Production Note
Cornell University Library pro-
duced this volume to replace the
irreparably deteriorated original.
It was scanned using Xerox soft-
ware and equipment at 600 dots
per inch resolution and com-
pressed prior to storage using
CCITT Group 4 compression. The
digital data were used to create
Cornell's replacement volume on
paper that meets the ANSI Stand-
ard Z39.48-1984. The production
of this volume was supported in
part by the Commission on Pres-
ervation and Access and the Xerox
Corporation. Digital file copy-
right by Cornell University
Library 1991.Curnrll •Umnmnftf Ctlirart?
31 Irani, Xeu* Ijitrk
COMSTOCK MEMORIAL LIBRARY:
ENTOMOLOGY'
Illllllllllllllllllllllllllll^iP^i^lllS^ll^i^^lliP^i^i^ip^^ilii^iiilllllllllillllllllllllll
'Qli\ HENRY COMSTOCK
fTHE LIFE OF A BUTTERFLY.OTHER WORKS ON BUTTERFLIES BY THE AUTHOR.
The Butterflies of the Eastern United States
and Canada, with Special Reference to New England.
3 vols., Imp. 8vo, 2,000 pp. 96 Plates and Maps, plain
and colored. Half Levant, gilt top, #75.00.
Published by Houghton, Mifflin, <5r» Co., Boston,, Mass,
Butterflies : Their Structure, Changes, and Life Histories,
with Special Reference to American Forms. With an
Appendix of Practical Instructions. i2mo, 332 pp. 201
Figures. Cloth, #1.50.
Published by Henry Holt <5r» Co., New York, N. Y,
A Brief Guide to the Commoner Butterflies of
the Northern United States. i2mo, 12 + 206 pp.
Published by Henry Holt &= Co., New York, N. Y.
Fossil Butterflies. 4to, 100 pp. 3 Plates. Paper, #2.00.
Published $y the A meric an A ssociation for the
A dvancemtnt of Science, Salem, Mass.
Historical Sketch of the Generic Names pro-
posed for Butterflies : A Contribution to System-
atic Nomenclature. 8vo, 203 pp. Paper, #1.00.
Sold by the Cambridge Entomological Club, Cambridge, Mass.
IN PREPARATION.
A Students’ Manual of the Butterflies of North
America, north of Mexico.THE
LIFE OF A BUTTERFLY
~zA Chapter in Statural History for
the General Trader
BY
NEW YORK
HENRY HOLT AND COMPANY
i893Copyright, 1893,
By Samuel H. Scudder.TO
THE FOREMOST STUDENT OF THE LIFE HISTORIES
OF AMERICAN BUTTERFLIES,
WILLIAM HENRY EDWARDS,
OF WEST VIRGINIA,
SKjfe 3LtttIe ESEorfe
CONTAINING THE STORY OF ONE OF THEM,
IS RESPECTFULLY DEDICATED.
VI
?CONTENTS
PAGE
Preface ............................... 9
I.	General Account of the Milkweed
Butterfly......................11
II.	The Tongue, and How it Works . .	23
III.	The Course of its Life : One Phase
of the “Struggle for Existence”
among Butterflies.................30
IV.	Its Vagrancy: A Lesson in Geograph-
ical Distribution.................40
V.	The Critical Periods of its Life. .	66
VI.	A Favored Race: Mimicry and Pro-
tective Resemblance....................79
VII.	Scent Scales : A Question of Sexual
Selection.........................96
VIII.	Insect Vision. Does Sight or Smell
most control the Action of But-
terflies ?. . . .........................no
IX.	The Fore Legs of the Male and the
Hanging of the Chrysalis: a Les-
son in Classification.................133
X.	What is its Proper Name? A Brief
Chapter on Nomenclature as used
by Naturalists........................1488	Contents.
PAGE
XI. Some Points not touched upon, and
some Things we do not yet suf-
ficiently know :	A Suggestion
for Future Study................157
Explanation of the Plates.............180
Index ................................183PREFACE.
In the following work I have tried to present
in untechnical language the story of the life
of one of our most conspicuous American
butterflies. At the same time, by introducing
into the account of its anatomy, development,
distribution, enemies, and seasonal changes
some comparisons with the more or less dis-
similar structure and life of other butterflies,
and particularly of our native forms, I have
endeavored to give, in some fashion and in
brief space, a general account of the lives of
the whole tribe. By using a single butterfly as
a special text, one may discourse at pleasure
of many; and in the limited field which our
native butterflies cover, this method has a cer-
tain advantage from its simplicity and direct-
ness, and I trust may in this instance be of
service.
There is little in the volume, either in the
statement of facts or their discussion, whichIO
Preface.
has not already been published. But as such
accounts are scattered in many bulky works,
which only occasionally come under the eye
of the general reader, for whom this work is
intended, the presentation of the facts within
the compass of a single volume may interest
a larger audience, and perhaps gain for butter-
flies the serious study of some who had before
looked at them as merely pretty creatures,—
types of the frivolous.
There is, however, one point barely suggested
before, which is here pretty fully discussed, —
the question whether the particular butterfly
chosen as the centre-piece of the work can
strictly be regarded as a proper denizen of
regions as far north as New England, where
it nevertheless occurs abundantly year after
year. In short, in a volume devoted to our
native butterflies and meant to inspire an in-
terest in them, I am attempting to show that
my principal figure is, after all, but a tropical
interloper.
SAMUEL H. SCUDDER.
Cambridge, Mass.THE LIFE OF A BUTTERFLY.
I.
General Account of the Milkweed Butterfly.
HERE is probably no other but-
terfly so well known in every
part of the United States as
the Milkweed Butterfly, Anosia plexippus.
This is due to its large size, conspicuous
coloring, wide distribution, and its occa-
sional remarkable abundance. Of easy,
quiet flight when undisturbed, often sail-
ing smoothly with wide-spread wings, yet
ever ready to do battle with a tempestu-
ous wind, a reckless adventurer in its
contrasted livery of orange and black,
it seems the very beau-ideal of the con-12
The Life of a Butterfly.
tented, happy-go-lucky butterfly. It has
a spread of wing of between four and five
inches.
Examined close at hand we note the
different form of the fore and hind
wings, — both broad and ample, but the
former somewhat falcate, with smoothly
rounded curves; the latter rather trian-
gular, but with the outer margin strongly
arched. A deep orange red forms the
ground color, but there is a wide margin
of black on both wings more or less
dotted with pure white, while the veins
forming the framework of the wing are
all traced in black, and across the falcate
tip of the fore wings is a broad oblique
black band almost or quite merging into
the outer black border, and within it a
couple of rows of more or less sordid
white spots of a squarish form. The
body is black, but the thorax is dotted
with round white spots, and the abdomen
is narrowly ringed with the same. The
antennae, nearly as long as the abdomen,The Milkweed Butterfly.
13
are not scaled at all, and have a drooping
club; and the black palpi are conspicu-
ously marked with white, the apical joint
protruding like a pin tip. Beneath, the
wings are paler than above, but are very
similarly marked ; the sides of the body
are conspicuously dotted with white; and
the legs are purplish black with some
white markings and very long, hardly
divergent claws. The fore pair, however,
are very different, being much shortened,
the tarsal joints reduced to a single one
(with signs of division in the female) and
without apical claws. The male is further
distinguished by possessing at the tip of
the abdomen an extensile brush of long
hairs, and on the lower and inner branch
of the vein, forming the inner margin of
the discal