallic plate with parallel diamond rulings of microscopic
fineness. The particular color produced depends in both cases upon the
distance between the striae. Though almost all lepidopterous scales are
striated, it is only now and then that the striae are sufficiently close to-
gether to give diffraction colors. In a Brazilian species of Apatura the
iridescent scales have 1050 striae to the millimeter, and in a species of
Morpho, according to Kellogg, the iridescent pigmented scales have 1400
striae per millimeter, the striae being only .0007 mm. apart; while in some
of the finest Rowland gratings they number only 700 per millimeter.

These interference colors of butterfly scales may be due, not only to
surface markings, but also to the lamination of the scale and to the over-
lapping of two or more scales. In beetles the metallic blues and greens,
and iridescence in general, are often produced by minute lines or pits that
diffract the light. Purely structural colors, however, are not so common
as might be supposed, according to Tower, who says, "The pits alone,
however, are powerless to produce any color; it is only when they are
combined with a highly reflecting and refractive surface lamella and a
pigmented layer below that the iridescent color appears. The action of

159

l6o ENTOMOLOGY

light is in this case the same as in the plain metallic coloring, excepting
that each pit acts as a revolving prism to disperse different wave-lengths
of light in different directions, and the combined result is iridescence.
The existence of minute pits over the body surface is of common occur-
rence, but it is only when they are combined as above that iridescent
colors occur."

Silvery white effects are usually caused by the total reflection of light
from scales or other sacs that are filled with air ; the same silvery appear-
ance is given also by air-filled tracheae and by the air bubbles that many
aquatic insects carry about under water.

Violet, blue-green, coppery, silver and gold colors are, with few excep-
tions, structural colors. (Mayer.)

Pigmental Colors. These are either cuticular or hypodermal. The
predominant brown and black colors of insects are made by pigment dif-
fused in the outer layer of the cuticula (Fig. 88). Cockroaches are almost
white just after a moult, but soon become brown, and many beetles change
gradually from brown to black. In these cases it is apparently significant
that the cuticular pigments lie close to the surface of the skin, i. e., where
they are most exposed to atmospheric influences. Tower holds that
cuticular colors "are not due to drying, oxidation, secretion, or like pro-
cesses," but are due to "some katalytic agent or enzyme [formed by the
hypodermis] which, passing out through the pore canals, comes in con-
tact with the primary cuticula and there becomes the active factor in the
production of cuticular colors." Gortner finds, however, that the black
cuticular pigment in Leptinotarsa belongs to the group of melanins and
is produced by oxidation, induced by an oxidase; that when all oxygen is
absent no pigmentation takes place.

The cuticular .pigments are derived, of course, from the underlying
hypodermis cells, and these cells themselves, moreover, usually contain
(i) colored granules or fatty drops which give red, yellow, orange and
sometimes white or gold colors as seen through the skin; (2) diffused
chlorophyll (green) or xanthophyll (yellow), taken from the food plant.
Unlike the structural colors, which are persistent, these hypodermal
colors often change after death, though less rapidly when the pigments are
tightly enclosed, as in scales or hairs. Though white and green are
structural colors as a rule, they are due to pigments in Pieridae, Lycaeni-
dae and some Geometridae.

Frequently a color pattern consists partly of cuticular and partly of
hypodermal colors, the hypodermal or sub-hypodermal color forming "a
groundwork upon which the pattern is cut out by the cuticular color."

COLOR AND COLORATION l6l

(Tower.) Thus in Leptinotarsa decemlineata the pattern "is composed
of a dark cuticular pigment upon a yellow hypodermal background."

Combination Colors. -The splendid changeable hues of Apatura,
Euplcea and other tropical butterflies depend upon the fact that their
scales are both pigmented and striated. Under the microscope, certain
Apatura scales are brown by transmitted light and violet by reflected
light, and to the unaided eye the color of the wing is either brown or
violet, according as the light is received respectively from the pigment
or from the striated surfaces of the scales. According to Tower, chemico-
physical colors " which are of exceedingly wide occurrence, are also the
most brilliant and varied of all those found in insects. To this class be-
long all metallic iridescent, pearly, and translucent colors, as well as blue,
green, and violet in almost every case."

Nature of Pigments. Some pigments are taken bodily from the
food; others are manufactured indirectly from the food, and some of
these are excretory products.

The green color of many caterpillars and grasshoppers is due to chloro-
phyll, which tinges the blood and shows through the transparent integu-
ment. Mayer has found that scales of Lepidoptera contain only blood
while the pigment is forming; that the first color to appear upon the pupal
wings is a dull ochre or drab the same color that the blood assumes when
it is removed from the pupa and exposed to the air; also that pigments
like those of the wings may be manufactured artificially from pupal
blood. Pieridas are peculiar in the nature of their pigments, as Hopkins
has shown. The white pigment of this family is uric acid and the reds
and yellows of Pieris, C alias and Papilla are due to derivatives of uric
acid; the yellow pigment, termed lepidotic acid, precedes the red in time
of appearance, the latter being probably a derivative of the former. The
green pigments of some Papilionidae, Noctuidae, Geometridae and Sphin-
gidae are also said by some investigators to be products of uric acid, which
in insects as in other animals is primarily an excretory, or waste, product.

Effects of Food on Color. Besides chlorophyll, to which various
caterpillars, aphids and other forms owe their green color, the yellow con-
stituent of chlorophyll, namely xanthophyll, frequently imparts its color
to plant-eating insects, while some phytophagous species are dull yellow
or brown from the presence of tannin, taken from the food plant. Most
pigments, however, are elaborated from the food by chemical processes
that are not well understood.

Many who have reared Lepidoptera extensively know that the color
of the imago is influenced by the character of the larval food, other con-

12

1 62 ENTOMOLOGY

ditions being equal, and are able at will to effect certain color changes
simply by feeding the larvae from birth upon particular kinds of plants.
In this country we have few observations upon the subject, but in Europe
the effects of food upon coloration have been ascertained in the case of
many species of Lepidoptera. According to Gregson, Hybernia defoliaria
is richly colored when fed upon birch, but is dull colored and almost un-
marked when fed on elm. Pictet, by feeding larvae of Vanessa urticce, on
the flowers instead of the leaves of the nettle obtained the variety known
as urticoides. Food affects the color of the larva also, as Poulton found
in the case of caterpillars of Tryphcena pronuba, all from the same batch
of eggs. When fed with only the white midribs of cabbage leaves, the
larvae remained almost white for a time, but afterward showed a moderate
amount of black pigment ; when fed with the yellow etiolated heart-leaves
or the dark green external leaves, however, the larvae all became bright
green or brown the same pigment being derived indifferently from etio-
lin (probably the same substance as xanthophyll) or chlorophyll.

Though the pigments may differ in color or amount according to the
kind of food, the color patterns vary without regard to food. Thus
Callosamia promethea, Leptinotarsa decemlineata (Colorado potato beetle),
Coccinellidae (lady-bird beetles) and a host of other insects exhibit ex-
tensive individual variations in coloration under precisely the same food
conditions. Caterpillars of the same kind and age are often very dif-
ferently marked when feeding upon the same plant; for example, Helio-
this obsoleta (corn worm) and the sphingid Deilephila lineata. Further-
more, striking changes of coloration accompany each moult in most cater-
pillars, but particularly those of butterflies, and these changes may prove
to have an important phylogenetic significance. Individual differences
of coloration apart from those due to the direct action of food, light,
temperature and other environmental conditions are to be explained by
heredity.

Effects of Light and Darkness. Sunlight is an important factor in
the development of most animalpigments, as they will not develop in its
absence. The collembolan Anurida maritima is white at hatching, but
soon becomes indigo blue, unless shielded from sunlight, in which event
it remains white until exposed to the sunlight, when it assume the blue
color. Subterranean or wood-boring larvae are commonly white or yel-
low, but never highly colored. The most notable instances, however,
are furnished by cave insects. These, like other cavernicolous animals,
are characteristically white or pale from the absence of pigment, if they
live in regions of continual darkness, but have more or less pigmentation

COLOR AND COLORATION 163

in proportion respectively to the greater or less amount of sunlight to
which they have access.

Curiously enough, light often hastens the destruction of pigment in
insects that are no longer alive, for which reason it is necessary to keep
cabinet specimens in the dark as much as possible. Life is evidently es-
sential for the sustention or renewal of the pigments.

A chrysalis not infrequently matches its surroundings in color. This
phenomenon has been investigated by Poulton, who has proved that the
color of the chrysalis is determined largely by the prevalent color of the
surroundings during the last few days of larval life. Larvae of Pieris rapce,
raised upon the same food plant (all other conditions being made as nearly
equal as possible) produced dark pupae if kept in darkness for a few days
just before pupation; yellow light arrested the formation of the dark
pigment and gave green pupae; while light colors in general gave light-
colored pupae. This color resemblance is commonly assumed to be of
protective value, and perhaps it is. Nevertheless, it is a direct effect of
light, and does not need to be explained by natural selection, even though
it cannot be denied that natural selection may have helped in its produc-
tion.

Poulton extended his studies to the adaptive coloration of caterpillars
and has published the results of an extensive series of experiments which
prove that the colors of certain caterpillars also are directly produced by
the same colors in the surrounding light. Gastro pacha quercifolia, which
always rests by day on the older wood of its food plant, was given black
twigs, reddish brown sticks, lichens, etc., to rest upon, and though all the
larvae were from the same cluster of eggs, and had been fed in the. same
way, each larva gradually assumed the color or colors of its resting place,
resulting in exquisite examples of protective resemblance, the most re-
markable of which were those in which the larvae assumed the variegated
coloration of lichens. Only the younger larvae, however, proved to be
susceptible to the colors of the environment; unlike those of Amphidasis
betularia, in which the older larvae also were sensitive to the surrounding
light. Here again, natural selection is unnecessary, even if not superfluous,
as an explanation of this kind of protective coloration.

Effects of Temperature. The amount of a pigment in the wing of a
butterfly depends in great measure upon the surrounding temperature
during the pupal stage, when the pigments are forming. Black or brown
spots have been enlarged artificially by subjecting chrysalides to cold;
hence it is probable that the characteristically large black spots on the
under side of the wings of the spring brood of our Cytiniris pscudargiolus

164 ENTOMOLOGY

are simply a direct effect of cold upon the wintering chrysalides. Simi-
larly the spring brood (variety marcia) of Phyciodes tharos owes its dis-
tinctive coloration to cold, as Edwards has proved experimentally.
Lepidoptera have been the subject of very many temperature experi-
ments, some of which will be mentioned presently in the consideration of
seasonal coloration.

Speaking generally, warmth (except in melanism] tends to induce a
brightening and cold a darkening of coloration, the darkening being due
to an increased amount of black or brown pigment. Temperature,
whether high or low, seldom if ever produces new pigments, but simply
alters the amount and distribution of pigments that are present already.

Effects of Moisture. Very little is known as to the effects of mois-
ture upon coloration. The dark colors of insular or coastal insects as
contrasted with inland forms, and the predominance of dull or suffused
species in mountainous regions of high humidity, have led observers oc-
casionally to ascribe melanism and suffusion to humidity. In these cases,
however, the possible influence of low temperature and other factors must
be taken into consideration. The experiments of Merrifield and of
Standfuss showed no effect of moisture upon lepidopterous pupae.

Pictet has found, however, that humidity acting on the cater-
pillars of Vanessa urticce and V. polychloros has a conspicuous effect on
the coloration of the butterflies. Thus when the caterpillars were fed
for ten days with moist leaves, the resulting butterflies had abnormal
black markings on the wings, and the same results followed when the
larvae were kept in an atmosphere saturated with moisture.

Climatal Coloration. The brilliant and varied colors of tropical
insects are popularly ascribed to intense heat, light and moisture; and
the dull monotonous colors of arctic insects, similarly, to the surrounding
climatal conditions. Climate undoubtedly exerts a strong influence upon
coloration, but the precise nature of this influence is obscure and will re-
main so until more is known about the effects separately produced by each
of the several factors that go to make up what is called climate.

The prevalence of intense and varied colors among tropical insects is
doubtless somewhat exaggerated, for the reason that the highly colored
species naturally attract the eye to the exclusion of the less conspicuous
forms. Indeed, Wallace assures us that, although tropical insects present
some of the most gorgeous colors in the whole realm of nature, there are
thousands of tropical species that are as dull colored as any of the tem-
perate regions. Carabidae, in fact, attain their greatest brilliancy in the
temperate zone, according to Wallace, though butterflies certainly show

COLOR AND COLORATION 165

a larger proportion ot vivid and varied colors in the tropics. Mayer
finds, in the widely distributed genus Papilio, that 200 South American
species display but 36 colors, while 22 North American species show 17.
While the number of species in South America is nine times as great as in
North America, the number of colors displayed is only a little more than
twice as great; hence Mayer concludes that the richer display of colors in
the tropics may be due to the far greater number of species, which gives
a better opportunity for color sports to arise; and not to any direct in-
fluence of the climate. Furthermore, the number of broods which occur
in a year is much greater in the tropics than in the temperate zones, so
that the tropical species must possess a correspondingly greater oppor-
tunity to vary.

Albinism and Melanism. These interesting phenomena, wide-
spread among the higher animals, are little understood, but have often
been attributed to temperature.

Albinism is exceptional whiteness or paleness of coloration, and is due
usually to lack or deficiency of pigment, but in some instances (Pieridae)
to the presence of a white pigment.

The common yellow butterfly, Colias philodice, and its relatives, are
frequently albinic. Scudder observed that albinism among butterflies in
America appears to be confined to a few Pieridse, and to be restricted to
the female sex; is more common in subarctic and subalpine regions than
in lower latitudes and altitudes, and only in the former places includes all
the females. At low altitudes, however, instead of appearing early in
the year as might be expected, the albinic forms appear during the warmer
months.

The experiments made by Gerould on C. philodice show that the
number of albinic female offspring from white females crossed with yellow
males is in accordance with Mendelian law. Albinism is not entirely
confined to the female as Scudder thought, for white males occur, though
they are extremely rare. ''They may be expected in regions where the
white female is especially abundant" (Gerould).

In Europe there are many albinic species of butterflies, and they are
by no means confined to the family Pieridae.

Melanism is unusual blackness or darkness of coloration. As to how
it is produced little is known, though warmth is probably the most po-
tent influence, and some attribute it to moisture, as was mentioned.
Pictet obtained partial melanism in Vanessa urtictz and V. polychloros by
subjecting the larva? to moisture.

In warm latitudes, some females of our Papilio glaucus are blackish

1 66 ENTOMOLOGY

brown with black markings, instead of being, as usual, yellow with black
markings. In the South, some males of the spring brood of Cyaniris
pseudargiolus are partly or wholly brown instead of blue. A melanic
male of Colias philodice occurs as an extremely rare mutation.

Seasonal Coloration. When butterflies have more than one brood
in a year, the broods usually differ in aspect, sometimes so much that their
specific identity is revealed only by rearing one brood from another.
The same species may exist under two or more distinct forms during the
same season in other words, may be seasonally dimorphic, trimorphic
or polymorphic.

Thus Polygonia interrogation's has two forms, fabricii and umbrosa,
which differ not only in coloration, but even in the form of the wings and
the genitalia. In New England fabricii hibernates and produces um-
brosa, as a rule, while umbrosa usually yields fabricii.

FIG. 232. Cyaniris pseudargiolus; A, form lucia; B, violacea; C, pseudargiolus proper.

Natural size.

The little blue butterfly, Cyaniris pseudargiolus (Fig. 232), is poly-
morphic to a remarkable degree. In the high latitudes of Canada a
single brood (lucia) occurs. About Boston the same spring brood ap-
pears, but under two forms: an earlier variety (lucid), which is small,
with large black markings beneath; and a later variety (violacea}, which
is typically larger, with smaller black spots, though it varies into the form
lucia. Finally, in summer, a third form (pseudargiolus proper) appears,
as the product of lucia or else the joint product of lucia and violacea, and
this is still larger, but the black spots are now faint. In the warm South
the spring form is violacea, but while some of the males are blue, others are
melanic, as just mentioned a dimorphic condition which does not occur
in the North. Violacea then produces pseudargiolus, in which, however,
all the males are blue.

Iphiclides ajax (Fig. 233) is another polymorphic butterfly whose
life history is complex. The three principal varieties of this species,
known respectively as marcellus, telamonides and ajax, differ not only in

COLOR AND COLORATION

167

coloration, but also in size and form; marcellns appears first, in spring;
tela mo nides appears a little later (though before marcellus has disappeared) ;
and dja.v is the summer form; as the season advances the varieties be-
come successively larger, with longer tails to the hind wings.

Now Edwards submitted chrysalides of the summer form aja.v to
cold and thereby obtained, in
the same summer, butterflies
with the form of ajax but the
markings of the spring form
telamonides. Some of the
chrysalides, however, lasted
over until the next spring and
then gave telamonides.

In Phyciodcs tliaros (Fig.
234) the spring and summer
broods, termed respectively
marcia and morpheus, were
at first regarded as distinct
species. In marcia the hind
wings are heavily and diffusely
marked beneath with strongly
contrasting colors, while in

morpheus they are plain and but faintly marked. Edwards placed upon
ice eighteen chrysalides that normally would have produced morpheus;
but instead of this, the fifteen imagines that emerged were all of the
spring form marcia and were smaller than usual. Pupa? derived from eggs

FIG. 233. Iphidides ajax, form telamonides, on
flower or button bush. Reduced.

\ * '

A

B

FIG. 234. Pliyciinli's lliaros; .1, spring form, marcia; B, summer form, morpheus; under

surfaces. Natural size.

of marcia gave, after artificial cooling, not morpheus, but marcia again.
The evident conclusion is that the distinctive coloration of the spring
variety is brought about by low temperature. In Labrador, only one

1 68 ENTOMOLOGY

brood occurs marcia; in New York, the species is digoneutic (two-
brooded) and in West Virginia polygoneutic (several-brooded).

Extensive temperature experiments upon seasonal dimorphism in
Lepidoptera have been conducted in Europe by some of the most com-
petent biologists. Weismann found that pupae of the summer form of
Pieris napi, if placed on ice, disclosed the darker winter form, usually
in the same season, though sometimes not until the next spring. It was
found impossible, however, to change the winter variety into the sum-
mer one by the application of heat. Similar results have attended the
important and much-discussed experiments of Dorfmeister, Weismann
and others upon Vanessa levana-prorsa and other species, from which it
has been inferred by Weismann that the winter form is the primary, older,
and more stable of the two forms, and the summer form a secondary,
newer, and less stable variety; since the latter form only, as a rule, re-
sponds much to thermal influences. Weismann argues that, in addition
to the direct effect of temperature, alternative inheritance also plays an
important part in the production of seasonal varieties. He tries to show,
moreover, that each seasonal variety is colored in adaptation to its
particular environment and that this adaptation may have been brought
about by natural selection though he does not succeed in this respect.

In several instances, local varieties have been artificially produced
as results of temperature control. Thus Standfuss produced in Germany,
by the application of cold, individuals of Vanessa urticoz which were
indistinguishable from the northern variety polaris; and from pupae of
Vanessa cardui, by warmth, a very pale form like that found in the
tropics; and, by cold, a dark variety similar to one found in Lapland.

These investigators and others, notably Merrifield and Fischer, have
accumulated a considerable mass of experimental evidence, the inter-
pretation of which is in many respects difficult, involving as it does, not
merely the direct effect of temperature upon the organism, but also deep
questions of heredity, including reversion, individual variation, and the
inheritance of acquired characters.

The seasonal increase in size that is noticeable, as in C. pseudargiolus
and /. ajax, is doubtless an expression of increasing metabolism due to
increasing temperature. Warmth, as is well known, stimulates growth,
and cold has a dwarfing effect. While this is true as a rule, there are
some apparent exceptions, however. Thus Standfuss found that some
caterpillars were so much stimulated by unusual warmth that they
pupated before they were sufficiently fed, and gave, therefore, under-

COLOR AND COLORATION

169

sized imagines. A moderate degree of warmth, however, undoubtedly
hastens growth.

Sexual Coloration. -The sexes are often distinguished by colora-
tional as well as structural differences. Colorational antigeny (this

FIG. 235. Picris protodicc; male (on the left) and female (on the right). Natural size.

word signifying secondary sexual differences of whatever sort) is most
prevalent among butterflies, in which it is the extreme phase of that
differentiation of ornamentation for which Lepidoptera are unrivaled.

The male of Pieris protodice (Fig. 235) has a few brown spots on the
front wings; the female is checkered with
brown on both wings. In Colias philodice
(Fig. 236) and C. eurytheme the marginal
black band of the front wings is sharp and
uninterrupted in the male, but diffuse and
interrupted by yellow spots in the female.
In the genus Papilio the sexes are often dis-
tinguished by colorational differences and in
Hesperiida? the males often have an oblique
black dash across the middle of each front
wing. Callosamia promethea (Fig. 237), the
gypsy moth and many other Lepidoptera
exhibit colorational antigeny. In not a few
Sesiidae the sexes differ greatly in coloration.
Thus in the male of the peach tree borer
(Sanninoidea exitiosa) all the wings are color-
less and transparent; while in the female the front wings are violet and
opaque and the fourth abdominal segment is orange above. The same
sex may present two types of coloration, as in males of Cyaniris pseudar-
giolus and females of Papilio glaucus, already mentioned. Papilio meropc,

FIG. 236. Colias pliilodice:
right fore wing of male (above)
and of female (below). Xat-
ural size.

170

ENTOMOLOGY

of South Africa, is remarkable in having three females, which are entirely
different in coloration from one another and from the male. There
is no longer any doubt, it may be added, as to the specific identity of
these forms.

Next to Lepidoptera, Odonata most frequently show colorational
antigeny. The male of Calopteryx maculata is velvety black; the female

smoky, with a white ptero-
stigmatal spot. Among
Coleoptera, the male of
Hoplia trifasciata is grayish
and the female reddish
brown; a few more ex-
amples might be given,
though sexual differences in
coloration are comparative-
ly rare among beetles. Of
Hymenoptera, some of the
Tenthredinidae exhibit col-
orational antigeny.

Among tropical butter-
flies there are not a few in-
stances in which the special
coloration of the female is
a d a p t i v e harmonizing
with the surroundings or
else imitating with remark-
able precision the colora-
tion of another species
which is known to be im-
mune from the attacks of
birds as described beyond.
In this way, as Wallace sug-
gests, the egg-laden females

may escape destruction, as they sluggishly seek the proper plants upon
which to lay their eggs. Here would be a fair field for the operation of
natural selection.

In most insects, however, sexual differences in coloration are ap-
parently of no protective value and are usually so trivial and variable
as probably to be of no use for recognition purposes. The usual state-
ment that these differences facilitate sexual recognition is a pure as-

FIG. 237. Callosamia promethca; A , male, clinging to
cocoon; B, female. Reduced.

COLOR AND COLORATION iyi

sumption, in the case of insects, and one that is inadequate in spite of
its plausibility, for (i) it is extremely improbable from our present knowl-
edge of insect vision that insects are able to perceive colors except in
the broadest way, namely, as masses; (2) the great majority of insect
species show no sexual differences in coloration; (3) when colorational
antigeny is present it is probably unnecessary, to say the least, for sexual
recognition. Thus, notwithstanding the marked dissimilarity of colora-
tion in the two sexes of C. promctJiea, the males, guided by an odor, seek
out their mates even when the wings of the female have been amputated
and male wings glued in their place, as Mayer found.

Hence, when useless, colorational antigeny cannot have been de-
veloped by natural selection and may be due simply to the extended
action of the same forces that have produced variety of coloration in
general.

Origin of Color Patterns. Tower, who has written an important
work on the colors and color patterns of Coleoptera, finds that each of
the black spots on the pronotum of the Colorado potato beetle (Fig. 238)
"is developed in connection with a muscle, and marks the point of at-
tachment of its fibers to the cuticula." Thus the color pattern, in its
origin, is not necessarily useful. This point is so important that we
quote Tower's conclusions in full. "The most important and widely
disseminated of insect colors are those of the cuticula . . . these
colors develop as the cuticula hardens, and appear first, as a rule, upon
sclerites to which muscles are attached. In one of the earlier sections of
this paper I showed that the pigment develops from before backward
and, approximately, by segments, excepting that it may appear upon
the head and most posterior segments simultaneously.

"In ontogeny color appears first, as a rule, over the muscles which
become active first, or upon certain sclerites of the body. These are
usually the head muscles, although exceptions are not infrequent. It
should be remembered that as the color appears the cuticula hardens,
and, considering that muscles must have fixed ends for their action, it
seems that there is a definite relation between the development of color,
the hardening of the cuticula, and the beginning of muscular activity;
the last being dependent upon the second, and, incidentally, accompanied
by the first. As muscular activity spreads over the animal the cuticula
hardens and color appears, so that color is nearly, if not wholly, seg-
mentally developed.

"The relation which exists between cuticular color and the stiffening
of the cuticula is thus a physiological one, the cuticula not being able to

172 ENTOMOLOGY

harden without becoming yellow or brown. What bearing has this
upon the origin of color patterns? In the lower forms of tracheates,
such as the Myriapods, colors appear as segmental repetitions of spots
or pigmented areas which mark either important sclerites or muscle
attachments. On the abdomens of insects, where segmentation is best
observed, color appears as well-defined, segmentally arranged spots,
but on the thorax segmentation is obscured and lost upon the head. Of
what importance, then, is pigmentation? And how did it arise? If the
ontogenetic stages offer any basis for phylogenetic generalization, we
may conclude that cuticula color originated in connection with the
hardening of the integument of the ancestral tracheates as necessary to
the muscular activity of terrestrial life. The primitive colors were yellows,
browns and blacks, corresponding well with the surroundings in which
the first terrestrial insects are supposed to have lived. The color pattern
was a segmental one, showing repetition of the same spots upon suc-
cessive segments, as upon the abdomen of Coleoptera.

"So firmly have these characters become ingrained in the tracheate
series, and so important is this relation of the hardening of the cuticula
to the musculature and to the formation of body sclerites, that even the
most specialized forms show this primitive system of coloration; and,
although there may be spots and markings which have no connection
with it, still the chief color areas are thus closely associated."

Development of Color Patterns. Although the causes of colora-
tion are, for the most part, obscure, it is possible, nevertheless, to point
out certain paths along which coloration appears to have developed.
These paths have been determined by the comparison of color patterns
in kindred groups of insects and the study of colorational variations in
adults of the same species. The development of coloration in the in-
dividual, however, has as yet received but little attention excepting
the excellent studies of Mayer and of Tower. Butterflies, moths and
beetles have naturally been preferred by most students of the subject.

The most primitive colors among moths are uniform dull yellows,
browns and drabs the same colors that the pupal blood assumes when
it is dried in the air. These simple colors prevail on the hind wings of
most moths and on the less exposed parts of the wings of highly colored
butterflies. The hind wings of moths are, as a rule, more primitively
colored than the front ones because, as Scudder says, "all differentiation
in coloring has been greatly retarded by their almost universal conceal-
ment by day beneath the overlapping front wings." Exceptions to
this statement are found in Geometridae and such other moths as rest

COLOR AND COLORATION 173

with all the wings spread. "In such hind wings we find that the simplest
departure from uniformity consists in a deepening of the tint next the
outer margin of the wing; next we have an intensification of the deeper
tint along a line parallel to the margin; it is but a step from this condi-
tion to a distinct line or band of dark color parallel to the margin. Or the
marginal shade may, in a similar way, break up into two or more trans-
verse and parallel submarginal lines, a very common style of ornamenta-
tion, especially in moths. Or, again, starting with the submarginal
shade, this may send shoots or tongues of dark color a short distance
toward the base, giving a serrate inner border to the marginal shade;
when now this breaks up into one, two, or more lines or narrow stripes,
these stripes become zigzag, or the inner ones may be zigzag, while the
outer ones are plain a very common phenomenon.

"A basis such as this is sufficient to account for all the modifications
of simple transverse markings which adorn the wings of Lepidoptera."

Briefly, one or more bands may break up into spots or bars, the
breaks occurring either between the veins or, more commonly, at the
veins; and in the latter event, short bars or more or less quadrate or
rounded spots arise in the interspaces. From simple round spots there
may develop, as Darwin and others have shown, many-colored eye-like
spots, or ocelli.

Mayer gives the following laws of color pattern: "(a) Any spot
found upon the wing of a butterfly or moth tends to be bilaterally sym-
metrical, both as regards form and color; and. the axis of symmetry is a
line passing through the center of the interspace in which the spot is
found, parallel to the longitudinal nervures. (b) Spots tend to appear
not in one interspace only, but in homologous places in a row of adjacent
interspaces, (c) Bands of color are often made by the fusion of a row
of adjacent spots, and, conversely, chains of spots are often formed by
the breaking up of bands, (d) When in process of disappearance, bands
of color usually shrink away at one end. (e) The ends of a series of spots
are more variable than the middle. (/) The position of spots situated
near the outer edges of the wing is largely controlled by the wing folds

or creases. '

These results have been arrived at chiefly by the study of the varia-
tions presented by color patterns.

Variation in Coloration. It is safe to say that no two insects are
colored exactly alike. Some species, however, are far more variable
than others. Catocala ilia, for example, occurs under more than fifty
varieties, each of which might be given a distinctive name, were it not

1 74 ENTOMOLOGY

for the fact that these varieties run into one another. One may examine
hundreds of potato beetles (L. decemlineata) without finding any two
that have precisely the same pattern on the pronotum. The range of
this variation in this species is partially indicated in Fig. 238, and that
of Cicindela in Fig. 239.

Individuals of Cicindela vary in pattern in a few definite directions,
and the patterns that characterize the various species appear to be
fixations of individual variations. In the words of Dr. Horn: "(i)
The type of marking is the same in all our species. (2) Assuming a well-
marked species (vulgaris, Fig. 239, i) as a central type, the markings of
other species vary from that type, (a) by a progressive spreading of the
white, (6) by a gradual thinning or absorption of the white, (c) by a
fragmentation of the markings, (d) by linear supplementary extension.
(3) Many species are practically invariable (i. e., the individual varia-
tions are small in amount as compared with those in other species).
These fall into two series: (a) those of the normal type, as vulgaris,
hirticollis and tenuisignata; (b) those in which some modification of the
type has become permanent, probably through isolation, as margini-
pennis, togata and lemniscata. (4) Those species which vary do so in
one direction only." New types of pattern, of specific value, appear
to have arisen by the isolation and perpetuation of individual variations.

Variations in general fall into two classes: continuous (individual
variations] and discontinuous mutations. The former are always present,
are slight in extent and intergrade with one another; they are distributed
symmetrically about a mean condition. The latter are occasional, of
considerable extent and sharply separated from the normal condition.

R. H. Johnson has published an important statistical study on evo-
lution in the color pattern of the lady-beetles. He finds both con-
tinuous and discontinuous variations present; that the color pattern is
capable of modification by the environment; that some modifications
are hereditary characters and others not.

Replacements. Examples of the replacement of one color by
another are familiar to all collectors. The red of Vanessa atalanta and
Coccinellidae may be replaced by yellow. These two colors in many
butterflies and beetles are due to pigments that are closely related to
each other chemically. Thus in the chrysomelid Melasoma lapponica
the beetle at emergence is pale but soon becomes yellow with black
markings, and after several hours, under the influence of sunlight, the
yellow changes to red; the change may be prevented, however, by keep-
ing the beetle in the dark. After death, the red fades back through

COLOR AND COLORATION

175

FIG. 238. Colorational variations of the pronotum of the Colorado potato beetle, Leplinotarsa

dccemlineala.

i 7 6

ENTOMOLOGY

FIG. 239. Elytral color patterns of species of Cicindela. i-S illustrate reduction of dark
area; 9-14, extension of dark area; 15, 16, formation of longitudinal vitta; 17, iS, linear ex-
tension of markings, i, C. vulgar is; 2, generosa; j, generosa; 4, pamphila; j, limbata; 6,
togata; 7, gratiosa; 8, canosa; p, tcnuisignata; 10, marginipennis; u, licntzn; 12, scxgitttata;
13, hamorrhagica; 14, splendida; 15, impcrfecta; 16, Icmnixcata; 77, gabbii; 18, saulcyi.
After HORN, from Entomological News.

COLOR AND COLORATION 177

orange to yellow, especially as the result of exposure to sunlight. Yellow
in place of red, then, may be attributed to an arrested development of
pigment in the living insect and to a process of reduction in the dead
insect, metabolism having ceased.

Yellow and green are similarly related. The stripes of Pascilocapsus
lineatus are yellow before they become green, and after death fade back
to yellow. As the green pigment in most, if not all, phytophagous in-
sects is chlorophyll, these color changes are probably similar to those
that occur in leaves. Leaves grown in darkness are yellow, from the
presence of etiolin, and do not turn green until they are exposed to sun-
light (or electric light), without which chlorophyll does not develop;
and as metabolism ceases, chlorophyll disintegrates, as in autumn,
leaving its yellow constituent, xanthophyll, which is very likely the
same substance as etiolin.

Cicindela sexguttata and Calosoma scrutator are often blue in place of
green. Here, however, these colors are structural, and their variations
are to be attributed to slight differences in the spacing of the surface
elevations or depressions.

Green grasshoppers occasionally become pink toward the close of
summer. No explanation has been offered for this phenomenon, though
it may be remarked that when grasshoppers are killed in hot water the
normal green pigment turns to pink.

These changes of color are apparently of no use to the insect, being
merely incidental effects of light, temperature or other inorganic in-
fluences.

j j CHAPTER VI

ADAPTIVE COLORATION

Protective Resemblance. Every naturalist knows of many ani-
mals that tend to escape detection by resembling their surroundings.
This phenomenon of protective resemblance is richly exemplified by in-
sects, among which one of the most remarkable cases is furnished by the
Kallima butterflies, especially K. inachis of India and K. paralekta of the
Malay Archipelago. The former species (Fig. 240) is conspicuous when on
the wing; its bright colors, however, are confined to the upper surfaces
of the wings, and when these are folded together, as in repose, the insect

B

FIG. 240. Kallima inachis; A, upper surface; B, with wings closed, showing resemblance to

a leaf. X 1 A.

resembles to perfection one of the dead leaves among which it is accus-
tomed to hide. The form, size and color of the leaf are accurately re-
produced, the petiole being simulated by the tails of the wings. Two
parallel shades, one light and one dark, represent, respectively, the
illuminated and the shaded side of a mid-rib, and the side-veins as well
are imitated; there are even small scattered black spots resembling
those made on the leaf by a species of fungus. Furthermore, the butterfly
habitually rests, not among green leaves, where it would be conspicu-
ous, but among leaves with which it harmonizes in coloration. Not-

ADAPTIVE COLORATION

179

withstanding some discussion as to whether it usually rests in pre-
cisely the same position as a leaf, this insect certainly deceives experi-
enced entomologists and presumably eludes birds and other enemies by
means of its deceptive coloration.

Some of the tropicaJ Phasmidre counterfeit sticks, green leaves, or
dead leaves with minute accuracy. Our common phasmids, Diaphero-

FIG. 241. Mcmomera blalchlcyi, on a twig.
Natural size.

FIG. 242. Catocala lacrymosa; A, upper surface;
R, with wings closed, and resting on bark. Re-
duced.

mera femorata and Manomera blatchleyi (Fig. 241), are well known as
''stick insects"; indeed, it is not necessary to go beyond the temperate
zone to find plenty of examples of protective resemblance. Geometrid
caterpillars imitate twigs, holding the body stiffly from a branch and
frequently reproducing the form and coloration of a twig with striking
exactitude; and the moths of the same family are often colored like the
bark against which they spread their wings. Even more perfectly do the
Catocala moths resemble the bark upon which they rest (Fig. 242), with

i8o

ENTOMOLOGY

their conspicuous and usually showy hind wings concealed under the pro-
tectively colored front wings. The caterpillars of Basilarchia archippus
and Papilio thoas, as well as other larvae and not a few moths, resemble
closely the excrements of birds. Numerous grass-eating caterpillars are
striped with green, as is also a sphingid species (Ellema harrisii] that lives
among pine needles. The large green sphinx caterpillars perhaps owe
their inconspicuousness partly to their oblique lateral stripes, which cut a
mass of green into smaller areas. The caterpillar of Schizura ipomcea
(Fig. 243), which is green with brown patches, rests for hours along the
eaten or torn edge of a basswood leaf, in w r hich position it bears an ex-

FIG. 243. Caterpillar of Schizura ipomoea clinging to a torn leaf. Natural size.

tremely deceptive resemblance to the partially dead border of a leaf.
The weevils that drop to the ground and remain immovable are often
indistinguishable to the collector on account of their likeness to bits of
soil or little pebbles. Everyone has noticed the extent to which some
of the grasshoppers resemble the soil in color; Trimerotropis maritinia
is practically invisible against the gray sand of the seashore or other
places to which it restricts itself; and Dissosteira Carolina, which varies
greatly in color, ranging from ashy gray to yellowish or to reddish brown,
is commonly found on soil of its own color.

Adventitious Resemblance. If, instead of hastily ascribing all
cases apparently of protective resemblance to the action of natural

ADAPTIVE COLORATION l8l

selection, one inquires into the structural basis of the resemblance in
each instance, it is found that some cases can be explained, without the
aid of natural selection, as being direct effects of food, light or other
primary factors. Such cases, then, are in a sense accidental. For ex-
ample, many inconspicuous green insects are green merely because
chlorophyll from the food-plant tinges the blood and shows through the
skin. If it be argued that natural selection has brought about a thin
and transparent skin, it may be replied that the skin of a green cater-
pillar is by no means exceptional in thinness or transparency. More-
over, many leaf-mining caterpillars are green, simply because their food
is green; for, living as they do within the tissues of leaves, and surrounded
by chlorophyll, their own green color is of no advantage, but is merely
incidental.

Again, in the "protectively" colored chrysalides experimented upon
by Poulton, the color was directly influenced by the prevailing color
of the light that surrounded the larva during the last few days before
pupation. Of course, it is conceivable that natural selection may have
preserved such individuals as were most responsive to the stimulus of
the surrounding light; nevertheless the fact remains that these resem-
blances do not demand such an explanation, which is, in other words,
superfluous.

Indeed, a great many of the assumed examples of "protective re-
semblance" are very far-fetched. On the other hand, when the re-
semblance is as specific and minutely detailed as it is in the Kallima
butterflies where, moreover, special instincts are involved the phe-
nomenon can scarcely be due to chance; the direct and uncombined
action of such factors as food or light is no longer sufficient to explain
the facts although these and other factors are undoubtedly important
in a primary, or fundamental; way. Here natural selection becomes
useful, as enabling us to understand how original variations of structure
and instinct in favorable directions may have been preserved and ac-
cumulated until an extraordinary degree of adaptation has been attained.

Value of Protective Resemblance. The popular opinion as to
the efficiency of protective resemblances in undoubtedly an exaggerated
one, owing mainly to the false assumption that the senses of the lower ani-
mals are co-extensive in range with our own. As a matter of fact, birds
detect insects with a facility far superior to that of man, and destroy them
by the wholesale, in spite of protective coloration. Thus, as Judd has
ascertained, no less than three hundred species of birds feed upon pro-
tectively colored grasshoppers, which they destioy in immense numbers.

I 82 ENTOMOLOGY

and more than twenty species prey upon the twig-like geometric! larvae;
while the weevils that look like particles of soil, and the green-striped
caterpillars that assimilate with the surrounding foliage are constantly
to be found in the stomachs of birds.

After all,. however, protective resemblance may be regarded as ad-
vantageous upon the whole, even if it is ineffectual in thousands of in-
stances. An adaptation may be successful even if it does fall short of
perfection; and it should be borne in mind that the evolution of protect-
ive resemblances among insects has probably been accompanied on the
part of birds by an increasing ability to discriminate these insects from
their surroundings.

Warning Coloration. In strong contrast to the protectively
colored species, there are many insects which are so vividly colored as
to be extremely conspicuous amid their natural surroundings. Such
are many Hemiptera (Lygceus, Murgantia), Coleoptera (Necrophoms,
Lampyrida?, Coccinellidae, Chrysomelidae), Hymenoptera (Mutillidae,
Vespidae), and numerous caterpillars and butterflies. Conspicuous col-
ors, being frequently though not always associated with qualities
that render their possessors unpalatable or offensive to birds or other
enemies, are advantageous if, by insuring ready recognition, they ex-
empt their owners from attack.

Efficiency of Warning Colors. Owing to much disagreement as
to the actual value of "warning" colors, several investigators have made
many observations and experiments upon the subject. Tests made by
offering various conspicuous insects to birds, lizards, frogs, monkeys and
other insectivorous animals have given diverse results, according to
circumstances. Thus, one gaudy caterpillar is refused by a certain bird,
at once, or else after being tasted, but another and equally showy cater-
pillar is eaten without hesitation. Or, an insect at first rejected may at
length be accepted under stress of hunger; or a warningly colored form
disregarded by some animals is accepted by others. Moreover, some
of the experiments with captive insectivorous animals are open to ob-
jection on the score of artificiality.

Nevertheless, from the data now accumulated, there emerge some
conclusions of definite value. Frank Finn, whose conclusions are quoted
beyond, has found in India that the conspicuous colors of some butter-
flies (Danainse, Acr&a violce, Delias eucharis, Papilio aristolochice) are
probably effective as "warning" colors. Marshall found in South
Africa that mantids, which would devour most kinds of butterflies, in-
cluding warningly colored species, refused Acrcea, which appeared to be

ADAPTIVE COLORATION 183

not only distasteful but even unwholesome; Acraa is eaten, however,
by the predaceous Asilidae, which feed indiscriminately upon insects
for example, beetles, dragon flies and even stinging Hymenoptera. The
masterly studies of Marshall and Poulton strongly support the general
theory of warning coloration.

In this country, much important evidence upon the subject has been
obtained by Dr. Judd from an extensive examination of the stomach-
contents of birds, supplemented by experiments and field observations.
Judd says that Murgantia histrionica and other large showy bugs are
usually avoided by birds; that the showy, ill-flavored Coccinellidae, and
Chrysomelidae such as the elm leaf beetle, Diabrotica, and Leptinotarsa
(Doryphora), possess comparative immunity from birds; and that
Macrodactylus, Chauliognathus and Cyllene are highly exempt from
attack. Such cases, he adds, are comparatively few among insects,
however, and in general, warning colors are effective against some ene-
mies but ineffective against others.

Generally speaking, hairs, stings and other protective devices are
accompanied by conspicuous colors though there are many exceptions
to this rule. These warning colors, however, fail to accomplish their
supposed purpose in the following instances, given by Judd. Taking in-
sects that are thought to be protected by an offensive odor or a dis-
agreeable taste: Heteroptera in general are eaten by all insectivorous
birds, the squash bug by hawks and the pentatomids by many birds;
among Carabidae with their irritating fluids, Harpalus caliginosus and
pennsylvanicus are food for the crow, catbird, robin and six others;
Carabus and Calosoma are relished by crows and blackbirds; Silphidae
are taken by the crow, loggerhead shrike and kingbird ; and Leptinotarsa
decemlineata is eaten by at least six kinds of birds: wood thrush, rose-
breasted grosbeak, quail, crow, cuckoo and catbird. Of hairy and spiny
caterpillars, Arctiidas are eaten by the robin, bluebird, catbird, cuckoo
and others; the larvae of the gypsy moth are food for the blue-jay, robin,
chickadee, Baltimore oriole and many others [thirty-one birds, in Massa-
chusetts]; and the spiny caterpillars of Vanessa antiopa are taken by
cuckoos and orioles. Of stinging Hymenoptera, bumble bees are eaten
by the bluebird, blue-jay and two flycatchers; the honey bee, by the
wood pewee, phoebe, olive-sided flycatcher and kingbird; Andrena by
many birds, and Vespa and Polistes by the red-bellied woodpecker, king-
bird, and yellow-bellied flycatcher.

These facts by no means invalidate the general theory, but they do
show that "disagreeable" qualities and their associated color signals

1 84

ENTOMOLOGY

are of little or no avail against some enemies. The weight of evidence
favors the theory of warning coloration in a qualified form. While con-
spicuous colors do not always exempt their owners from destruction,
they frequently do so, by advertising disagreeable attributes of one sort
or another.

The evolution of warning coloration is explained by natural selec-
tion; in fact, we have no other theory to account for it. The colors

A

B

FIG. 244..! , Anoxia plcxippns, the "model"; B, Basilarcliia arclrippns, the "mimic.'

Natural size.

themselves, however, must have been present before natural selection
could begin to operate; their origin is a question quite distinct from that
of their subsequent preservation.

Protective Mimicry. This interesting and highly involved phe-
nomenon is a special form of protective resemblance in which one species
imitates the appearance of another and better protected species, there-

ADAPTIVE COLORATION 185

by sharing its immunity from destruction. Though it attains its highest
development in the tropics, mimicry is well illustrated in temperate
regions. A familiar example is furnished by Basilarchia archippus
(Fig. 244, B), which departs widely from the prevailing dark coloration
of its genus to imitate the milkweed butterfly, Anosia plcxippus. The
latter species, or "model," appears to be unmolested by birds, and the
former species, or ''mimic," is thought to secure the same exemption
from attack by being mistaken for its unpalatable model. The common
drone-fly, Eristalis tenax (Fig. 245, B) mimics a honey bee in form, size,
coloration and the manner in which it buzzes about flowers, in company
with its model; it does not deceive the kingbird and the flicker, however.
Some Asilidae are remarkably like bumble bees in superficial appearance
and certain Syrphus flies mimic wasps with more or less success. The

A

FIG. 245. Protective mimicry. A, drone bee. Apis mdlifcra; B, drone fly, Erislalis tcnax.

Natural size.

beetle Casnonia bears a remarkable resemblance to the ants with
which it lives.

The classic cases are those of the Amazonian Heliconiidae and Pieridas,
in which mimicry was first detected by Bates. The Heliconiidae are
abundant, vividly colored and eminently free from the attacks of birds
and other enemies of butterflies, on account of their disagreeable odor
and taste. Some of the Pieridae a family fundamentally different from
Heliconiidae imitate the protected Heliconiidae so successfully, in
coloration, form and flight, that while other Pieridas are preyed upon by
many foes, the mimicking species tend to escape attack.

The family Heliconiidae, referred to by Bates, comprised what are
now known as the subfamilies Heliconiinae, Ithomiinae and Danainae;
similarly, Pieridae and Papilionidae are now often termed respectively
Pierinas and Papilioninae. Ithomiinae are mimicked also by Papilio-
ninae and by moths of the families Castniidae and Pericopidae.

1 86 ENTOMOLOGY

The discoveries of Bates in tropical South America were paralleled
and supported by those of Wallace in India and the Malay Archipelago
(where Danainae are the chief "models"), and of Trimen in South Africa
(where Acrseinae and Danainae serve as models). Trimen discovered a
most remarkable case, in which three species of Dana is are mimicked,
each by a distinct variety of the female of Papilio cenea (merope) .

So much for that kind of mimicry but how is the following kind to
be explained? The Ithomiinas of the Amazon valley have the same form
and coloration as the Heliconiinse, but the Ithomiinae themselves are
already highly protected. The answer is that this resemblance is of
advantage to both groups, as it minimizes their destruction by birds
these having to learn but one set of warning signals instead of two. This
is the essence of Miiller's famous explanation, which will presently be
stated with more precision. There are two kinds of mimicry, then: (i) the
kind described by Bates, in which an edible species obtains security by
counterfeiting the appearance of an inedible species; (2) that observed
by Bates and interpreted by Miiller, in which both species are inedible.
These two kinds are known respectively as Batesian and Miillerian
mimicry, though some writers prefer to limit the term mimicry to the
Batesian type.

Wallace's Rules. The chief conditions under which mimicry
occurs have been stated by Wallace as follows:

" i. That the imitative species occur in the same area and occupy the
very same station as the imitated.

"2. That the imitators are always the more defenceless.

"3. That the imitators are always less numerous in individuals.

"4. That the imitators differ from the bulk of their allies.

"5. That the imitation, however minute, is external and visible
only, never extending to internal characters or to such as do not affect
the external appearance."

These rules relate chiefly to the Batesian form of mimicry and need
to be altered to apply to the Miillerian kind.

The first criterion given by Wallace is evidently an essential one and
it is sustained by the facts. It is also true that mimic and model occur
usually at the same time of year; Marshall found many new instances
of this in South Africa. In some cases of mimicry, strange to say, the
precise model is unknown. Thus some Nymphalidae diverge from their
relatives to mimic the Euplceinae, though no particular model has been
found. In such instances, as Scudder suggests, the prototype may
exist without having been found; may have become extinct; or the

ADAPTIVE COLORATION 187

species may have arrived at a general resemblance to another group
without having as yet acquired a likeness to any particular species of
the group, the general likeness meanwhile being profitable.

The second condition named by Wallace is correct for Batesian but
not for Miillerian mimicry . .

The fulfilment of the third condition is requisite for the success of
Batesian mimicry. Bates noted that none of the pierid mimics were so
abundant as their heliconiid models. If they were, their protection
would be less; and should the mimic exceed its model in numbers, the
former would be more subject to attack than the latter. Sometimes,
indeed, as Miiller found, the mimic actually is more common than the
model; in which event, the consequent extra destruction of the mimic
would at least theoretically reduce its numbers back to the point of
protection.

In Miillerian mimicry, however, the inevitable variation in abundance
of two or more converging and protected species is far less disastrous;
though when two species, equally distasteful, are involved, the rarer of
the two has the advantage, as Fritz Muller has shown. His lucid ex-
planation is essentially as follows:

Suppose that the birds of a region have to destroy 1,200 butterflies
of a distasteful species before it becomes recognized as such, and that
there exist in this region 2,000 individuals of species A and 10,000 of
species B; then, if they are different in appearance, each will lose 1,200
individuals, but if they are deceptively alike, this loss will be divided
among them in proportion to their numbers, and A will lose 200 and B
1,000. A accordingly saves 1,000, or 50 per cent, of the total number
of individuals of the species, and B saves only 200, or 2 per cent. Thus,
while the relative numbers of the two species are as i to 5, the relative
advantage from their resemblance is as 25 to i.

If two or more distasteful species are equally numerous, their re-
semblance to one another brings nearly equal advantages. In cases of
this kind and many are known it is sometimes impossible to dis-
tinguish between model and mimic, as all the participants seem to have
converged toward a common protective appearance, through an inter-
change of features the "reciprocal mimicry" of Dr. Dixey.

Marshall argues, however, against this diaposematism, maintaining
that in the case of two participants in Miillerian mimicry the evolution
of the mimetic pattern has been in one direction only toward the more
abundant species any variations in the opposite direction being dis-
advantageous.

1 88 ENTOMOLOGY

From this explanation, the superior value of Miillerian as compared
with Batesian mimicry is evident.

The fourth condition that the imitators differ from the bulk of
their allies holds true to such a degree that even the two sexes of the
same species may differ extremely in coloration, owing to the fact that
the female has assumed the likeness of some other and protected species.
The female of Papilio cenea, indeed, occurs (as was just mentioned)
under three varieties, which mimic respectively three entirely dissimilar
species of Danais, and none of the females are anything like their male
in coloration.

The generally accepted explanation for these remarkable but numer-
ous cases in which the female alone is mimetic, is that the female, bur-
dened with eggs and consequently sluggish in flight and much exposed
to attack, is benefited by imitating a species which is immune; while

the male has had no such incentive so to speak-
to become mimetic. Of course, there has been no
conscious evolution of mimicry.

Wallace's fifth stipulation is important, but should
read this way: "The imitation, however minute, is but
FIG 246 -- \ external and visible usually, and never extends to in-
locustid, Myrmc- ternal characters which do not affect the external ap-

cophana fallax, which . ,11

resembles an ant. pearance. For, as Poulton points out, the alertness

Twice natural length. o f a ^gg^e w hich mimics a wasp, implies appropriate

From BRUNNER VON

WATTENWYL. changes in the nervous and muscular systems. In

its intent, however, Wallace's rule holds good, and
by disregarding it some writers strain the theory of mimicry beyond
reasonable limits. Some have said, for example, that the resem-
blance between caddis flies and moths is mimicry; when the fact is
that this resemblance is not merely superficial but is deep-seated; the
entire organization of Trichoptera shows that they are closely related
to Lepidoptera. This likeness expresses, then, not mimicry, but affinity
and parallel development. The same objection applies to the assumed
cases of mimicry within the limits of a single family, as between two
genera of Heliconiidas or between the chrysomelid genera Lema and
Diabrotica. The more nearly two species are related to each other, the
more probable it becomes that their similarity is due not to mimicry but
to their common ancestry.

On the other hand, the resemblance frequently occurs between species
of such different orders that it cannot be attributed to affinity. Illus-
trations of this are the mimicry of the honey bee by the drone fly, and

ADAPTIVE COLORATION 189

the many other instances in which stinging Hymenoptera are counter-
feited by harmless flies or beetles. A locustid of the Soudan resembles
an ant (Fig. 246), and the resemblance, by the way, is obtained in a
most remarkable manner. Upon the stout body of this orthopteron
the abdomen of an ant is delineated in black, the rest of the body being
light in color and inconspicuous by contrast with the black. Indeed
the various means by which a superficial resemblance is brought about
between remotely related insects are often extraordinary.

Irrespective of affinity, insects of diverse orders may converge in
wholesale numbers toward a central protected form. The most com-
plete examples of this have been brought to light by Marshall and
Poulton, in their splendid work on the bionomics of South African
insects, in which is given, for instance, a colored plate showing how
closely six distasteful and dominant beetles of the genus Lycus are imi-
tated by almost forty species of other genera a remarkable example of
convergence involving no less than eighteen families and five orders,
namely, Coleoptera, Hymenoptera, Hemiptera, Lepidoptera and Diptera.
Excepting a few unprotected, or Batesian, mimics (a fly and two or
three beetles), this association is one between species that are already
protected, by stings, bad tastes or other peculiarities. In other words,
here is Miillerian mimicry on an immense scale; and if Miillerian mim-
icry is profitable when only two species are concerned, what an enormous
benefit it must be to each of forty participants!

Strength of the Theory. Evidently the theory of mimicry rests
upon the assumption that the mimics, by virtue of their mimicry, are
specially protected from insectivorous foes. Until the last few years,
however, there was altogether too little positive evidence bearing upon
the assumption itself, though this was supported by such scattered ob-
servations as were available. The oft-repeated assertion that this lack
of evidence was due simply to inattention to the subject, has been proved
to be true by the decisive results recently gained in the tropics by several
competent investigators who have been able to give the subject the
requisite amount of attention.

From his observations and experiments in India, Frank Finn con-
cludes:

"i. That there is a general appetite for butterflies among insec-
tivorous birds, even though they are rarely seen when wild to attack
them.

"2. That many, probably most, species dislike, if not intensely, at
any rate in comparison with other butterflies, the warningly-colored

I QO ENTOMOLOGY

Danainae, Acr&a violce, Delias cucharis, and Papilio aristolochicE; of
these the last being the most distasteful, and the Danainae the least so.

"3. That the mimics of these are at any rate relatively palatable,
and that the mimicry is commonly effectual under natural conditions.

"4. That each bird has separately to acquire its experience, and well
remembers what it has learned.

"That therefore on the whole, the theory of Wallace and Bates is
supported by the facts detailed in this and my former papers, so far as
they deal with birds (and with the one mammal used). Professor
Poulton's suggestion that animals may be forced by hunger to eat un-
palatable forms is also more than confirmed, as the unpalatable forms
were commonly eaten without the stimulus of actual hunger generally,
also, I may add, without signs of dislike."

Though insects have many vertebrate and arthropod enemies, it is
probable that the evolution of mimetic resemblance, implying warning
coloration, has been brought about chiefly by insectivorous birds.

Neglecting papers of minor importance, we may pass at once to the
most important contribution upon this subject the voluminous work
of Marshall and Poulton upon mimicry and warning colors in South
African insects. These investigators have found that birds are to be
counted as the principal enemies of butterflies; that the Danainae and
Acraeinae, which are noted as models, are particularly immune from de-
struction, while unprotected forms suffer; and that mimicking, though
palatable, species share the freedom of their models. The same is true
of beetles, of which Coccinellidse, Malacodermidae (notably Lycus),
Cantharidae and many Chrysomelidae serve as models for many other
Coleoptera, being "conspicuous and constantly refused by insect-
eaters. " In short, the splendid work of Marshall and Poulton tends to
place the theory of Batesian and Miillerian mimicry upon a substantial
foundation of observational and experimental evidence.

In regard to the important question do birds avoid unpalatable
insects instinctively or only as the result of experience the evidence is
all one way. Several investigators, including Lloyd Morgan, have
found that newly-hatched birds have no instinctive aversions as regards
food, but test everything, and (except for some little parental guidance)
are obliged to learn for themselves what is good to eat and what is not.
This experimental evidence that the discrimination of food by birds
is due solely to experience, was evidently highly necessary to place the
theory of mimicry especially the Mtillerian theory upon a sound
basis.

ADAPTIVE COLORATION 191

Though butterflies as a group are much subject to the attacks of
birds in the tropics, it has been asserted that butterflies in temperate
regions are as a whole almost exempt from the attacks of birds, and that
consequently the mimicry of the monarch (Fig. 244) by the viceroy is
of no advantage. In answer to this assertion Marshall has published a
long list of references showing that butterflies are attacked by birds
more commonly than has been generally supposed. At the same time
there is no proof that the viceroy profits at present by its mimetic pattern,
though it may have done so in the past. In any event, the departure of
archippus from its congeners toward one of the Danainae a famous
group of " models" in the tropics is unintelligible except as an instance
of mimicry.

Granting that mimicry is upon the whole advantageous, it becomes
important to learn just how far the advantage extends; and we find that
mimicry is not of universal effectiveness. Even the highly protected
Heliconiinse and Danainae are food for some predaceous insects. In
this country, as Judd has observed, the drone-fly (Eristalis tenax), which
mimics the honey bee, is eaten by the kingbird and the phcebe; the
kingbird, indeed, eats the honey bee itself, but is said to pick out the
drones; chickens also discriminate between drones and workers, eating
the former and avoiding the latter. Bumble bees and wasps, imitated
by many other insects, are themselves eaten by the kingbird, catbird and
several other birds, though it is not known whether the stingless males
of these are singled out or not. Such facts as these do not discredit the
general theory of mimicry but point out its limits.

Evolution of Mimicry. Natural selection gives an adequate ex-
planation of the evolution of a mimetic pattern. Before accepting this
explanation, however, we must inquire: (i) What were the first stages
in the development of a mimetic pattern? (2) What evidence is there
that every step in this development was vitally useful, as the theory de-
mands that it should be? These pertinent questions have been answered
by Darwin, Wallace, Miiller, Dixey and several other authorities.

The incipient mimic must have possessed, to begin with, colors or
patterns that were capable of mimetic development; evidently the raw
material must have been present. Now Miiller and Dixey in particular
have called attention to the fact that many pierids have at least touches
of the reds, yellows and other colors that are so conspicuous in the heli-
conids. More than this, however, Dixey has demonstrated as appears
clearly from his colored figures a complete and gradual transition from
a typical non-mimetic pierid, Pieris locusta, to the mimetic pierid

IQ 2 ENTOMOLOGY

Mylothris pyrrha. the female of which imitates Heliconius numata. He
traces the transition chiefly through the males of several pierid species
for the males, though for the most part white (the typical pierid color),
"show on the under surface, though in varying degrees, an approach
towards the Heliconiine pattern that is so completely imitated by their
mates. These partially developed features on the under surface of the
males enable us to trace the history of the growth of the mimetic pattern."
Starting from Pieris locusta, it is an easy step to Mylothris lypera, thence
to M. lorena, and from this to the mimetic M. pyrrha. "Granted a
beginning, however small, such as the basal red touches in the normal
Pierines, an elaborate and practically perfect mimetic pattern may be
evolved therefrom by simple and easy stages."

Furthermore (in answer to the second question), it does not tax the
imagination to admit that any one of these color patterns has at least
occasionally been sufficiently suggestive of the heliconid type to pre-
serve the life of its possessor; especially when both bird and insect were
on the wing and perhaps some distance apart, when even a momentary
flash of red or yellow from a pierid might be enough to save it from
attack.

It is highly desirable, of course, that this plausible explanation should
be tested as far as possible by observations in the field and by experiments
as well.

Mimicry and Mendelism. The weight of evidence is at present
vastly in favor of the theory of mimicry as against any other explanation
of the facts, even though the theory is sometimes stretched to impossible
limits by some of its enthusiastic adherents. The only opposing opinion
that has sufficient plausibility to demand much consideration as yet
is that of Punnett.

In India and Ceylon the butterfly Papilio polytes has in addition to
the normal female a second form of female which mimics P. aristolochia
and a third which imitates P. hector; polytes being palatable to birds
and its two models unpalatable.

This case, described by Wallace almost fifty years ago, is one of the
classic examples of mimicry. Punnett holds, however, that these re-
semblances are of no practical value and that natural selection has played
no part in the formation of these polymorphic forms and suggests that
Mendelism offers a better explanation of the phenomenon a suggestion
that should be tested experimentally.

Adaptive Colors in General. Several classes of adaptive colors
have been discriminated and defined by Poulton, whose classification,

ADAPTIVE COLORATION

193

necessarily somewhat arbitrary but nevertheless very useful, is given
below, in its abridged form.

I. APATETIC COLORS. Colors resembling some part of the environment or

the appearance of another species.
A. CRYPTIC COLORS. Protective and Aggressive Resemblances.

1. Procryptic colors. Protective Resemblances. Concealment as a pro-

tection against enemies. Example: Kallima butterfly.

2. Aitticryptic colors. Aggressive Resemblances. Concea'ment in order

to facilitate attack. Example: Mantids with leaf-like appendages.
B. PSEUDOSEMATIC COLORS. False warning and signalling colors.

1. Pseudaposematic colors. Protective Mimicry. Example: Bee-like fly.

2. Pseudepisematic colors. Aggressive Mimicry and Alluring Coloration.

Examples: Vohiccllct, resembling bees (Fig. 247); Flower-like mantid.
II. SEMATIC COLORS. Warning and Signalling Colors.

1. Aposematic colors. Warning Colors. Examples: Gaudy colors of

stinging insects.

2. Episcmatic colors. Recognition Markings.
III. EPIGAMIC COLORS. Colors Displayed in Courtship.

Such of these classes as have not already been discussed need brief
reference.

Aggressive Resemblances. The resemblance of a carnivorous
animal to its surroundings may not only be protective but may also

FIG. 247. Aggressive mimicry. On the left, a bee, Bombiis mastntcatiis; on the right, a fly,

Volucrlla bonib\'l<jns. Natural size.

enable it to approach its prey undetected, as in the case of the polar bear

or the tiger. Among insects, however, the occurrence of aggressive

resemblance is rather doubtful, even in the case of the leaf-like mantids.

Aggressive Mimicry. Under this head are placed those cases in

which one species mimics another to which it is hostile. The best known

instance is furnished by European flies of the genus Volucella, whose

larvae feed upon those of bumble bees and wasps. The flies bear a close

14

ENTOMOLOGY

resemblance to the bees, owing to which it is supposed that the former
are able to enter the nests of the latter and lay their eggs.

Alluring Coloration. The best example of this phenomenon is
afforded by an Indian mantid, Gongylus gongyloides, which resembles so
perfectly the brightly colored flowers among which it hides that insects
actually fly straight into its clutches.

Recognition Markings. Though these are apparently important
among mammals and birds, as enabling individuals of the same species
quickly to recognize and follow one another, no special markings for
this purpose are known to occur among insects, not excepting the gregari-
ous migrant species, such as Anosia plexippus and the Rocky Mountain
locust.

Epigamic Colors. Among birds, frequently, the bright colors of
the male are displayed during courtship, and their evolution has been
attributed by Darwin and many of his followers to sexual selection a
highly debatable subject. Among insects, however, no such phenomenon
has been found ; whenever the two sexes differ in coloration the difference
does not appear to facilitate the recognition of even one sex by the other.

Evolution of Adaptive Coloration. Natural selection is the only
theory of any consequence that explains the highly involved phenomena
of adaptive coloration. Against such vague and unsupported theories
as the action of food, climate, laws of growth or sexual selection, natural
selection alone accounts for the multitudinous and intricate correlations
of color, pattern, form, attitude, movement, place, time, etc., that are
necessary to the development of a perfect case of protective resemblance
or mimicry. Natural selection cannot, of course, originate colors or any
other characters, its action being restricted to the preservation and
accumulation of such advantageous variations as may arise, from what-
ever causes. As Poulton says, the vast body of facts, utterly meaning-
less under any other theory, become at once intelligible as they fall har-
moniously into place under the principle of natural selection, to which,
indeed, they yield the finest kind of support.

CHAPTER VII

INSECTS IN RELATION TO PLANTS

Insects, in common with other animals, depend for food primarily
upon the plant world. No other animals, however, sustain such intimate
and complex relations to plants as insects do. The more luxuriant and
varied the flora, the more abundant and diversified is its accompanying
insect fauna.

Not only have insects become profoundly modified for using all kinds
and all parts of plants for food and shelter, but plants themselves have
been modified to no small extent in relation to insects, as appears in their
protective devices against unwelcome insects, in the curious formations
known as "galls','' the various insectivorous plants, and especially the
omnipresent and often intricate floral adaptations for cross-pollination
through the agency of insect visitors. Though insects have laid plants
under contribution, the latter have not only vigorously sustained the
attack but have even pressed the enemy into their own service, as it were.

Numerical Relations. The number of insect species supported by
one kind of plant is seldom small and often surprisingly large. The
poison ivy (Rkus toxicodendron] is almost exempt from attack, though
even this plant is eaten by a leaf -mining caterpillar, two pyralid larvae
and the larva of a scolytid beetle (Schwarz, Dyar). Horse-chestnut
and buckeye have perhaps a dozen species at most; elm has eighty;
birches have over one. hundred, and so have maples; pines are known to
harbor 170 species and may yield as many more; while our oaks sus-
tain certainly 500 species of insects and probably twice as many. Turn-
ing to cultivated plants, the clover is affected, directly or indirectly, by
about 200 species, including predaceous insects, parasites, and flower-
visitors. Clover grows so vigorously that it is able to withstand a great
deal of injury from insects. Corn is attacked by about 200 species, of
which 50 do notable injury and some 20 are pests. Apple insects num-
ber some 400 species.

Not uncommonly, an insect is restricted to a single species of plant.
Thus the caterpillar of Heodes hypophlccas feeds only on sorrel (Rumex ace-
toselld), so far as is known. The chrysomelid Chrysochus auratus appears
to be limited to Indian hemp (Apocynum andros&mifoliuwi) and to milk-

195

ENTOMOLOGY

weed (Asclepias) . In many instances an insect feeds indifferently upon
several species of plants provided these have certain attributes in com-
mon. Thus Argynnis cybele, aphrodite and atlantis eat the leaves of
various species of violets, and the Colorado potato beetle eats different
species of Solatium. Papilio thoas feeds upon orange, prickly ash and
other Rutaceae. Anosia plcxippns eats the various species of Asclepias
and also Apocynum androscemifolium; while Chrysochus also is limited
to these two genera of plants, as was said. These plants agree in having
a milky juice; in fact the two genera are rather nearly related botanically.
The common cabbage butterfly (Pieris rapes) though confined for the
most part to Crucifera, such as cabbage, mustard, turnip, radish, horse-
radish, etc., often develops upon Tro-
p&olum, which belongs to Geraniaceae;
all its food plants, however, have a
pungent odor, which is probably the
stimulus to oviposition.

Most phytophagous insects, how-
ever, range over many food plants.
The cecropia caterpillar has more than
sixty of these, representing thirty-one
genera and eighteen orders of plants;
and the tarnished plant bug (Lygus
pratensis) feeds indifferently on all sorts
of herbage, as does also the caterpillar
of Diacrisla mrginica. Many of the
insects of apple, pear, quince, plum,
peach, and other plants of the family
Rosaceae occur also on wild plants of
the same family; and the worst of our corn and wheat insects have
come from wild grasses. As regards number of food plants, the gypsy
moth "holds the record," for its caterpillar will eat almost any plant.
In Massachusetts, according to Forbush and Fernald, it fed in the field
upon 78 species of plants, in captivity upon 458 species (30 under stress
of hunger, the rest freely), and refused only 19 species, most of which
(such as larkspur and red pepper) had poisonous or pungent juices, or
were otherwise unsuitable as food. The migratory locust is notoriously
omnivorous, and perhaps eats even more kinds of plants than the gypsy
moth.

Galls. Most of the conspicuous plant outgrowths known as "galls"
are made by insects, though many of the smaller plant galls are made

FIG. 248. Holcaspis globulus. A ,
galls on oak, natural size; B, the gall-
maker, twice natural length.

INSECTS IN RELATION TO PLANTS

197

by mites (Acarina) and a few plant excrescences are due to nematode
worms and to fungi.

Among insects, Cynipidae (Hymenoptera) are pre-eminent as gall-

FIG. 249. Galls of Holcaspis ditricoria, on oak. Natural size.

makers and next to these, Cecidomyiidae (Diptera), Aphididae and
Psyllidae (Hemiptera); a few gall-insects occur among Tenthredinidae
(Hymenoptera) and Trypetidae (Diptera),
and one or two among Coleoptera and Lepi-
doptera.

Cynipidae affect the oaks (Figs. 248, 249)
far more often than any other plants, though
not a few species select the wild rose. Ceci-
domyiid galls occur on a great variety of
plants, and those of aphids on elm (Fig. 250),
poplar, and many other plants; while psyllid
galls are most frequent on hackberry. The
galls may occur anywhere on a plant, from
the roots to the flowers or seeds, though each
gall-maker always works on the same part of
its plant, root, stem, bud, leaf, leaf-vein,
flower, seed, etc.

Galls present innumerable forms, but the
form and situation of a gall are usually char-
acteristic, so that it is often possible to classify
galls as species even before the gall-maker is
known.

Gall-Making. The female simply lays the egg on the epidermis, or
else punctures the plant and deposits an egg in or near the cambium, or any
other tissue capable of growth; the egg hatches and the surrounding plant
tissue is stimulated to grow rapidly and abnormally into a gall, which

FIG. 250. Cockscomb gall
of Colo pita itlniicola, on elm.
Slightly reduced.

198 ENTOMOLOGY

serves as food for the larva; this transforms within the gall and escapes
as a winged insect. The physiology of gall-formation is far from being
understood. It has been found that the mechanical irritation from the
ovipositor is not the initial stimulus to the development of a gall; neither
is the fluid which is injected by the female during oviposition, this fluid
being probably a lubricant; if the egg is removed, the gall does not
appear. Ordinarily the gall does not begin to grow until the egg has
hatched, and then the gall grows along with the larva; exceptions to
this are found in some Tenthredinidae in which the egg itself increases
in volume, when the gall may grow with the egg. It appears that the
larva exudes some fluid which acts upon the protoplasm of certain plant
cells (the cambium and other cells capable of further growth and multi-
plication) in such a way as to stimulate their increase in size and number.
The following recent observations on this subject by A. Cosens are im-
portant. The cells of the plant that immediately surround the larva
are known as nutritive cells. In Cynipidae the larva gradually with-
draws the contents of these cells, by means of the mouth and not by
absorption, and the cells gradually collapse. The proportion of sugar
to starch decreases from the inside of the nutritive zone (nearest the
larva) to the outside. This is owing to an enzyme that changes starch
into sugar, the source of this enzyme being probably a pair of salivary
glands that open externally on each side just below the mouth of the
larva. The larva by accelerating the rate of change from starch to
sugar renders available to the plant more food than usual and therefore
stimulates the activity of the protoplasm toward greater cell-growth
and more rapid cell-reproduction. Thus the gall as well as the larva
draws food from the nutritive zone.

Why the gall should have a distinctive, or specific, form, it is not yet
known. There is no evidence that the form is of any adaptive impor-
tance, and the subject probably admits of a purely mechanical explana-
tion. One factor in determining the form of the gall is the direction in
which the stimulus is applied ; a spherical cynipid gall arising when the
influence is about equally distributed in all directions (Cosens).

Gall Insects. The study of gall insects is in many respects difficult.
It is not at all certain that an insect which emerges from a gall is the
species that made it; for many species, even of Cynipidae, make no galls
themselves but lay their eggs in galls made by other species. Such
guest-insects are termed inquilines. Furthermore, both gall-makers
and inquilines are attacked by parasitic Hymenoptera, making the in-
terrelations of these insects hard to determine. Many species of insects

INSECTS IN RELATION TO PLANTS 199

feed upon the substance of galls; thus Sharp speaks of as many as thirty
different kinds of insects, belong to almost all the orders, as having been
reared from a single species of gall.

Parthenogenesis and Alternation of Generations. Partheno-
genesis has long been known to occur among Cynipidoe. It has repeat-
edly been found that of thousands of insects emerging from galls of the
same kind, all were females. In one such instance the females were
induced by Adler to lay eggs on potted oaks, when it was found that the
resulting galls were quite unlike the original ones, and produced both
sexes of an insect which had up to that time been regarded as another
species. Besides parthenogenesis and this alternation of generations,
many other complications occur, making the study of gall-insects an
intricate and highly interesting subject.

Plant-Enemies of Insects. Most of the flowering plants are com-
paratively helpless against the attacks of insects, though there are many
devices which prevent "unwelcome" insects from entering flowers, for
instance, the sticky calyx of the catchfly (Silene mrginica), which
entangles ants and small flies. A few plants, however, actually feed upon
insects themselves. Thus the species of Drosera, as described in Dar-
win's classic volume on insectivorous plants, have specialized leaves for
the purpose of catching insects. The stout hairs of these leaves end
each in a globular knob, which secretes a sticky fluid. When a fly alights
on one of these leaves the hairs bend over and hold the insect; then a
fluid analogous to the gastric juice of the human stomach exudes, digests
the albuminoid substances of the insect and these are absorbed into the
tissues of the leaf; after which the tentacles unfold and are ready for the
next insect visitor. The Venus's flytrap is another well known example;
the trap, formed from the terminal portion of a leaf, consists of two
valves, each of which bears three trigger-like bristles, and when these are
touched by an insect the valves snap together and frequently imprison
the insect, which is eventually digested, as before. In the common
pitcher-plants, the pitcher, fashioned from a leaf, is lined with downward
pointing bristles, which allow an insect to enter but prevent its escape.
The bottom of the pitcher contains water, in which may be found the
remains of a great variety of insects which have drowned. There are
even nectar glands and conspicuous colors, presumably to attract in-
sects into these traps, where their decomposition products are more or
less useful to the plant. In Pinguicula the margin of a leaf rolls over
and envelops insects that have been caught by the glandular hairs of the
upper surface of the leaf, a copious secretion digests the softer portions of

2OO

ENTOMOLOGY

the insects, and the dissolved nitrogenous matter is absorbed into the
plant. Utricularia has little bladders which entrap small aquatic in-
sects. These plants are only partially dependent on insect-food, how-
ever, for they all possess chlorophyll.

Bacteria cause epidemic diseases among insects, as in the flacherie of
the silkworm; and fungi of a few groups are specially
adapted to develop in the bodies of living insects.

Those who rear insects know how frequently cater-
pillars and other larvae are destroyed by fungi that
give the insects a powdered appearance. These fungi,
referred to the genus Isaria, are in some cases known
to be asexual stages of forms of Cordyceps, which forms
appear from the bodies of various larvae, pupae and
imagines as long, conspicuous, fructifying sprouts (Fig.

The chief fungus parasites of insects belong to
the large family Entomophthoraceas, represented by
the common Empusa musca (Fig. 252) which affects
various flies. In autumn, especially in warm moist
weather, the common house fly may often be seen in a
dead or dying condition, sticking to a window-pane, its
abdomen distended and presenting alternate black and
white bands, while around the fly at a little distance
is a white powdery ring, or halo. The white interseg-
mental bands are made by threads of the fungus just
named, and the white halo by countless asexual spores
known as comdia, which have been forcibly discharged
from the swollen threads that bore them (Fig. 252) by
pressure, resulting probably from the absorption of
moisture. These spores, ejected in all directions, may
infect another fly upon contact and produce a growth
of fungus threads, or hyp/tec, in its body. The fungus
may be propagated also by means of resting spores,
as found by Thaxter, our authority upon the fungi of insects.

Empusa apJiidis is very common on plant lice and is an important
check upon their multiplication. Aphids killed by this fungus are
found clinging to their food plant, with the body swollen and discolored.
Empusa grylli attacks crickets, grasshoppers, caterpillars and other
forms. Curiously enough, grasshoppers affected by this fungus almost
always crawl to the top of some plant and die in this conspicuous position.

FIG. 251. Fruc-
tifying sprouts of
a fungus, Cordyccps
ravenelii, arising
from the body of a
white grub, Lachno-
slcrna. Slightly
reduced. After
RILEY.

INSECTS IN RELATION TO PLANTS

2OI

Sporotrichum, a genus of hyphomycetous fungi, affects a great variety
of insects, spreading within the body of the host and at length emerging
to form on the body of the insect a dense white felt-like covering, this
consisting chiefly of myriads of spores, by means of which healthy in-
sects may become infected. Under favorable conditions, especially in
moist seasons, contagious fungus diseases constitute one of the most
important checks upon the increase of insects and are therefore of vast
economic importance. Thus the termination (in 1889) f a disastrous
outbreak of the chinch bug in Illinois and neighboring states "was ap-
parently due chiefly, if not altogether, to parasitism by fungi. " Arti-

B

' ''^'^fW

Yefsgwv- * K -*.

:*>..' . ^

A:^.,K.V i

"i':ciKfet--. ' ' '- ^

FIG. 252. EwpHsn waster, the common fly-fungus. .4, house fly (Miisca donicslica), sur-
rounded by fungus spores (conidia); B, group of conidiophores showing conidia in several
stages of development; C, basidium (b) bearing conidium (c) before discharge. B and C
after THAXTER.

ficial cultures of the common Sporotrichum globuUfcrnm have been used
extensively as a means of spreading infection among chinch bugs and
grasshoppers, with, however, but moderate success as yet.

Insects in Relation to Flowers. Among the most marvelous
phenomena known to the biologist are the innumerable and complex
adaptations by means of which flowers secure cross pollination through
the agency of insect visitors. Cross fertilization is actually a necessity
for the continued vigor and fertility of flowering plants, and while some
of them are adapted for cross pollination by wind or water, the majority
of flowering plants exhibit profound modifications of floral structure for
compelling insects (and a few other animals, as birds or snails) to carry

202

ENTOMOLOGY

pollen from one flower to another. In general, the conspicuous colors
of flowers are for the purpose of attracting insects, as are also the odors
of flowers. Night-blooming flowers are often white or yellow and as a
rule strongly scented. Colors and odors, however, are simply indica-
tions to insects that edible nectar or pollen is at hand. Such is the usual
statement, and it is indeed probable that insects actually do associate
color and nectar, even though they will fly to bits of colored paper almost
as readily as they will to flowers of the same colors. It is not to be sup-

FIG. 253. Bumble bee (Bowbns) entering flower of blue-flag (Iris versicolor).

reduced.

Slightly

posed, however, that insects realize that they confer any benefit upon
the plant in the flowers of which they find food. At any rate, most
flowers are so constructed that certain insects cannot get the nectar or
pollen without carrying some pollen away, and cannot enter the next
flower of the same kind without leaving some of this pollen upon the
stigma of that flower. Take the iris, for example, which is admirably
adapted for pollination by a few bees and flies.

Iris. In the common blue-flag (Iris versicolor, Fig. 253) each of the

INSECTS IN RELATION TO PLANTS

203

an

three drooping sepals forms the floor of an arched passageway leading
to the nectar. Over the entrance and pointing outward in a movable
lip (Fig. 254, /), the outer surface of which is stigmatic. An entering
bee hits and bends down the free edge of this lip, which scrapes pollen
from the back of the insect and then springs back into place. Within
the passage, the hairy back of the bee rubs against an overhanging anther
(an) and becomes powdered with grains of pollen as the insect pushes
down towards the nectar. As the bee backs out of the passage it en-
counters the guardian lip again, but as this side of the lip can not re-
ceive pollen, immediate close pollination
is prevented. Of course, it is possible for
bees to enter another part of the same
flower or another flower of the same plant,
but as a matter of fact, they habitually
fly away to another plant; moreover, as
Darwin found, foreign pollen is prepotent
over pollen from the same flower. It may
be added that bees and other pollenizing
insects ordinarily visit in succession several
flowers of the same kind.

Orchids. The orchids, with their
fantastic forms, are really elaborate traps
to insure cross pollination. In some
orchids (Habenaria and others) the nec-
tar, lying at the bottom of a long tube,
is accessible only to the long-tongued
Sphingidae. While probing for the nectar,
a sphinx moth brings each eye against a
sticky disk to which a pollen mass is
attached, and flies away carrying the mass
on its eye. Then these pollinia bend down
on their stalks in such a way that when

the moth thrusts its head into the next flower they are in the proper
position to encounter and adhere to the stigma. The orchid Angracum
sesquipedale, of Madagascar, has a nectary tube more than eleven inches
long, from which Darwin inferred the existence of a sphinx moth with a
tongue equally long.

Milkweed. The various milkweeds are fascinating subjects to the
student of the interrelations of flowers and insects. The flowers, like
those of orchids, are remarkably formed with reference to cross pollina-

FIG. 254. Section to illustrate
cross pollination of Iris, an,
anther; /, stigmatic lip; ,
nectary; s, sepal.

2O4

ENTOMOLOGY

tion by insects. As a honey bee or other insect crawls over the flowers
(Fig. 255, ,4) to get the nectar, its legs slip in between the peculiar nec-
tariferous hoods situated in front of each anther. As a leg is drawn up-
ward one of its claws, hairs, or spines frequently catches in a V-shaped
fissure (/", Fig. 255, B) and is guided along a slit to a notched disk, or cor-
puscle (Fig. 255, C, d). This disk clings to the leg of the insect, which
carries off by means of the disk a pair of pollen masses, or pollinia (Fig.
2 55> C}. When first removed from their enclosing pockets, or anthers,
these thin spatulate pollinia lie each pair in the same plane, but in a few
seconds the two pollinia twist on their stalks and come face to face in
such a way that one of them can be easily introduced into the stigmatic

B

FIG. 255. Structure of milkweed flower (Asclcpias incarnata) with reference to cross
pollination. .4, a single flower; r, corolla; //, hood; B, external aspect of fissure (/) leading
up to disk and also into stigmatic chamber; h, hood; C, pollinia; d, disk. Enlarged.

chamber of a new flower visited by the insect. Then the struggles of the
insect ordinarily break the stem, or retinaculum, of the pollinium and
free the insect. Often, however, the insect loses a leg or else is per-
manently entrapped, particularly in the case of such large-flowered
milkweeds as Asclepias cornuti, w r hich often captures bees, flies and
moths of considerable size. Pollination is accomplished by a great
variety of insects, chiefly Hymenoptera, Diptera, Lepidoptera and Cole-
optera. These insects when collected about milkweed flowers usually
display the pollinia dangling from their legs, as in Fig. 256.

The details of pollination may be gathered by a close observer from
observations in the field and may be demonstrated to perfection by using
a detached leg of an insect and dragging it upward between two of the

INSECTS IN RELATION TO Pf.AX I'S

20:

FIG. 256. A wasp, Spl/cx iclim n-
mnuca, with pollinia of milkweed
;iti;irhed to its legs. Slightly en-
larged.

hoods of a flower; first to remove the pair of pollinia and then again

to introduce one of them into an empty stigmatic chamber.

Yucca. An extraordinary example of the interdependence of plants

and insects was made known by Riley,

whose detailed account is here summa-
rized. The yuccas of the southern United

States and Mexico are among the few

plants that depend for pollination each

upon a single species of insect. The

pollen of Yucca jilumcntosa cannot be

introduced into the stigmatic tube of

the flower without the help of a little

white tineid moth, Pronuba yuccasella,

the female of which pollenizes the flower

and lays eggs among the ovules, that

her larvae may feed upon the young

seeds. While the male has no unusual

structural peculiarities, the female is adapted for her special work by

modifications which are unique among Lepidoptera, namely, a pair of

prehensile and spinous maxillary
"tentacles" (Fig. 257, A) and a
long protrusible ovipositor (B)
which combines in itself the func-
tions of a lance and a saw.

The female begins to work
soon after dark, and will con-
tinue her operations even in the
light of a lantern. Clinging to a
stamen (Fig. 258) she scrapes off
pollen with her palpi and shapes
it into a pellet by using the front
legs. After gathering pollen from
several flowers she flies to another
flower, as a rule, thrusts her long
flexible ovipositor into the ovary
(Fig. 259) and lays a slender egg
alongside seven or eight of the
ovules. After laying one or more

eggs she ascends the pistil and actually thrusts pollen into the stigmatic

tube and pushes it in firmly. The ovules develop into seeds, some of

FIG. 257. Pronuba yuccasella. A , maxillary
tentacle and palpus; B, ovipositor. After
RILEY. Figures 257-259 are republished from
the Third Report of the Missouri Botanical
Garden, by permission.

206

ENTOMOLOGY

which are consumed by the larvae, though plenty are left to perpetuate
the plant itself. Three species of Pronuba are known, each restricted to
particular species of Yucca. Riley says that Yucca never produces seed
where Pronuba does not occur or where she is excluded artificially, and
that artificial pollination is rarely so successful as the normal method.

Why does the insect do this ? The little nectar secreted at the base
of the pistil appears to be of no consequence, at present, and the stigmatic
fluid is not nectarian; indeed, the tongue of Pronuba, used in clinging to
the stamen, seems to have lost partially or entirely its sucking power,
and the alimentary canal is regarded as functionless. Ordinarily it is
the flower which has become adapted to the insect, which is enticed by

FIG. 258. Pronuba yuccasella, fe-
male, gathering pollen from anthers of
Yucca. Enlarged.

FIG. 259. Pronuba moth ovipositing in flower of
Yucca. Slightly reduced.

means of pollen or nectar, but here is a flower which though entomo-
philous in general structure has apparently adapted itself in no way
to the single insect upon which it is dependent for the continuance of its
existence. More than this, the insect not only labors without compensa-
tion in the way of food, but has even become highly modified with refer-
ence to the needs of the plant, its special modifications being unparal-
leled among insects with the exception of bees, and being more puzzling
than the more extensive adaptations of the bees when we take into con-
sideration the impersonal nature of the operations of Pronuba. Further
investigation may render these extraordinary interrelations more in-
telligible than they are at present.

INSECTS IN RELATION TO PLANTS

207

The bogus Yucca moth (Prodoxus quinqucpunctella) closely resembles
and associates with Pronuba but oviposits in the flower stalks of Yucca
and has none of the special adaptive structures found in Pronuba.

As regards floral adaptations, these examples are sufficient for present
purposes; many others have been described by the botanist; in fact, the
adaptations for cross pollination by insects are as varied as the flowers
themselves.

Insect Pollenizers. The great majority of entomophilous flowers
are pollenized by bees of various kinds; the apple, pear, blackberry,
raspberry and many other rosaceous plants depend chiefly upon the
honey bee, while clover cannot set seed without the aid of bumble bees
or honey bees, assisted possibly by
butterflies. Lilies and orchids fre-
quently employ butterflies and
moths, as well as bees, and the
milkweed is adapted in a remark-
able manner for pollination by
butterflies, moths and some wasps,
as was described. Honeysuckle,
lilac, azalea, tobacco, Petunia,
Datura and many other strongly
scented and conspicuous nocturnal
flowers attract for their own uses
the long-tonged sphinx moths (Fig.
260); the evening primrose, like
milkweed, is a favorite of noctuid
moths. Umbelliferous plants are
pollenized chiefly by various flies,

but also by bees and wasps. Pond lilies, golden rod and some other flowers
are pollenized largely by beetles, though the flowers exhibit no special
modifications in relation to these particular insects. It is noteworthy
that pollination is performed only by the more highly organized insects,
the bees heading the list.

Of all the insects that haunt the same flower, it frequently happens
that only a few are of any use to the flower itself; many come for pollen
only; many secure the nectar illegitimately; thus bumble bees puncture
the nectaries of columbine, snapdragon and trumpet creeper from the
outside, and wasps of the genus Odynerus cut through the corolla of
Pentestemon Icevigatus, making a hole opposite each nectary; then there
are the many insects that devour the floral organs, and the insects which

FIG. 260. PliJc^ctlionlins sexta visiting flower
of Petunia. Reduced.

208

ENTOMOLOGY

are predaceous or parasitic upon the others. In the Iris, according to
Needham, two small bees (Clisodon terminalis and Osmia distincta) are
the most important pollenizers, and next to them a few syrphid flies,
while bumble bees also are of some importance. The beetle Trie hi us
piger and several small flies obtain pollen without assisting the plant,
and Pamphila, Eudamus, Chrysophanus and some other butterflies
succeed after many trials in stealing the nectar from the outside (Fig. 261).
A weevil (Mononychus vulpeculus) punctures the nectary, and the flowing

FIG. 261. A butterfly, Polites peckiiis, stealing nectar from a flower of Irix verxicolor.

reduced.

Slightly

nectar then attracts a great variety of insects. Grasshoppers and cater-
pillars eat the flowers, an ortalid fly destroys the buds, and several par-
asitic or predaceous insects haunt the plant; in all, over sixty species of
insects are concerned in one way or another with the Iris.

Modifications of Insects with Reference to Flowers. While
the manifold and exquisite adaptations of the flower for cross pollination
have engaged universal attention, very little has been recorded concern-
ing the adaptations of insects in relation to flowers. In fact, the adapta-
tion is largely one-sided; flowers have become adjusted to the structure

INSECTS IX RELATION TO PLANTS

209

of insects as a matter of vital necessity to put it that way while in-
sects have had no such urgent need so to speak in relation to floral
structure. They have been influenced by floral structure to some extent,
however, and in some cases to a very great extent, as appears from their
structural and physiological adaptations for gathering and using pollen
and nectar.

Among mandibulate insects, beetles and caterpillars that eat the
floral envelopes show no special modifications for this purpose; pollen-
feeding beetles, however, usually have the mouth parts densely clothed
with hairs, as in Euphoria (Fig. 262). In suctorial insects, the mouth

FIG. 262. A, right mandible; B, right maxilla; C. FIG. 263. Pollen-gather-

hypopharynx, of a pollen-eating beetle, Euphor ia inda. En- ing hair from a worker honey
larged. (The mandibles are remarkable in being two-lobed.) bee, with a pollen grain at-
tached. Greatly magnified.

parts are frequently formed with reference to floral structure; this is the
case in many butterflies and particularly in Sphingidae, in which the
length of the tongue bears a direct relation to the depth of the nectar}*
in the flowers that they visit. According to Miiller, the mouth parts of
Syrphidae, Stratyomyiidae and Muscidae are specially adapted for feed-
ing on pollen. In Apidae, the tongue as compared with that of other
Hymenoptera, is exceptionally long, enabling the insect to reach deep
into a flower, and is exquisitely specialized (Fig. 127) for lapping up and
sucking in nectar.

Pollen-gathering flies and bees collect pollen in the hairs of the body
or the legs; these hairs, especially dense and often twisted or branched
15

210

ENTOMOLOGY

(Figs. 263, 89) to hold the pollen, do not occur on other than pollen-
gathering species of insects. Caudell found that out of 200 species of
Hymenoptera only 23 species had branched hairs and that these species
belonged without exception to the pollen-gathering group Anthophila,
no representative of which was found without such hairs. Similar
branched hairs occur also on the flower-frequenting Bombyliidae and
Syrphidae.

The most extensive modifications in relation to flowers are found in

CO

3

FIG. 264. Adaptive modifications of the legs of the worker honey bee. A , outer aspect of
left hind leg; B, portion of left middle leg; C, inner aspect of tibio-tarsal region of left hind
leg; D, tibio-tarsal region of left fore leg; a, antenna comb; an, auricle; b, brush; c, coxa;
co, corbiculum; /, femur; p, pecten; pc, pollen combs; s, spur; sp, spines; ss, spines; t,
trochanter; //, tibia; v, velum; w, so-called wax pincers; 1-5, tarsal segments; /, metatarsus,
or planta.

Pronuba, as already described, and above all in Apidse, especially the
honey bee.

Honey Bee. The thorax and abdomen and the bases of the legs
are clothed with flexible branching hairs (Fig. 263), which entangle
pollen grains. These are combed out of the gathering hairs by means
of special pollen combs (Fig. 264, C, pc) on the inner surface of the planta
of the hind tarsus, the middle legs also assisting in this operation. From
these combs, the pollen is transferred to the pollen baskets, or corbicula
(Fig. 264, A, co), of the outer surface of each hind tibia, the pollen from

INSECTS IN RELATION TO PLANTS 2 I I

one side being transferred to the corbiculum of the opposite side. This
is accomplished in the following manner: the left pecten combs out the
pollen from the right planta and a mass of pollen forms just above the
left pecten at the lower end of the corbiculum; this mass gradually
grows larger and is pushed up along the corbiculum by the upward
movement of the auricle: Further details are given by Casteel, whose
admirably precise and thorough studies on the manipulation of pollen
and wax by the honey bee have corrected certain prevalent errors and
added much to our knowledge of the subject. Arriving at the nest, the
hind legs are thrust into a cell and the mass of pollen on each corbiculum
is pried out by means of a spur situated at the apex of the middle tibia
(Fig. 264, B, s), this lever being slipped in at the upper end of the corbicu-
lum and then pushed along the tibia under the mass of pollen; the spur is
used also in cleaning the wings, which explains its presence on queen
and drone, as well as worker, but the pollen-gathering structures of the
hind legs are confined to the worker. The so-called wax-pincers of the
hind legs (Fig. 264, A, C,w) at the tibio-tarsal articulation, have nothing
to do with the transfer of wax scales from the abdomen to the mouth,
according to Casteel; a wax scale being removed from its pocket by
becoming impaled on stiff spines at the distal end of the inner face of the
planta.

For cleaning the antennae, a front leg is passed over an antenna,
which slips into a semicircular scraper (Fig. 264, D, a) fashioned from the
basal segment of the tarsus; when the leg is bent at the tibio-tarsal
articulation, an appendage, or velum (v) of the tibia falls into place to
complete a circular comb, through which the antenna is drawn. This
comb is itself cleaned by means of a brush of hairs (b} on the front margin
of the tibia. A series of erect spines (sp) along the anterior edge of the
metatarsus is used as an eye brush, to remove pollen grains or other
foreign bodies from the hairs of the compound eyes. The labium, hypo-
pharynx and maxillae (Fig. 54) are exquisitely constructed with reference
to gathering and sucking nectar; the maxillae are used also to smooth
the cell walls of the comb; the mandibles (Fig. 45, C), notched in queen
and drone but with a sharp entire edge in the worker, are used for cut-
ting, scraping and moulding wax, as well as for other purposes. The
entire digestive system of the honey bee is adapted in relation to nectar
and pollen as food; the proventriculus forms a reservoir for honey and
is even provided at its mouth with a rather complex apparatus for strain-
ing the honey from the accompanying pollen grains, as described by
Cheshire. The wax glands (Fig. 102) are remarkable specializations in

212

ENTOMOLOGY

correlation with the food habits, as are also the various cephalic glands,
the chief functions of which are given as: (i) digestion, as the conversion
of cane sugar into grape sugar, and possibly starch into sugar; (2) the
chemical alteration of wax; (3) the production of special food substances,
which are highly important in larval development.

Numerous special sensory adaptations also occur. In fact, the
whole organization of the honey bee has become profoundly modified
in relation to nectar and pollen. Many other insects have the same food
but none of them sustain such intimate relations to the flowers as do the
bees.

Ant-Plants. There are several kinds of tropical plants which are
admirablv suited to the ants that inhabit them. Indeed, it is often as-

'b

FIG. 265. Acacia splueroccphaia. an ant-piant. b, one of the "Beit's bodies"; g, gland;
s, s, hollow stipular thorns, perforated by ants. Reduced. From Strasburger's Lehrbuch
der Botanik.

serted that these plants have become modified with special reference to
their use by ants, though this is a gratuitous and improbable assumption.

Belt found several species of Acacia in Nicaragua and the Amazon
valley which have large hollow stipular thorns, inhabited by ants of the
genus Pseudomyrma . The ants enter by boring a hole near the apex of
a thorn (Fig. 265, s}. The plant affords the ants food as well as shelter,
for glands (g) at the bases of the petioles secrete a sugary fluid, while
many of the leaflets are tipped with small egg-shaped or pear-shaped
appendages (b) known as "Belt's bodies," which are rich in albumin,
fall off easily at a touch, and are eaten by the ants. These ants drive
away the leaf-cutting species, incidentally protecting the tree in which
they live.

The ant-trees (Cecropia adenopus) of Brazil and Central America
have often been referred to bv travelers. When one of these trees is

INSECTS IN RELATION TO PLANTS

213

handled roughly, hosts of ants rush out from small openings in the stems
and pugnaciously attack the disturber. Just above the insertion of
each leaf is a small pit (Fig. 266, a, b] where the wall is so thin as to form
a mere diaphragm, through which an ant (probably a fertilized female)
bores and reaches a hollow internode. To establish communication be-
tween the internodal chambers, the ants bore through the intervening
septa (Fig. 267). They seldom leave the Cecropia plant, unless disturbed,
and even keep herds of aphids in their abode. The base of each petiole

-b

FIG. 266. Portion of young stem of Cecropia adctwpiis, FIG. 267. Cecropia adenopus.

showing internodal pits, a and b. Natural size. Portion of a stem, split so as to show

Figures 266-268 are from Schimper's Pflanzengeographie. internodal chambers and the inter-
vening septa perforated by ants.

bears (Fig. 268) tender little egg-like bodies ("Muller's bodies") which
the ants detach, store away and eat; the presence of these bodies is a
sure sign that the tree is uninhabited by these ants, which, by the way,
belong to the genus Aztec a.

It is too much to assert that the ants protect the Cecropia plant /;/
return for the food and shelter which they obtain. All ants are hostile
to all other species of ants, with few exceptions, and even to other col-
onies of their own species; so that their assaults upon leaf-cutting ants
are by no means special and adaptive in their nature, and any protec-

214

ENTOMOLOGY

tion that a plant derives thereby is merely incidental. Furthermore,
hollow stems, glandular petioles and pitted stems are of common oc-
currence when they bear no relation to the needs of ants. These inter-
relations of ants and plants are too often misinterpreted in popular and
uncritical accounts of the subject.

The interesting habits of the leaf-cutting ants in relation to the
plants that they attack are described in a subsequent chapter, where
will be found also an account of the harvesting ants.

FIG. 268. Cecropia adenopus. Base of
petiole showing "Miiller's bodies."
Slightly reduced.

FIG. 269.- Hydnophytum montannm. Sec-
tion of pseudo-bulb, to show chambers inhabited
by ants. One-fourth natural size. After FOREL.

The epiphytic plants Myrmecodia and Hydnophytum, of Java, form
spongy bulb-like masses, the chambers of which are usually tenanted by
ants, which rush forth when disturbed. These lumps (Fig. 269) are
primarily water-reservoirs, but the ants utilize them by boring into them
and from one chamber into another. In plants of the genus Hwnboldtia
the ants can enter the hollow internodes through openings that already
exist.

CHAPTER VIII

INSECTS IN RELATION TO OTHER ANIMALS

On the one hand, insects may derive their food from other animals,
either living or dead; on the other hand, insects themselves are food for
other animals, especially fishes and birds, against which they protect
themselves by various means, more or less effective. These topics form
the principal subject of the present chapter.

Predaceous Insects. Innumerable aquatic insects feed largely or
entirely upon microscopic Protozoa, Rotifera, Entomostraca, etc.; this
is especially the case with culicid and chironomid larvae. Many aquatic
Hemiptera and Coleoptera prey upon planarians, nematodes, annelids,
molluscs and crustaceans; Belostoma sometimes pierces the bodies of tad-
poles and small fishes ; Dytiscus also kills young fishes occasionally and is
distinctly carnivorous both as larva and imago. Among terrestrial
insects, Carabidae are notably predaceous, preying not only upon other
insects but also upon molluscs, myriopods, mites and spiders. Ants do
not hesitate to attack all kinds of animals; in the tropics the wandering
ants (Eciton) attack lizards, rats and other vertebrates, and it is said that
even huge serpents, when in a torpid condition, are sometimes killed by
armies of these pugnacious insects.

Mosquitoes affect not only mammals but also, though rarely, fishes
and turtles. The gadflies (Tabanidas) torment horses and cattle by
their punctures; and the black-flies, or buffalo gnats (Simulium), per-
secute horses, mules, cattle, fowls, and frequently become unendurable
even to man. The notorious tsetse fly (Glossina morsitans) of South
Africa spreads a deadly disease among horses, cattle and dogs, by inocu-
lating them with a protozoan blood-parasite, to the effects of which,
fortunately, man is not susceptible.

Parasitic Insects. Insects belonging to several diverse orders have
become peculiarly modified to exist as parasites either upon or within the
bodies of birds or mammals.

Almost all birds are infested by Mallophaga, or bird lice, of which
Kellogg has catalogued 264 species from 257 species of North American
birds. Sometimes a species of Mallophaga is restricted to a single
species of bird, though in the majority of cases this is not so. Several

215

2l6 ENTOMOLOGY

mallophagan species often infest a single bird; thus nine species occur
on the hen, and no less than twelve species, representing five genera, on
the American coot. These parasites spread by contact from male to
female, from old to young, and from one bird to another when the birds
are gregarious. When a single species of bird louse occurs on two or
more hosts, these are almost always closely allied, and Kellogg has sug-
gested the interesting possibility that such a species has persisted un-
changed from a host which was the common ancestor of the two or more
present hosts. Mallophaga are not altogether limited to birds, however,
for they may be found on cattle, horses, cats, dogs, and some other
mammals; Kellogg records eighteen species from fifteen species of mam-
mals. These biting lice feed, not upon blood, but upon epidermal cells
and portions of feathers or hairs. They have flat tough bodies (Fig. 17),
with no traces of wings, and a large head with only simple eyes; the eggs
are glued to feathers or hairs.

Mammals only are infested by the sucking lice, or Pediculidae (Hemip-
tera). These (Fig. 23) have a large oval or rounded abdomen, no wings,
a small head, minute simple eyes or none, and claws that are adapted to
clutch hairs; the eggs are glued to hairs. Sucking lice affect horses,
cattle, sheep, dogs, monkeys, seals, elephants, etc., and man is para-
sitized by three species, namely, the head louse (Pediculus capitis], the
body louse (Pediculus vestimenti) , and the crab louse (Phthirius pubis] ,
though the first two are possibly the same species.

An anomalous beetle, Platypsyllus castoris, occurs throughout North
America and also in Europe as a parasite of the beaver.

The fleas, allied to Diptera but constituting a distinct order (Siphon-
aptera), are familiar parasites of chickens, cats, dogs and human beings.
These insects (Fig. 30) are well adapted by their laterally compressed
bodies for slipping about among hairs, and their saltatory powers and
general elusiveness are well known. Their wings are reduced to mere
rudiments, their eyes when present are minute and simple and their
mouth parts are suctorial.

Among Diptera, there are a few external parasites, the best known
of which is the sheep tick (Melophagus ovinus), though several highly
interesting but little-studied forms are parasitic upon birds and bats.

The larvae of the bot-flies (CEstridae) are common internal parasites of
mammals. The sheep bot-fly (CEstrus oms} deposits her eggs or larvae on
the nostrils of sheep; the maggots develop in the frontal sinuses of the
host, causing vertigo or even death, and when full grown escape through
the nostrils and pupate in the soil. The horse bot-fly (Gastrophilus cqui)

INSECTS IX RELATION TO OTHER ANIMALS 21 7

glues its eggs to the hairs of horses, especially on the fore legs and shoul-
ders, whence the larva? are licked off and swallowed; once in the stomach,
the bots fasten themselves to its lining, by means of special hooks, and
withstand almost all efforts to dislodge them; though when the bots have
attained their growth they release their hold and pass with the excrement
to the soil. Bots of the genus Hypodcrma form tumors on cattle and
other mammals, domesticated or wild. The ox-warble (//. lincalii,
Fig. 211, /) reaches the oesophagus of its host in the same manner as the
horse bot, according to Curtice, but then makes its way into the sub-
cutaneous tissue and causes the well-known tumors on the back of the
animal; when full grown the bots squirm out of these tumors and drop
to the ground, leaving permanent holes in the hide.

Parasitism in General. Parasitic insects evidently do not consti-
tute a phylogenetic unit, but the parasitic habit has arisen independently
in many different orders. These insects do, however, agree superficially,
in certain respects, as the result of what may be termed convergence of
adaptation. Thus a dipterous larva, living as an internal parasite, in
the presence of an abundant supply of food, has no legs, no eyes or anten-
nae, and the head is reduced to a mere rudiment, sufficient simply to
support a pair of feeble jaws; the skin, moreover, is no longer armor-like
but is thin and delicate, the body is compact and fleshy, and the digestive
system is of a simplified type. The same modifications are found in
hymenopterous larvae, under similar food-conditions, except that the
head usually undergoes less reduction. The various external parasites
lack wings, almost invariably, and the eyes, instead of being compound,
are either simple or else absent. In some special cases, however, as in a
few dipterous parasites of birds and bats, the wings are present, either
permanently or only temporarily, enabling the insects to reach their
hosts.

This so-called parasitic degeneration, widespread among animals in
general and consisting chiefly in the reduction or loss of locomotor and
sensory functions in correlation with an immediate and plentiful supply
of food, results in a simplicity of organization which is to be regarded
not as a primitive condition but as an expression of what is, in one
sense, a high degree of specialization to peculiar conditions of life. This
exquisite degree of adaptation to a special environment, however, sacri-
fices the general adaptability of the animal, makes it impossible for a
parasite to adapt itself to new conditions; and while parasitism may be
an immediate advantage to a species, there are few parasites that have
attained any degree of dominance among animals. Ichneumonidae, to

2l8 ENTOMOLOGY

be sure, are remarkably dominant among insects, but their parasitic
adaptations are limited for the most part to the larval stage and the
adults may be said to be as free for new adaptations as any other
Hymenoptera.

Scavenger and Carrion Insects. Not a few families of Diptera
and Coleoptera derive their food from dead animal matter. The aquatic
families Dytiscidae and Gyrinidae are largely scavengers. Among terres-
trial forms, Silphidae feed on dead animals of all kinds; the burying
beetles (Necrophorus), working in pairs, undermine and bury the bodies
of birds, frogs and other small animals, and lay their eggs in the carcasses;
Histeridse and Staphylinidae are carrion beetles, and Dermestidae attack
dried animal matter of almost every description, their depredations upon
furs, feathers, museum specimens, etc., being familiar to all. Ants are
famous as scavengers, destroying decaying organic matter in immense
quantities, particularly in the tropics. Many Scarabaeidae feed upon
excrementitious matter, for example the " tumble-bugs," which are
frequently seen in pairs, laboriously rolling along or burying a large ball
of dung, which is to serve as food for the larva.

Insects as Food for Vertebrates. Lizards, frogs and toads are
insectivorous, especially toads. The American toad feeds chiefly upon
insects, which form 77 per cent, of its food for the season, the remainder
consisting of myriopods, spiders, Crustacea, molluscs and worms, accord-
ing to the observations of A. H. Kirkland, who states that Lepidoptera
form 28 per cent, of the total insect food, Coleoptera 27, Hymenoptera 19
and Orthoptera 3 per cent. The toad does not capture dead or motion-
less insects but uses its extensile sticky tongue to lick in moving insects
or other prey, which it captures with surprising speed and precision. In
the cities one often sees many toads under an arc-light engaged in catch-
ing insects that fall anywhere near them. Though its diet is varied and
somewhat indiscriminate, the toad consumes such a large proportion of
noxious insects, such as May beetles and cutworms, that it is unques-
tionably of service to man.

Moles are entirely insectivorous and destroy large numbers of white
grubs and caterpillars; field mice and prairie squirrels eat many insects,
especially grasshoppers, and the skunk revels in these insects, though it
eats beetles frequently, as does also the raccoon, which is to some extent
insectivorous. Monkeys are omnivorous but devour many kinds of
insects.

With these hasty references, we may pass at once to the subject of
the insect food of fishes and birds.

INSECTS IN RELATION TO OTHER ANIMALS 219

Insects in Relation to Fishes. Insects constitute the most im-
portant portion of the food of adult fresh water fishes, furnishing forty
per cent, of their food, according to Dr. Forbes, from whose valuable
writings the following extracts are taken.

"The principal insectivorous fishes are the smaller species, whose
size and food structures, when adult, unfit them for the capture of Ento-
mostraca, and yet do not bring them within reach of fishes or Mollusca.
Some of these fishes have peculiar habits which render them especially
dependent upon insect life, the little minnow Phenacobius,-for example,
which, according to my studies, makes nearly all its food from insects
(ninety-eight per cent.) found under stones in running water. Next are
the pirate perch, Aphredoderus (ninety-one per cent.), then the darters
(eighty-seven per cent.), the croppies (seventy-three per cent.), half-
grown sheepshead (seventy-one per cent.), the shovel fish (fifty-nine per
cent.), the chub minnow (fifty-six per cent.), the black warrior sunfish
(Chccnobryttus) and the brook silversides (each fifty-four per cent.), and
the rock bass and the cyprinoid genus Notropis (each fifty- two percent.).

"Those which take few insects or none are mostly the mud-feeders
and the ichthyophagous species, Amia (the dog-fish) being the only
exception noted to this general statement. Thus we find insects wholly
or nearly absent from the adult dietary of the burbot, the pike, the gar,
the black bass, the wall-eyed pike, and the great river catfish, and from
that of the hickory shad and the mud-eating minnows (the shiner, the
fathead, etc.). It is to be noted, however, that the larger fishes all go
through an insectivorous stage, whether their food when adult be almost
wholly other fishes, as \vith the gar and the pike, or molluscs, as with the
sheepshead. The mud-feeders, however, seem not to pass through this
stage, but to adopt the limophagous habit as soon as they cease to de-
pend upon Entomostraca.

"Terrestrial insects, dropping into the water accidentally or swept
'in by rains, are evidently diligently sought and largely depended upon
by several species, such as the pirate perch, the brook minnow, the top
minnows or killifishes (cyprinodonts) the toothed herring and several
cyprinoids (Semotilus, Pimephales and Notropis).

"Among aquatic insects, minute slender dipterous larvae, belonging
mostly to Chironomus, Corethra and allied genera are of remarkable
importance, making, in fact, nearly one tenth of the food of all the fishes
studied. They are most abundant in Phcnacobius and Ethcostoma,
which genera have become especially adapted to the search for these
insect forms in shallow rocky streams. Next I found them most gener-

220 ENTOMOLOGY

ally in the pirate perch, the brook silversides, and the stickleback, in
which they averaged forty-five per cent. They amounted to about one
third the food of fishes as large and important as the red horse and the
river carp, and made nearly one fourth that of fifty-one buffalo fishes.
They appear further in considerable quantity in the food of a number of
the minnow family (Notropis, Pimcphales, etc.), which habitually fre-
quent the swift waters of stony streams, but were curiously deficient in
the small collection of miller's thumbs (Cottidae) which hunt for food in
similar situations. The sunfishes eat but few of this important group,
the average of the family being only six per cent.

"Larvae of aquatic beetles, notwithstanding the abundance of some
of the forms, occurred in only insignificant ratios, but were taken by fifty-
six specimens, Belonging to nineteen of the species, more frequently by
the sunfishes than by any other group. The kinds most commonly
captured were larvae of Gyrinidae and Hydrophilidae; whereas the adult
surface beetles themselves (Gyrimts, Dineutes, etc.) whose zigzag-
darting swarms no one can have failed to notice were not once en-
countered in my studies.

"The almost equally well-known slender water-skippers (Hygro-
trechus) seem also completely protected by their habits and activity
from capture by fishes, only a single specimen occurring in the food of
all my specimens. Indeed, the true water bugs (Hemiptera) were gener-
ally rare, with the exception of the small soft-bodied genus Corisa,
which was taken by one hundred and ten specimens, belonging to twenty-
seven species, most abundantly by the sunfishes and top minnows.

"From the order Neuroptera [in the broad sense] fishes draw a larger
part of their food than from any other single group. In fact, nearly a
fifth of the entire amount of food consumed by all the adult fishes exam-
ined by me consisted of aquatic larvae of this order, the greater part of
them larvae of day flies (Ephemeridae) , principally of the genus Hexa-
genia. These neuropterous larvae were eaten especially by the miller's
thumb, the sheepshead, the white and striped bass, the common perch,
thirteen species of the darters, both the black bass, seven of the sunfishes,
the rock bass and the croppies, the pirate perch, the brook silversides,
the sticklebacks, the mud minnow, the top minnows, the gizzard shad,
the toothed herring, twelve species each of the true minnow family and
of the suckers and buffalo, five catfishes, the dog-fish, and the shovel
fish, seventy species out of the eighty-seven which I have studied.

"Among the above, I found them the most important food of the
white bass, the toothed herring, the shovel fish (fifty-one per cent.), and

INSECTS IN RELATION TO OTHER ANIMALS 221

the croppies; while they made a fourth or more of the alimentary con-
tents of the sheepshead (forty-six per cent.), the darters, the pirate perch,
the common sunfishes (Lepomis and Chcenobryttus), the rock bass, the
little pickerel, and the common sucker (thirty-six per cent.).

"Ephemerid larvas were eaten by two hundred and thirteen specimens
of forty-eight species not counting young. The larva of Hexagenia,
one of the commonest of the 'river flies,' was by far the most important
insect of this group, this alone amounting to about half of all the Neurop-
tera eaten. It made nearly one half of the food of the shovel fish,
more than one tenth that of the sunfishes, and the principal food resource
of half-grown sheepshead; but was rarely taken by the sucker family,
and made only five per cent, of the food of the catfish group.

"The various larvae of the dragon flies, on the other hand, were much
less frequently encountered. They seemed to be most abundant in the
food of the grass pickerel (twenty-five per cent.) and next to that, in the
croppie, the pirate perch, and the common perch (ten to thirteen per
cent.).

"Case-worms (Phryganeidae) were somewhat rarely found, rising to
fifteen per cent, in the rock bass and tw r elve per cent, in the minnows of
the Hybopsis group, but otherwise averaging from one to six per cent,
in less than half of the species."

Insects in Relation to Birds. From an economic point of view
the relations between birds and insects are extremely important, and from
a purely scientific standpoint they are no less important, involving as
they do biological interactions of remarkable complexity.

The prevalent popular opinion that birds in general are of inestimable
value as destroyers of noxious insects is a correct one, as Dr. Forbes
proved, from his precise and extensive studies upon the food of Illinois
birds, involving a laborious and difficult examination of the stomach
contents of many hundred specimens. All that follows is taken from
Forbes, when no other author's name is mentioned, and though the
percentages given by Forbes apply to particular years and would un-
doubtedly vary more or less from year to year, they are here for con-
venience regarded as representative of any year and are spoken of in the
present tense. About two thirds of the food of birds consists of insects.

Robin. The food of the robin in Illinois, from February to May
inclusive, consists almost entirely of insects; at first, larvae of Bibio albi-
pcnnis for the most part, and then caterpillars and various beetles.
When the small fruits appear, these are largely eaten instead of insects;
thus in June, cherries and raspberries form fifty-five per cent, and insects

222 ENTOMOLOGY

(ants, caterpillars, wire-worms and Carabidae) forty-two per cent, of the
food; and in July, raspberries, blackberries and currants form seventy-
nine per cent, and insects (mostly caterpillars, beetles and crickets) but
twenty per cent, of the food. In August, insects rise to forty-three per
cent, and fruits drop to fifty-six per cent., and these are mostly cherries,
of which two thirds are wild kinds. In September, ants form fifteen per
cent, of the food, caterpillars five per cent, and fruits (mostly grapes,
mountain-ash berries and moonseed berries) seventy per cent. In
October, the food consists chiefly of wild grapes (fifty-three per cent.),
ants (thirty-five per cent.), and caterpillars (six per cent.).

For the year, judging from the stomach contents of one hundred and
fourteen birds, garden fruits form only twenty-nine per cent, of the food
of the robin, while insects constitute two thirds of the food. The results
are confirmed by those of Professor Beal in Michigan, who found that
more than forty-two per cent, of the food of the robin consists of insects
with some other animal matter, the remainder being made up of various
small fruits, but notably the wild kinds.

Upon the whole, the robin deserves to be protected as an energetic
destroyer of cutworms, white grubs and other injurious insects, and the
comparatively few cultivated berries that the bird appropriates are
ordinarily but a meagre compensation for the valuable services rendered
to man by this familiar bird.

Catbird. Not so much can be said for the catbird, however, for,
though its food habits are similar to those of the robin, it arrives later and
departs earlier, with the result that it is less dependent than the robin
upon insects and that berries form a larger percentage of its total food.

In May, eighty-three per cent, of the food of the catbird consists of
insects, mostly beetles (Carabidae, Rhynchophora, etc.), crane-flies, ants
and caterpillars (Noctuidae); while dry sumach berries are eaten to the
extent of seven per cent. For the first half of June, the record is much
the same, with an increase, however, in the number of May beetles eaten;
in the second half of the month the food consists chiefly of small fruits,
especially raspberries, cherries and currants; so that for the month as a
whole, only forty-nine per cent, of the food is made up of insects. This
falls to eighteen per cent, in July, when three quarters of the food con-
sists of small fruits, mostly blackberries, however. In August, with the
diminution of the smaller cultivated fruits, the percentage of insects
rises to forty-six per cent., nearly one half of which is made up of ants
and the rest of caterpillars, grasshoppers, Hemiptera, Coleoptera, etc.
In September, with the appearance of wild cherries, elderberries, Virginia

INSECTS IN RELATION TO OTHER ANIMALS 223

creeper berries and grapes, these are eaten to the extent of seventy-six
per cent., the insect element of the food falling to twenty-one per cent.,
of which almost half consists of ants, and the remainder of beetles and a
few caterpillars.

For the entire year, as appears from the study of seventy specimens
by Forbes, insects form forty-three per cent, of the food of the catbird
and fruits fifty-two per cent. As the injurious insects killed are offset
by the beneficial ones destroyed, "the injury done in the fruit-garden by
these birds remains without compensation unless we shall find it in the
food of the young," says Professor Forbes. And this has been found, to
the credit of the catbird ; for Weed learned that the food of three nest-
lings consisted of insects, sixty-two per cent, of which were cutworms
and four per cent, grasshoppers; while Judd found that fourteen nest-
lings had eaten but four per cent, of fruit, the diet being chiefly ants,
beetles, caterpillars, spiders and grasshoppers. In fact, Weed believes
that, on the whole, the benefit received from the catbird is much greater
than the harm done, and that its destruction should never be permitted
except when necessary in order to save precious crops.

Bluebird. The excellent reputation which the bluebird bears every-
where as an enemy of noxious insects is well deserved. From a study of
one hundred and eight Illinois specimens, Forbes finds that seventy-
eight per cent, of the food for the year consists of insects, eight per cent.
of Arachnida, one per cent, of Julidae and only thirteen per cent, of vege-
table matter, edible fruits forming merely one per cent, of the entire food.
The insects eaten are mostly caterpillars (chiefly cutworms), Orthoptera
(grasshoppers and crickets) and Coleoptera (Carabidae and Scarabaeidae) .
Though some of the insects are more or less beneficial to man, such as
Carabidae and Ichneumonidae (respectively predaceous and parasitic),
the beneficial elements form only twenty-two per cent, of the food for
the year, as against forty-nine per cent, of injurious elements, the remain-
ing twenty-nine per cent, consisting of neutral elements. The food of
the nestlings, according to Judd, is essentially like that of the adults,
being "beetles, caterpillars, grasshoppers, spiders and a few snails."

Other Insectivorous Birds. Weed and Dearborn, from whose
excellent work the following notes are taken, find that the common
chickadee devours immense numbers of canker-worms, and that more
than half its food during winter consists of insects, largely in the form of
eggs, including those of the common tent caterpillar (C. americana),
the fall web-worm (H. cunea) and particularly plant lice, whose eggs, small
as they are, form more than one fifth of the entire food; more than four
hundred and fifty of them are sometimes eaten by a single bird in one day,

224 ENTOMOLOGY

and the total number destroyed annually is inconceivably large. The
house wren is almost exclusively insectivorous, feeding upon caterpillars
and other larvee, ants, grasshoppers, gnats, beetles, bugs, spiders, and
myriopods. The swallows, also, are highly insectivorous; "most of
their food is captured on the wing, and consists of small moths, two-
winged flies, especially crane-flies, beetles in great variety, flying bugs,
and occasionally small dragon-flies. The young are fed with insects."
Ninety per cent, of the food of the kingbird "consists of insects, includ-
ing such noxious species as May-beetles, click-beetles, wheat and fruit
weevils, grasshoppers, and leaf hoppers." The honey bees eaten by this
bird are insignificant in number. Woodpeckers destroy immense num-
bers of wood-boring larvae, bark-insects, ants, caterpillars, etc. The
cuckoos "are unique in having a taste for insects that other birds reject.
Most birds are ready to devour a smooth caterpillar that comes in their
way, but they leave the hairy varieties severely alone. The cuckoos,
however, make a specialty of devouring such unpalatable creatures;
even stink-bugs and the poisonous spiny larvae of the lo moth are freely
taken." Caterpillars form fifty per cent, of the food for the year;
Orthoptera (grasshoppers, katydids, and tree crickets), thirty per cent.;
Coleoptera and Hemiptera, six per cent, each; and flies and ants are
taken in small quantities. "The nestling birds are fed chiefly with
smooth caterpillars and grasshoppers, their stomachs probably being
unable to endure the hairy caterpillars. All in all, the cuckoos are of
the highest economic value. They do no harm and accomplish great
good. If the orchardist could colonize his orchards with them, he would
escape much loss." The quail feeds largely upon insects during the
summer, frequently eating the Colorado potato beetle and the army
worm; the prairie hen has similar food habits but lives almost exclusively
on grasshoppers, when these are abundant.

The Insect Food of Birds. "There are few groups of injurious
insects that enter so largely into the composition of the food of birds as
do the locusts, or short-horned grasshoppers, of the family Acridiidae.
The enormous destructive power of these insects is well known, but our
indebtedness to birds in checking their oscillations is less generally recog-
nized." Professor Aughey, who has made extensive studies upon the
relation of birds to the Rocky Mountain locust, found that upon one
occasion 6 robins had eaten 265 of these insects, 5 catbirds 152, 3 blue-
birds 67, 7 barn swallows 139, 7 night hawks 348, 16 yellow-billed cuckoos
416, 8 flickers 252, 8 screech owls 219, and i humming bird 4; while
crows and blue- jays had eaten large numbers of the locusts; and grouse,
quail and prairie hen, enormous numbers. Even shore birds, such as

INSECTS IN RELATION TO OTHER ANIMALS

22-

geese, ducks, gulls and pelicans came to share in the feast. Aughey
estimated that the locusts eaten in one day by the passerine birds of the
eastern half of Nebraska were sufficient to destroy in a single day 174,397
tons of crops, valued at $1,743.97.

Weed and Dearborn state that, of Hemiptera, Jassichu are very often
found in the stomachs of birds, and that aphids and their eggs form a
large part of the food of many of the smaller birds, such as the warblers,
nuthatches, kinglets and chickadees. "A large proportion of the cater-
pillars of the Lepidoptera are eagerly devoured by birds, forming an
important element of the food of many species. The hairy caterpillars
are eaten by cuckoos and blue-jays and the large saturniid caterpillars,
such as cecropia and polyphemus, by some of the hawks. Almost all
kinds of Coleoptera are food for birds, but especially the grubs of Scara-
baeidae, which are eagerly devoured by robins, blackbirds, crows and
other birds. Of the Diptera, Cecidomyiidae and other gnats are eaten
by swallows, swifts and night hawks; while Tipulidae are often found in
the stomachs of birds. Among Hymenoptera, ants are eaten exten-
sively by woodpeckers, catbirds and many other species, as are also
Ichneumonidae and other parasitic forms these last by the flycatchers
in particular.

The Regulative Action of Birds upon Insect Oscillations. -
The worst injuries by insects are done by species that fluctuate exces-
sively in number as the result of variations in those manifold forces that
act as checks upon the multiplication of the species.

In order to determine whether birds do anything to reduce existing
oscillations of injurious insects, Professor Forbes made some admirable
studies upon the food of birds which were shot in an Illinois apple or-
chard which was being ravaged by canker-worms. In this orchard,
birds were present in extraordinary number and variety, there being at
least thirty-five species, most of which were studied by Forbes, from
whose exhaustive tables the following food-percentages are taken :

Birds Examined.

Robin, Q

Catbird, 14

Brown Thrush, 4

Bluebird,

Black-capped Chickadee,

House Wren, 5

Tennessee Warbler, i

Summer Yellow Bird, 5

Black-throated Green Warbler,. . . i

Maryland Yellow-throat,

Baltimore Oriole, 3

Insects.

Canker-worms.

93%

21%

98

15

94

12

98

12

IOO

61

9i

46

IOO

80

94

67

IOO

70

IOO

37

IOO

40

16

226 ENTOMOLOGY

To quote Forbes: "Three facts stand out very clearly as results of
these investigations: i. Birds of the most varied character and habits,
migrant and resident, of all sizes, from the tiny wren to the blue-jay, birds
of the forest, garden and meadow, those of arboreal and those of terres-
trial habits, were certainly either attracted or detained here by the
bountiful supply of insect food, and were feeding freely upon the species
most abundant. That thirty-five per cent, of the food of all the birds
congregated in this orchard should have consisted of a single species of
insect, is a fact so extraordinary that its meaning can not be mistaken.
Whatever power the birds of this vicinity possessed as checks upon
destructive irruptions of insect life was being largely exerted here to
restore the broken balance of organic nature. And while looking for
their influence over one insect outbreak we stumbled upon at least two
others, less marked, perhaps incipient, but evident enough to express
themselves clearly in the changed food ratios of the birds.

"2. The comparisons made show plainly that the reflex effect of this
concentration on two or three unusually numerous insects was so widely
distributed over -the ordinary elements of their food that no especial
chance was given for the rise of new fluctuations among the species com-
monly eaten. That is to say, the abnormal pressure put upon the canker-
worm and vine-chafer was compensated by a general diminution of the
ratios of all the other elements, and not by a neglect of one or two alone.
If the latter had been the case, the criticism might easily have been made
that the birds, in helping to reduce one oscillation, were setting others on
foot.

"3. The fact that, with the exception of the indigo bird, the species
whose records in the orchard were compared with those made -else where
had eaten in the former situation as many caterpillars other than canker-
worms as usual, simply adding their canker-worm ratios to those of other
caterpillars, goes to show that these insects are favorites with a majority
of birds."

The Relations of Birds to Predaceous and Parasitic Insects.
The false assumption is often made that a bird is necessarily inimical to
man's interest whenever it destroys a parasitic or a predaceous insect.
Weed and Dearborn attack this assumption as follows:

"Suppose an ichneumon parasite is found in the stomach of a robin
or other bird: it may belong to any one of the following categories:

4 1. The primary parasite of an injurious insect.

"2. The secondary parasite of an injurious insect.

"3. The primary parasite of an insect feeding on a noxious plant.

INSECTS .IN RELATION TO OTHER ANIMALS 227

"4. The secondary parasite of an insect feeding on a noxious plant.

''5. The primary parasite of an insect feeding on a wild plant of no
economic value.

"6. The secondary parasite of an insect feeding on a wild plant of no
economic value.

"7. The primary parasite of a predaceous insect.

"8. The primary parasite of a spider or a spider's egg.

''This list might easily be extended still farther, and the assumption
that the parasite belongs to the first of these categories is unwarranted
by the facts and does violence to the probabilities of the case.

"A correct idea of the economic role of the feathered tribes may be
obtained only by a broader view of nature's methods, a view in which
we must ever keep before the mind's eye the fact that all the parts of
the organic world, from monad to man, are linked together in a thousand
ways, the net result being that unstable equilibrium commonly called
'the balance of nature."

This broader view was first elaborated by Professor Forbes, in his
masterly paper, "On Some Interactions of Organisms," the substance of
which is given below.

' l Evidently a species can not long maintain itself in numbers greater
than can find sufficient food, year after year. If it is a phytophagous
insect, for example, it will soon dwindle if it seriously lessens the numbers
of the plants upon which it feeds, either directly, by eating them up, or
indirectly, by so weakening them that they labor under a marked dis-
advantage in the struggle with other plants for foothold, air, light and
food. The interest of the insect is therefore identical with the interest of
the plant it feeds upon. Whatever injuriously affects the latter, equally
injures the former; and whatever favors the latter, equally favors the
former. This must, therefore, be regarded as the extreme normal limit
of the numbers of a phytophagous species, a limit such that its depre-
dations shall do no especial harm to the plants upon which it depends for
food, but shall remove only the excess of foliage or fruit, or else super-
fluous individuals which must perish otherwise, if not eaten, or, surviv-
ing, must injure their 'species by over-crowding. If the plant-feeder
multiply beyond the above limit, evidently the diminution of its food
supply will soon react to diminish its own numbers; a counter reaction
will then take place in favor of the plant, and so on through an oscillation
of indefinite continuance.

"On the other hand, the reduction of the phytophagous insect below
the normal number will evidently injure the food plant by preventing a

228 ENTOMOLOGY

i

reduction of its excess of growth or numbers, and will also set up an
oscillation like the preceding, except that the steps will be taken in re-
verse order.

"I next point out the fact that precisely the same reasoning applies
to predaceous and parasitic insects. Their interests also are identical
with the interests of the species they parasitize or prey upon. A diminu-
tion of their food reacts to decrease their own numbers. They are thus
vitally interested in confining their depredations to the excess of indi-
viduals produced, or to redundant or otherwise unessential structures.
It is only by a sort of unlucky accident that a destructive species really
injures the species preyed upon.

"The discussion has thus far affected only such organisms as are
confined to a single species. It remains to see how it applies to such as
have several sources of support open to them, such, for instance, as
feed indifferently upon several plants or upon a variety of animals, or
both. Let us take, first, the case of a predaceous beetle feeding upon a
variety of other insects, either indifferently, upon whatever species is
most numerous or most accessible, or preferably upon certain species,
resorting to others only in case of an insufficiency of its favorite food.

"It is at once evident that, taking the group of its food-insects as a
unit, the same reasoning applies as if it were restricted to a single species
for food; that is, it is interested in the maintenance of these food-species
at the highest number consistent with the general conditions of the
environment, interested to confine its own depredations to that sur-
plus of its food which would otherwise perish if not eaten interested,
therefore, in establishing a rate of reproduction for itself which will not
unduly lessen its food supply. Its interest in the numbers of each species
of the group it eats will evidently be the same as its interest in the group
as a whole, since the group as a whole can be kept at the highest number
possible only by keeping each species at the highest number possible. . . .

"This argument holds for birds as well as for insects, for animals of
all kinds, in fact, whether their food be mixed or simple, animal or vege-
table, or both. It also applies to parasitic plants. The ideal adjust-
ment is one in which the reproductive rate of each species should be so
exactly adapted to its food supply and to the various drains upon it that
the species preyed upon should normally produce an excess sufficient for
the species it supports. And this statement evidently applies through-
out the entire scale of being. Among all orders of plants and animals,
the ideal balance of Nature is one promotive of the highest good of all
the species. In this ideal state, towards which Nature seems continually

IXSECTS IN RELATION TO OTHER ANIMALS 22Q

striving, every food-producing species of plant or animal would grow and
multiply at a rate sufficient to furnish the required amount of food, and
every depredating species would reproduce at a rate no higher than just
sufficient to appropriate the food thus furnished. .

"Exact adjustment is doubtless never reached anywhere, even for a
single year. It is usually closely approached in primitive nature, but
the chances are practically infinite against its becoming really complete,
and mal-adjustment in some degree is therefore the general rule. All
species must oscillate more or less."

Professor Forbes then shows that oscillations are injurious to a species
and that the tendency of things is toward a healthy equilibrium. If the
rate of reproduction, as in a parasite for instance, is too small in relation
to the food supply, the species will eventually yield to its more prolific
competitors in the general struggle for existence. If, on the other hand,
its rate of multiplication is too high, the species will be at a disadvantage
in the search for food, as compared with better adjusted species, and must
again suffer. "The fact of survival is therefore usually sufficient evi-
dence of a fairly complete adjustment of the rate of reproduction to the
drains upon the species." . . . "We may be sure, therefore, that, as a
general rule, in the course of evolution, only those species have been able
to survive whose parasites, if any, were not prolific enough sensibly to
limit the numbers of their hosts for any length of time.

"We notice incidentally that it is thus made unlikely that an injuri-
ous species can be exterminated, can even be permanently lessened in
numbers, by a parasite strictly dependent upon it, a conclusion which
remarkably diminishes the economic role of parasitism. The same line
of argument will, of course, apply, with slight modifications, to any
animal, or even to any plant dependent upon any other animal or any
other plant for existence.

"It is a general truth, that those animals and plants are least likely
to oscillate widely which are preyed upon by the greatest number of
species, of the most varied habit. Then the occasional diminution of a
single enemy will not greatly affect them, as any consequent excess of
their own numbers will be largely cut down by their other enemies, and
especially as, in most cases, the backward oscillations of one set of ene-
mies will be neutralized by the forward oscillations of another set. But
by the operations of natural selection, most animals are compelled to
maintain a varied food habit, so that if one element fails, others may be
available. Thus each species preyed upon is likely to have a number of
enemies, which will assist each other in keeping it properly in check.

230 ENTOMOLOGY

"Against the uprising of inordinate numbers of insects, commonly
harmless but capable of becoming temporarily injurious, the most valu-
able and reliable protection is undoubtedly afforded by those predaceous
birds and insects which eat a mixed food, so that in the absence or diminu-
tion of any one element of their food, their own numbers are not seriously
affected. Resorting, then, to other food supplies, they are found ready,
on occasion, for immediate and overwhelming attack against any threat-
ening foe. Especially does the wonderful locomotive power of birds,
enabling them to escape scarcity in one region which might otherwise
decimate them, by simply passing to another more favorable one, with-
out the loss of a life, fit them, above all other animals and agencies, to
arrest disorder at the start, to head off aspiring and destructive rebel-
lion before it has had time fairly to make head. But we should not
therefrom derive the general, but false and mischievous notion, that the
indefinite multiplication of either birds or predaceous insects is good.
Too many of either is nearly or quite as harmful as too few.

'There is a general consent that primeval nature, as in the unin-
habited forest or the untilled plain, presents a settled harmony of inter-
action among organic groups which is in strong contrast with the many
serious mal-adjustments of plants and animals found in countries occu-
pied by man.

'To man, as to nature at large, the question of adjustment is of vast
importance, since the eminently destructive species are the widely oscil-
lating ones. Those insects which are well adjusted to their environ-
ments, organic and inorganic, are either harmless or inflict but moderate
injury (our ordinary crickets and grasshoppers are examples); while
those that are imperfectly adjusted, whose numbers are, therefore, sub-
ject to wide fluctuations, like the Colorado grasshopper, the chinch-bug
and the army worm, are the enemies which we have reason to dread.
Man should then especially address his efforts, first, to prevent any
unnecessary disturbance of the settled order of the life of his region
which will convert relatively stationary species into widely ocsillating
ones; second, to destroy or render stationary all the oscillating species
injurious to him; or, failing in this, to restrict their oscillations within the
narrowest limits possible.

u For example, remembering that every species oscillates to some
extent, and is held to relatively constant numbers by the joint action of
several restraining forces, we see that the removal or weakening of any
check or barrier is sufficient to widen and intensify this dangerous oscil-
lation; may even convert a perfectly harmless species into a frightful

INSECTS IN RELATION TO OTHER ANIMALS 231

pest. Witness the maple bark louse, which is so rare in natural forcM-
as scarcely ever to be seen, limited there as it is by its feeble locomotive
power and the scattered situation of the trees it infests. With the multi-
plication and concentration of its food in towns, it has increased enor-
mously, and, if it has not done the gravest injury, it is because the trees
attacked by it are of comparatively slight economic value, and because
it has finally reached new limits which hem it in once more.

" We are therefore sure that the destruction of any species of insectivo-
rous bird or predaceous insect is a thing to be done, if at all, only after
the fullest acquaintance with the facts. The natural presumptions are
nearly all in their favor. It is also certain 'that the species best worth
preserving are the mixed feeders and not those of narrowly restricted
dietary (parasites, for instance), that while the destruction of the latter
would cause injurious oscillations in the species affected by them, they
afford a very uncertain safeguard against the rise of such oscillations.
In fact, their undue increase would be finally as dangerous as their
diminution.

"Notwithstanding the strong presumption in favor of the natural
system, when we remember that the purposes of man and what, for con-
venience' sake, we may call the purposes of Nature do not fully harmo-
nize, we find it incredible that, acting intelligently, we should not be able
to modify existing arrangements to our advantage, especially since
much of the progress of the race is due to such modifications made in the
past. . . .

'But far the most important general conclusion we have reached is a
conviction of the general beneficence of nature, a profound respect for
the natural order, a belief that the part of wisdom is essentially that of
practical conservatism in dealing with the system of things by which we
are surrounded."

Efficiency of Protective Adaptations of Insects. Interesting
from a scientific point of view are the various adaptations by means of
which insects are protected more or less from their bird enemies. Color-
ational adaptations having been discussed in another chapter, there
remain for consideration (i) hairs, (2) stings, (3) odors, flavors and
irritants. Most of what follows is from an admirable paper by Dr. Judd,
whose data are based upon his examination of the stomach contents of
fifteen thousand birds.

Hairs. ; ' Excepting two species of cuckoos, no species of bird in the
eastern United States, so far as I am aware, makes a business of feeding
upon hairy caterpillars." Judd observed that Hyphantria cunea infest-

232 ENTOMOLOGY

ing a pear tree was not at all molested, in spite of the fact that the tree
was tenanted by three broods of birds at the time, namely, kingbirds,
orchard orioles and English sparrows. The hairy arctiid caterpillars,
however, are eaten by a few birds: the robin, bluebird, catbird, sparrow-
hawk, cuckoos and shrikes; and the spiny larvae of Vanessa antiopa by
cuckoos and the Baltimore oriole; while the hairy caterpillars of the gypsy
moth are known to be eaten in Massachusetts by no less than thirty-one
species of birds, notably cuckoos, Baltimore oriole, catbird, chickadee,
blue-jay, chipping sparrow, robin, vireos and the crow, these birds being
of no little assistance in the suppression of this pest. These are excep-
tional cases, however, and in general the hairiness of caterpillars appears
to be a highly effective protection against most birds.

Stings. Some birds (chewink, young ducks) are fatally affected by
eating honey bees. The blue-jays, however, will eat Bombus and Xylo-
copa, and flycatchers and swallows feed habitually upon stinging Hymen-
optera, particularly Scoliidae, while a great many birds eat Myrmicidae,
or stinging ants. The formic acid of ants does not protect them from
wholesale destruction by birds; Judd found three thousand ants in the
stomach of a flicker. . "Stingless ants pretend to sting but many birds
they do not deceive." The stinging caterpillar of Automeris io is occa-
sionally eaten by the yellow-billed cuckoo. Aside from these exceptions,
however, the stings of insects are an extremely efficient means of defence.

Odors, Flavors and Irritants. The malodorous Heteroptera in
general are food for most birds; Lygus, Reduviidae and Pentatomidae are
eaten by song sparrows, and Euschistus by blackbirds and crows. The
odors of Heteroptera are by no means universally protective.

Among Coleoptera, the showy, ill-scented or ill-flavored Coccinellidae
are eaten by but very few birds the flycatchers and swallows and are
refused by caged blue-jays and song sparrows even when these birds are
hungry. Of Chrysomelidae, the Colorado potato beetle is refused by the
catbird, blue-jay and song sparrow, and Diabrotica is not often eaten,
except by catbirds and thrushes. "The smaller Carabidae, whether
stinking or not, are eaten by practically all land birds." Crows, black-
birds and jays eagerly swallow Calosoma scrutator, and the first two birds
are especially fond of Harpalus caliginosus and H. pennsylvanicus, and
feed Galerita to their young. "A score of smaller Carabidae and Chryso-
melidae, metallic and conspicuously colored, are habitually eaten by
birds that have an abundance of other insect food to pick from."

The stenches of Lampyridae appear to be more effective than those of
Carabidae. Telephorus is occasionally eaten, but Photinus rarely if at

INSECTS IN RELATION TO OTHER ANIMALS 233

all. Chanliognatlms is not eaten by many birds (though flycatchers and
swallows select this insect) and the genus is regarded unfavorably by
caged catbirds and blue-jays.

In regard to other insects, Judd finds that Epicauta. with its irritant
fluid, is immune from all but the kingbird; Cyllene seldom occurs in the
stomachs of birds; May flies and caddis flies, however, are terribly perse-
cuted, but swiftly flying Diptera and Oclonata are highly immune.

From such facts as these, Judd properly infers, ''not cases of protec-
tion and non-protection, but cases of greater and lesser efficiency of
protective devices."

CHAPTER IX

TRANSMISSION OF DISEASES BY INSECTS

It is now known that several kinds of insects are of vital importance
to man as agents in the transmission of certain diseases. This recently
demonstrated role of insects now commands universal attention.

MALARIA

So far as is known, malaria is transmissible only through the agency
of mosquitoes.

The malaria "germ," discovered in 1880 by the French army surgeon
Laveran, may be found as a pale, amoeboid organism (Plasmodium,
Fig. 270) in the red blood corpuscles of persons afflicted with the disease.
This organism (schizont, 2) grows at the expense of the haemoglobin of
the corpuscle (3-5) and its growth is accompanied by an increasing
deposit of black granules (melanin), which are doubtless excretory in
their nature. At length, the amcebula divides into many spores (mero-
zoites, 6), which by the disintegration of the corpuscle are set free in the
plasma of the blood. Here many if not most of the spores, and the
pigment granules as well, are attacked and absorbed by leucocytes, or
white blood corpuscles, while some of the spores may invade healthy red
corpuscles and develop as before. The period of sporulation, as Golgi
found, is coincident with that of the "chill" experienced by the patient;
and quinine is most effective when administered just before the sporula-
tion period. The destruction of red blood corpuscles explains the pallid,
or ancemic, condition which is characteristic of malarial patients. In
three or four days the number of red corpuscles may be reduced from
5,000,000 per cubic millimeter the normal number to 3,000,000; and
in three or four weeks of intermittent fever, even to 1,000,000.

Three' types of malaria are recognized: (i) the tertian, in which the
paroxysm recurs every two days; (2) the quartan, in which it happens
every third day; and (3) the aestivo-autumnal type (Fig. 270). These
three kinds are by some investigators thought to be due to different
species of parasites; and when, as often happens, the malarial chill occurs
every jiay, this is attributed to two sets of tertian amcebulae, sporulating
on alternate days.

234

FIG. 270. Life history of malaria parasite, Plasmodium pracox. i, sporozoite, introduced
by mosquito into human blood; the sporozoite becomes a schizont. 2, young schizont,
which enters a red blood corpuscle, j, young schizont in a red blood corpuscle. 4, full-
grown schizont, containing numerous granules of melanin. 5, nuclear division preparatory
to sporulation. 6, spores, or merozoites, derived from a single mother-cell. 7, young macro-
gamete (female), derived from a merozoite and situated in a red blood corpuscle. 70, young
microgametocyte (male) derived from a merozoite. 8, full-grown macrogamete. Sa, full-
grown microgametocyte. In stages 8 and Sa the parasite is taken into the stomach of a
mosquito; or else remains in the human blood, g, mature macrogamete, capable of fertiliza-
tion; the round black extruded object may probably be termed a "polar body." ga, mature
microgametocyte, preparatory to forming microgametes. gb, resting cell, bearing six flagellate
microgametes (male). 10, fertilization of a macrogamete by a motile microgamete. The
macrogamete next becomes an ookinete. //, ookinete, or wandering cell, which penetrates
into the wall of the stomach of the mosquito. 12, ookinete in the outer region of the wall of
the stomach, i. e., next to the body cavity. 13, young oocyst, derived from the ookinete.
14, oocyst, containing sporoblasts, which are to develop into sporozoites. 75, older oocyst.
16, mature oocyst, containing sporozoites, which are liberated into the body cavity of the
mosquito and carried along in the blood of the insect. 17, transverse section of salivary gland
of an Anophc'cs mosquito, showing sporozoites of the malaria parasite in the gland cells
surrounding the central canal.

1-6 illustrate schizogony (asexual production of spores) ; 7-16, sporogony (sexual production
of spores) .

After GRASSI and LEUCKART, by permission of Dr. Carl Chun.

235

236 ENTOMOLOGY

After several successive asexual generations, there are produced
merozoites which develop no longer into schizonts but into sexual
forms, or gametes. These occur in red blood corpuscles either as macro-
gametes (female, 7, 5) or as microgametocytes (male, 70, 8d), in which
forms the parasite is introduced into the stomach of a mosquito which
has been feeding upon the blood of a malarial patient. The macro-
gamete now leaves its blood corpuscle and becomes spherical (p), as does
also the microgametocyte (pa); but the latter puts forth a definite
number (six, in P. pracox, gb) of flagella, or micro gametes, which separate
off as motile male bodies, capable of fertilizing the macrogametes. A
microgamete penetrates a maciogamete (16) and the nucleus of the one
unites with that of the other. The fertilized macrogamete now becomes
a migrating cell, or ookinete (u), which penetrates almost through the
wall of the stomach of the mosquito (12) and then becomes a resting cell,
or cyst. This ob'cyst (13) grows rapidly and its contents develop, by
direct nuclear division, into sporoblasts (14, 15), which differentiate into
spindle-shaped sporozoites (16, 17). The sporozoites are liberated into
the body cavity of the mosquito, carried in the blood to the salivary
glands (as well as elsewhere) and thence along the hypopharynx into the
body of a human being, bird or other animal attacked by the insect.

The role of the mosquito as the intermediary host of malarial organ-
isms was discovered by Manson and Ross and confirmed by Koch, Stern-
berg and others. It has been found repeatedly that certain mosquitoes
(Anopheles) after feeding on the blood of a malarial patient can transmit
the disease by means of their "bites" to healthy persons. Thus, Anoph-
eles mosquitoes were fed on the blood of malarial subjects in Rome
and then sent to London, where a son of Dr. Manson allowed himself
to be bitten by the insects. Though previously free from the malarial
organism, he contracted a well-marked infection as the result of the
inoculation.

Furthermore, it is highly probable that malaria cannot be trans-
mitted to man except through the agency of the mosquito. This appears
from the oft-cited experiment of Doctors Sambon and Low on the Roman
Campagna, a place notorious for malaria. There the experimenters
lived during the malarial season of 1900, freely exposed to the emanations
of the marsh and taking no precautions except to screen their house
carefully against mosquitoes and to retire indoors before the insects
appeared in the evening. Simply by excluding Anopheles mosquitoes,
with which the Campagna swarmed, these investigators remained per-
fectly immune from the malaria which was ravaging the vicinity.

TRANSMISSION OF DISEASES BY INSECTS 237

In a later experiment on the island of Formosa, one company of
Japanese soldiers was protected from mosquitoes and suffered no malaria,
while a second and unprotected company contracted the disease.

The evident preventive measures to be taken against malaria are
(i) the avoidance of mosquito bites, by means of screens and washes of
eucalyptus oil, camphor, oil of pennyroyal, oil of tar, etc., applied to
exposed parts of the body; (2) the isolation of malarial patients from
mosquitoes, in order to prevent infection; (3) the destruction of mosqui-
toes in their breeding places, especially by the use of kerosene and by
drainage. During unavoidable exposure in malarious regions, quinine
should be-taken in doses of six to ten grains during the day at intervals of
four or five days (Sternberg).

Culex and Anopheles. The mosquitoes of North America number
one hundred and twenty-five known species. Of these only the genus
Anopheles transmits malaria to man, though in India, Ross found that
Culex transmits a form of malaria to sparrows. These two common
genera are easily distinguishable. In Culex the wings are clear; in
Anopheles they are spotted with brown. In Culex when resting, the axis
of the body forms a curved line, the insect presenting a hump-backed
appearance; in Anopheles the axis forms a straight line. Culex has short
maxillary palpi, while in Anopheles they are almost as long as the probos-
cis. The note of the female Anopheles is several tones lower than that of
Culex, and only the female is bloodthirsty, by the way. As regards eggs,
larvae and pupae, the two genera differ greatly. The eggs of Culex are
laid in a mass and those of Anopheles singly; the larvae of Culex hang
from the surface film of a pool. at an angle of about forty-five degrees,
while those of A nopheles are almost parallel with the surface of the water
in which they live.

The bite of an Anopheles is not necessarily injurious, of course, unless
the insect has had recent access to a malarious person. Anopheles may
be present where there is no malaria. On the other hand, it has been
found impossible to prove that malaria exists where there are no A noph-
eles mosquitoes. Finally, fevers are sometimes diagnosed as malarial
which are not so.

Possibly the malarial parasite can complete its cycle of development
in othei animals than man. It is also possible that originally the mala-
rial organism was derived by mosquitoes from the stems or other parts of
aquatic plants, and that its effects on man are incidental phenomena.

238 ENTOMOLOGY

YELLOW FEVER

From 1793 to 1900 there occurred in the United States not less than
half a million cases of yellow fever and one hundred thousand deaths
from the disease. New Orleans suffered the worst with more than forty-
one thousand deaths, followed by Philadelphia with ten thousand and
Memphis with almost eight thousand ; while Charleston, New York City
and Norfolk, Virginia, lost together more than ten thousand lives.

The enormous financial loss from all the epidemics of yellow fever is
beyond exact computation; the epidemic of 1878 cost New Orleans more
than ten million dollars.

Yellow fever is now within human control; with no thanks to those
who at first violently opposed the theory, and later denied the fact, of
its transmission by mosquitoes.

Until 1901 yellow fever was fought energetically, but fought in the
dark. An immense amount of energy was misdirected and millions of
dollars wasted in the fight. On the supposition that bacteria were the
cause of the disease, methods of quarantine, burning and fumigation were
employed that destroyed an enormous amount of property, including
valuable cargoes, and paralyzed the business and social activities of
great cities.

Official accounts of yellow fever published before 1900 often describe
the disease as due to some insidious poison borne by the air and intro-
duced into the human body probably through the respiratory system.
It was observed that the disease was often conveyed down the wind, that
it was not carried far from the nearest focus of infection, that infection
was less liable to occur in daylight than by night, and that cases arose on
shore when the only source of infection was a ship that had not yet
touched the land. These facts and many others which formerly involved
the disease in mystery, are now quite intelligible in the light of the mos-
quito-theory of transmission.

Finlay's Work. The pioneer work leading toward the control of
yellow fever was done by Dr. Charles J. Finlay, of Havana, Cuba, who
not only advocated the mosquito-theory strongly for many years, but
also inoculated by means of mosquitoes ninety human subjects, some
of whom came down with what he believed to be a mild form of yellow
fever. His valuable work prepared the way for the brilliant investi-
gations of Major Reed and his associates.

United States Yellow Fever Commission. Major Walter Reed
was president of the board of medical officers sent to Cuba in June, 1900,

TRANSMISSION OF DISEASES BY INSECTS 239

to study the acute infectious diseases of the island; his associates were
James Carroll, Jesse W. Lazear and A. Agramonte.

At that time Sanarelli's theory as to the bacillary causation of yellow
fever was in favor, though Reed and Carroll had already shown that the
bacillus of Sanarelli bore no special relation to the disease. After further
investigations on this subject in Cuba, with negative results, the commis-
sion "concluded to test the theory of Finlay," in Dr. Reed's words.
For this purpose General Leonard Wood, the military governor of Cuba,
gave permission for experiments on human beings and granted a liberal
sum of money for the reward of volunteer subjects.

The commission succeeded in demonstrating how yellow fever is
transmitted; after that the methods of prevention to be employed were
evident.

The experiments, planned and directed by Major Reed, are models
of their kind. All possible sources of error were excluded; hence there
was no uncertainty in the interpretation of the results, the accuracy of
which has been confirmed by subsequent commissions and by many
independent investigators.

In the value of his services Major Walter Reed ranks among the
greatest benefactors of mankind. Before his death, which occurred in
1902, he received great honors for his brilliant achievements.

Experiments in Cuba. For experimental purposes Major Reed
established a camp about four miles from Havana. To prevent the
introduction of the fever from the outside the inmates of the camp were
rigidly quarantined; non-immunes were confined to the camp or, if re-
leased, not allowed to return. In order that the study of yellow fever
might not be complicated by the presence of any other disease, a com-
plete record was kept of the health of every subject; furthermore, ample
time was allowed for any possible development of the disease within the
camp before the experiments were begun. In short, the precautions
taken were so thorough that yellow fever never appeared in the camp
except at the will of the experimenters.

Harmlessness of Fomites. In a specially constructed building,
which was screened against mosquitoes and purposely ill-ventilated,
volunteers slept for twenty nights with bedding and clothing that had
been contaminated by yellow fever patients, and tried in every other way
to contract the disease, if possible, from the fomites, or belongings, of
fever subjects; yet the health of these volunteers remained unimpaired;
though they were not immunes, for some of them were subsequently
infected artificially by means of mosquitoes.

240 ENTOMOLOGY

Transmission by Transfusion. It was found that the disease
could be conveyed to non-immunes by the subcutaneous injection of
blood taken from the veins of patients during the first three days of the
disease.

Experiments with Mosquitos. These experiments were made at
a time of the year when there was the least chance of acquiring the dis-
ease naturally. The mosquitoes used were bred from the egg and kept
active by being maintained at a summer temperature. From time to
time some of them were taken away to a yellow fever hospital, fed on the
blood of patients and applied to non-immunes in the camp at varying
intervals from the time of feeding. The occupants of the camp were, of
course, protected carefully from accidental mosquito bites. When a
subject came down with yellow fever as the result of an experimental
inoculation he was at once removed from the camp to a yellow fever
hospital.

In a mosquito-proof building a single room was divided into two
compartments simply by means of a partition of wire netting. On one
side of the screen infected mosquitoes were liberated; and a brave non-
immune, who had been in quarantine for thirty-two days, entered the
compartment, allowed himself to be bitten several times, and contracted
the disease. In the opposite compartment, free from mosquitoes, non-
immunes slept with perfect safety; and the other room became harmless
as soon as the mosquitoes were removed.

In another experiment the subject acquired the disease by thrusting
his arm into a jar of infected mosquitoes. Eighteen non-immunes were
inoculated, ten of them successfully. It was demonstrated that yellow
fever is transmitted by the. bite of a mosquito, and in no other way
except by the artificial injection of diseased blood. The mosquito can
obtain infected blood from a patient during only the first three days of
his disease; in other words, the patient is no longer a menace to other
persons after three days from the time when he comes down with yellow
fever, which is from three to six days after the bite.

After biting a patient the mosquito cannot convey the infection until
at least twelve days have elapsed; thereafter it can transmit the disease
for certainly six weeks and possibly eight weeks.

Dr. James Carroll allowed himself to be bitten by an infected mos-
quito and consequently suffered a severe attack of yellow fever. He
recovered from this, but was left with an affection of the heart from
which he died in 1907.

Dr. Lazear failed to acquire the disease artificially, early in the course

TRANSMISSION OF DISEASES BY INSECTS 241

of the experiments; but a little later, while visiting yellow fever patients
in a hospital, was bitten by a mosquito which he deliberately allowed to
remain on his hand. Five days later he came down with yellow fever,
which caused his death. His life was a sacrifice for the benefit of the
human race.

Yellow Fever Mosquito. The mosquito that transmits this fever
is Aedcs calopus (Stegomyia fasciata) and no other species is as yet known
to be concerned in the disease. A. calopus is limited to warm regions;
at a temperature less than 68 F. the eggs do not hatch, and below 62 F.
the female does not bite (Reed). The dependence of the insect upon
warmth for its development explains the cessation of the disease in New
Orleans in December, with a mean temperature of 55.3 F. and in cities
farther north when frost comes. In Cuba and Brazil the fever has
occurred every month in the year.

Causes of Yellow Fever. The specific cause of yellow fever has as
yet eluded detection and is regarded by many investigators as being
ultra-microscopic. The U. S. Commission produced the disease by the
injection of blood serum that had been passed through a bacteria-proof
filter. Blood from a subject in whom the disease had been produced by
transfusion was capable' of infecting a third person.

The weight of evidence indicates that the unknown cause of yellow
fever is an organism rather than a toxin.

Control of Yellow Fever. The preventive measures based upon
the facts learned by the U. S. Army Commission were wonderfully suc-
cessful. In February, 1901, Major W. C. Gorgas began a campaign to
eradicate the disease in Havana. His efforts were directed against
mosquitoes. Every case of fever had to be reported promptly to the
authorities. Then the patient was isolated and all the rooms in the
building and in neighboring houses fumigated and the doors and windows
screened. Standing water in which mosquitoes might develop was
drained or treated with petroleum and water tanks and barrels were
screened.

In September, 1901, the last case of yellow fever arose in Havana,
where the disease had prevailed for 150 years, with an annual mortality
of 500 to 1600 or more. Cases are now and then brought into Havana
from Mexico, but are treated under screens in the regular hospitals with
impunity.

Yellow Fever in New Orleans. In 1905 the last epidemic of yel-
low fever occurred in New Orleans. It might have been checked at its
inception had not the authorities adopted a policy of secrecy in regard

242 ENTOMOLOGY

to the presence of the disease. The city was freed from the fever before
frost came, by the same methods that had proved successful in Cuba;
but not without organized work of the most strenuous kind on the part
of the citizens, under the direction of the U. S. Public Health and Marine-
Hospital Service. At present the yellow fever mosquito is said to be a
rarity in Louisiana owing to the vigorous measures enforced in its sup-
pression throughout the state.

Fever in the Canal Zone. The Panama Canal zone was formerly
one of the most unhealthful places on earth, chiefly on account of the
prevalence of malaria and yellow fever. When the United States ac-
quired the zone in 1904 it was realized that the first step toward building
the great canal was to protect the health of all those immediately con-
cerned in the undertaking, and the sanitation of the isthmus was placed
in charge of one eminently qualified for the work, Colonel W. C. Gorgas.

He adapted the methods he had used in Cuba to the conditions
existing on the isthmus, with the result that every year the death rate
decreased until in 1908 it became, among eight thousand white Americans
living there, 9.72 per thousand, "a rate no higher than for a similar popu-
lation in the healthiest localities in the United States, and much lower
than that for most parts of the country." The Sanitary Department
has succeeded in driving yellow fever from the isthmus and in checking
malaria and other diseases to such a degree that the canal zone is no
longer an unhealthful place.

TYPHOID FEVER

The specific cause of typhoid fever is Bacillus typhosus. In the
human body this bacillus occurs chiefly in the intestines; but also in the
urinary bladder and usually in the blood of infected persons.

The excreta of typhoid subjects contain the virulent bacilli; and some
persons, even after recovery, continue to be "chronic carriers" of the
disease for many years.

Transmission. The typhoid bacillus is introduced into the human
system by eating or drinking. Most epidemics are due to infected
water and many to milk; occasionally the disease is acquired from raw
vegetables or from oysters contaminated with sewage. Often the bacillus
is conveyed to food by human hands and possibly it is sometimes carried
by dust, cockroaches or ants; but there is no doubt that the disease is
transmitted by certain flies, particularly the true house fly, Musca domes-
tica, which is by far the commonest fly found generally in houses, and
becomes a serious menace to health during epidemics of typhoid fever.

TRANSMISSION OF DISEASES BY INSECTS 243

The house fly is well adapted by its structure and habits to carry
bacteria. The adults often feed on substances contaminated with
typhoid or other bacteria and these infected substances cling readily
to the hairs of the insect, especially those of the feet, and to the pro-
boscis. The larvae develop chiefly in horse manure, but also in other
kinds of excreta, some of which may contain virulent typhoid bacilli.

Transmission by Flies. During the Spanish-American war ty-
phoid fever occurred in every American regiment and raged in many of
the concentration camps, in consequence of which a special commission
was appointed to investigate the origin and spread of the disease in the
army. A report by one of the members of the commission, Doctor
Vaughan, presents the following conclusions:

"a. Flies swarmed over infected fecal matter in the pits and then
visited and fed upon the food prepared for the soldiers at the mess tents.
In some instances where lime had recently been sprinkled over the con-
tents of the pits, flies with their feet whitened with lime were seen walk-
ing over the food.

"b. Officers whose mess tents were protected by means of screens
suffered proportionally less from typhoid than did those whose tents
were not so protected.

"c. Typhoid fever gradually disappeared in the fall of 1898, with
the approach of cold weather, and the consequent disabling of the fly.

"It is possible for the fly to carry the typhoid bacillus in two ways.
In the first place, fecal matter containing the typhoid germ may adhere
to the fly and be mechanically transported. In the second place, it is
possible that the typhoid bacillus may be carried in the digestive organs
of the fly and may be* deposited with its excrement."

Similar conclusions in regard to the agency of flies in the spread of
enteric fever among troops have been reached also by investigators in
Bermuda, South Africa and India.

Firth and Horrocks fed house flies on material contaminated with
Bacillus typhosus and then obtained cultures of the bacillus from objects
to which the flies had access. In another experiment they got cultures
from the heads, bodies, wings and legs of such flies. Other investigators
have obtained Bacillus typhosus from flies captured in rooms occupied
by typhoid cases.

Faichnie caught flies in a place where there was an outbreak of ty-
phoid fever, held them on a sterilized needle and passed them through a
flame until legs and wings were scorched; after which he obtained the

244 ENTOMOLOGY

typhoid bacillus from the mashed bodies of the flies, the bacilli having
been present in the alimentary tract, without doubt.

Faichnie also obtained cultures of Bacillus typhosus from the intes-
tines of flies which had developed from larvae fed on feces containing the
bacillus.

Jordan states that the bacilli survive the passage of the alimentary
canal of the fly.

Ficker recovered typhoid bacilli from flies twenty-three days after
they had been infected.

In fact, a great amount of evidence has accumulated proving that
flies transmit not only the bacilli of typhoid fever, but many other bac-
teria, and often in enormous numbers. For example, Esten and Mason
in their study of the sources of bacteria in milk, collected and examined
flies from stables, pig-pens, houses and other places, and found an average
of 1,222,570 bacteria per fly; the majority of these being objectionable
kinds of bacteria.

Musca domestica. A single female of the common house fly lays
in all some six hundred eggs. In midsummer, in Washington, D. C.,
the eggs hatch in about eight hours; the larval period is from four to
five days and the pupal period five days, making the cycle about ten
days in length. In cooler parts of the season the cycle requires more
time and in warm climates it may be as short as eight days. The number
of generations in Washington is probably not more than nine (Howard).

Control. One of the best baits for flies in houses is formalin, which
is poisonous to flies but harmless to man. This is prepared by diluting
formaldehyde with five or six times as much water .and exposing it in
shallow dishes, the addition of a little sugar or milk making the solution
more attractive to flies, which drink it and quickly die. Pyrethrum is
effective against flies, but only when it is pure and has been kept from
exposure to the air. Pyrethrum, the chief basis of all the common
insect powders, is applied by being puffed through a bellows or by being
burned. The powder may be moistened and shaped into cones which
when lighted at the top burn slowly and give off fumes that are suffocat-
ing to insects.

Dr. Howard estimates that more than ten million dollars are spent
every year in screening houses in the United States. Another enormous
sum is spent for fly papers and fly traps. The efficient way to deal with
the fly problem, however, is to prevent the insects from breeding. Ex-
crementitious substances should be enclosed in such a way as to prevent
the access of flies, or should be treated in a way to kill the larvae therein;

TRANSMISSION OF DISEASES BY INSECTS 245

one of the simplest methods of treating stable manure being to spread it
out to dry, since the maggots cannot develop without moisture.

For detailed information on everything of importance relating to the
house fly, and particularly on the mitigation of the fly-nuisance by con-
certed action in communities, Dr. Howard's admirable book on the house
fly should be consulted.

PLAGUE.

In the ancient history of Europe epidemics of plague occupy a large
place. In recent years this pestilence has thrived in China and India,
and following an outbreak in 1894 in Hong Kong, the plague reached
the western hemisphere for the first time, appearing in Brazil, Argentina
and other South American countries, in Mexico and San Francisco.

The cause of plague is Bacillus pestis, an organism abundant in the
secretions and excretions of plague-stricken animals.

Three varieties of the disease are distinguished as follows:

(1) the bubonic, in which the bacilli cause enlargements of lymphatic
glands;

(2) the scpticcemic, characterized by the presence of large numbers
of bacilli in the blood and highly virulent ;

(3) the pneumonic, in which the respiratory organs are affected, the
sputum showing the bacilli in enoimous numbers; this form, relatively
rare, is the most fatal.

Transmission. Plague is primarily a disease of rats, an epidemic
of plague in these animals having often been observed to precede as
well as accompany an epidemic among human beings. The disease
affects also mice, cats, dogs, calves, sheep, pigs, ducks, geese and many
other animals.

Though rats and other of the lower animals may contract the septi-
caemic type of the disease from feeding on parts of animals killed by
plague or on cultures of Bacillus pestis, the disease is commonly trans-
mitted among rats neither by contact nor through the atmosphere, but
by means of fleas. Healthy rats in association with diseased rats do
not become infected as long as fleas are excluded; but a transfer of fleas
from the latter to the former starts the disease. By various experiments
the Indian Plague Commission demonstrated the important part played
by rat-fleas in the transmission of plague. Zirolia found that the bacilli
even multiply in the mid-intestine of the flea, retaining their virulence for
a week or more.

The weight of evidence, both observational and experimental, shows

246 ENTOMOLOGY

that plague is transmitted from rats to man by several species of fleas
and also by bedbugs. Verjbitski, whose experiments on this subject
were particularly precise and thorough, found that plague can be con-
veyed by the bites of these insects and that the opening made by the bite
affords entrance to plague bacilli when the bodies of the insects are
crushed or when the infected feces are introduced by the rubbing or
scratching of the wound.

The species of rat-flea most common in the orient is the cosmopolitan
"plague flea," Lcemopsylla cheopus.

In the United Sates the most common rat flea is Ceratophyllus fascia-
tus. The common cat and dog flea, Ctenocephalus canis, affects rats as
does also the human flea, Culex irritans; and all these species are known
to bite man.

Plague in San Francisco. Plague, long dreaded in American sea-
ports, finally entered San Francisco in 1900, killed 114 persons in the
next four years, became dormant and broke forth again, with violence,
in 1907. The city, just beginning to recover from the great fire of the
year before, was in a frightful sanitary condition and most of the popula-
tion, engaged in the work of reconstruction, paid little attention to the
deaths from plague and at first gave little aid toward the suppression
of the disease. As may be imagined, the campaign against the disease
undertaken by the U. S. Public Health and Marine-Hospital Service was
carried on in the face of great odds. It was, however, conducted most
efficiently and successfully under the command of Dr. Rupert Blue
(now Surgeon-General), who wisely attacked the disease by attacking
the rat population.

The labor involved in starving out the rats, trapping or poisoning
them, and making buildings rat-proof by the use of concrete or sheet
iron, was immense; but the undertaking was nevertheless carried to a
successful conclusion. More than one million rats were killed and the
disease was checked.

In California plague affects ground squirrels, which doubtless con-
tract the disease from the rats that use the runways of the squirrels in the
fields.

TRYPANOSOMIASES

Some of the diseases known as trypanosomiases are among the dead-
liest that affect man and other vertebrates, and pathogenic trypanosomes
the organisms causing these diseases have received an immense
amount of study during the last fifteen years.

TRANSMISSION OF DISEASES BY INSECTS

247

m

Trypanosomes. The organisms under consideration are flagellate
protozoans. A typical trypanosome, for example, T. lewisi (Fig. 271)
of the rat, is essentially an elongated cell, tapering at each end, serpen-
tine in form and with no definite cell-wall. A round or oval nucleus is
present, also a peculiar chromatin body situated often near the posterior
end of the cell and termed the blepharoplast. Along one side of the cell
is a delicate protoplasmic contractile membrane, the undulating mem-
brane, along the edge of which is a marginal cord, which arises by growth
from the blepharoplast and is continued beyond
the anterior end of the cell as a vibratile flagellum.

Asexual reproduction is by means of a longi-
tudinal division of the cell body, preceded by divi-
sion of the flagellum, blepharoplast and nucleus,
the nucleus dividing amitotically. In regard to the
existence of sexual stages, or gametes, the results
of investigators seem to be inconclusive as yet.

In a film of fresh blood under the microscope,
any active trypanosomes in the field of view attract
attention as centers of commotion among the red
blood corpuscles,, which are pushed aside by the
lashing, twisting and other movements of the try-
panosomes.

The nutrition is by means of osmosis. Try-
panosomes have not been seen to attack erythro-
cytes, but according to MacNeal and Novy haemo-
globin is useful if not indispensable to them.

All five classes of vertebrates serve as hosts for
trypanosomes, of which more than seventy species
have received names. Most of these species are
carried from one vertebrate host to another by
means as yet unknown, but about twenty per cent, are known or suspected
to be transmitted by an intermediate invertebrate host. Thus trypano-
somes of frogs are conveyed by leeches; pigeons are infected by mosqui-
toes, rats by sucking lice and fleas, and many mammals through the
agency of blood-sucking flies of the genus Glossina, and probably also
by Stomoxys and certain Tabanidae.

Tsetse Flies. The name tsetse fly, originally limited to Glossina
morsitans (Muscidas) is now used for any of the eight known species of
the genus. These flies are a little larger than the common house fly
(Musca domestica). Their wings, in the resting position, overlap exactly

u

FIG. 271. Trypanoso-
ma lewisi. b, blepharo-
plast; /, flagellum; m,
marginal cord; u, nucleus;
w, undulating membrane.
Greatly magnified.

248

ENTOMOLOGY

(Fig. 272) instead of being separated at the tips. The proboscis projects
forward, and is stout, owing to the ensheathing palpi; the base of the
labium forms a prominent bulb. These are the more conspicuous char-
acters that serve to distinguish tsetse flies from other blood-sucking flies
with which they might be confused.

The mode of reproduction as described by Brauer is similar to that
of the group of parasitic flies known as Pupipara. The fly produces a
full-grown larva, which at once creeps to some resting place and forms a
black puparium.

Tsetse flies frequent hot, humid regions, near bodies of water, and
are restricted to shaded situations, never occurring on the open plains.
Both sexes are bloodthirsty but bite only during the daytime as a rule;
though they may bite at night when the moonlight is bright. Travelers

take advantage of the habits of the fly to
journey by night; spending the day in an open
uninfested place.

Nagana . The colonization of South Africa
has been greatly retarded by nagana, a disease
invariably fatal to the horse, donkey and dog,
and usually fatal to cattle, but not affecting
man. Livingstone and other explorers in re-
gions where nagana is prevalent record their
having been bitten by tsetse flies thousands
of times with no result other than a slight
irritation.

Bruce was the first to prove the identity of
nagana and tsetse-fly disease and to demon-
strate the role of the fly in the transmission of the disease. His investi-
gations, begun in Zululand in 1894, are of fundamental importance and
have given an immense]stimulus to the study of trypanosomes.

After finding that no bacteria were concerned in nagana, Bruce dis-
covered trypanosomes in the blood of cattle affected with the disease.
He inoculated their blood into healthy horses and dogs and in a few days
the blood of these animals was teeming with trypanosomes. Then he
took healthy animals from the mountain on which he had located down
into the ''fly country"; there they contracted the tsetse-fly disease and
showed in their blood trypanosomes indistinguishable from those of
nagana.

Horses taken into the fly country but not allowed to eat or drink
there, took the disease ^'furthermore, supplies of grass and water brought

FIG. 272. Tsetse
nwrsitans.

fly, Glossina

TRANSMISSION OF DISEASES BY INSECTS

249

from the fly country and fed to healthy horses failed to convey the
disease.

Then the influence of the fly was tested. Tsetse flies caught in the
lowland, carried to the mountain and placed at once on healthy animals
gave rise to the disease; but the flies never retained the power of infecting
a healthy animal for more than forty-eight hours after feeding upon a
sick animal. Thus wild flies, kept without food for three days and then
fed on a healthy dog, never gave rise to the disease. The fly alone trans-
mitted the disease; and this by means of trypanosomes adhering to the
proboscis either inside or out. Bruce found these organisms in the diges-
tive tract also, but with no change in their form.

He discovered further that buffaloes, antelopes and many other wild
animals carried the parasite in their blood, and was
able by injecting this blood to transmit the disease
to healthy domesticated animals. The parasites
were never numerous in the blood of their wild hosts,
however, and the latter seemed to be unaffected by
their presence. The "big game" of Africa serves,
generally speaking, as a reservoir for supplies of
trypanosomes.

The species of parasite that Bruce studied is
named Trypanosoma brucei (Fig. 273). The flies
concerned are Glossina morsitans, G. pallidipes and
G. fusca, particularly the first two, the distribution
of which coincides with that of nagana.

No certain remedies for the disease are yet
known. Human serum injected into infected ani-
mals causes the trypanosomes to disappear, at least
temporarily; but this fact is of more scientific in-
terest than practical importance. The precaution of traveling by night
is often adopted. Creolin and some other substances rubbed on animals
serve to repel the flies, and the smoke of encampments drives them away.
The protection of horses by means of screens is of course effective.

Human Trypanosomiasis. Sleeping sickness is most prevalent in
the Congo basin, whence it has spread rapidly in equatorial Africa, where
it kills about fifty thousand natives every year. The reported cases of
recovery are so extremely rare that the mortality is placed at one hundred
per cent.

In the first stage of the disease, marked by the appearance of trypano-
somes in the blood, negroes show no symptoms as a rule, though whites

FIG. 273. Trypano-
soma brucei. Greatly
magnified.

350 ENTOMOLOGY

are subject to fever. The symptoms may appear as early as four weeks
after infection or as late as seven years.

In the second stage trypanosomes appear in the cerebro-spinal fluid
and in large numbers in the lymphatic glands, those of the neck, axillse
and groins becoming enlarged. There is tremor of the tongue and
hands, drowsiness, emaciation and mental degeneration. The drowsi-
ness passes into periods of lethargy which become gradually stronger
until the patient becomes comatose and dies. Some victims do not
sleep excessively, but are lethargic, and "profoundly indifferent to all
going on around them."

There is some disagreement among authors as to the precise effects of
trypanosomes on human tissues and organs, but the evidence indicates
at least that trypanosomes produce a toxin which sets up irritations of
the lymphatic glands in general and those of the brain in particular.
Many of the symptoms of trypanosomiasis are traceable primarily to
inflammation of the lymphatics of the nervous system.

The specific cause of sleeping sickness is T. gambiense, discovered in
1901 by Forde and named by Button. Two eminent English investi-
gators of sleeping sickness, Button and Tullock, sacrificed their lives to
the disease they were studying.

As the result of the labors of many investigators, human trypano-
somiasis is now well understood. Bruce and Nabarro demonstrated by
means of inoculation experiments with monkeys that T. gambiense is
transmitted chiefly, if not solely, by a tsetse fly, Glossina palpalis.
They and Greig showed that the distribution of the disease in Uganda
coincided with that of the fly. In some regions where the fly is present
the disease is unknown; which means simply that cases of the disease
have not yet been introduced.

Notwithstanding the great activity in the study of this disease no
good remedy for it has been found. Wise travelers in tropical Africa
take every precaution against being bitten by tsetse flies. Much effort
is being exerted to check the spread of the disease among the natives in
some of the infected regions; chiefly by removing patients from the fly
region, by screening dwellings or by building them away from the damp
and marshy areas where the flies breed.

FILARIASIS

The first disease found to be transmitted by an insect was filariasis,
the subject of important investigations by Manson, Bancroft and others.
This disease of tropical and subtropical regions is caused by a thread-

TRANSMISSION: OF DISEASES BY INSECTS 251

worm, or nematode, known as Filaria bancrofti, which occurs in the
blood of man and of several of the lower animals as a slender larva
(microfilarid) about one-quarter of a millimeter in length. At night these
larvae swarm in the peripheral circulation, from which they are taken
into the alimentary canal of a blood-sucking mosquito (chiefly Culex
fatigans). In the mid-intestine of the mosquito the larva escapes from
its sheath and penetrates into muscular tissue, where it grows and devel-
ops for two or three weeks, after which it goes to some other part of the
mosquito's body, often to the base of the proboscis, whence the larvae
are carried into the blood of some vertebrate host, there to develop to
sexual maturity.

The larvae are often common in human blood without seeming to
injure the host in any way, but the adults (three or four inches long and
often found in groups) and ova that have escaped from the parent female
sometimes obstruct t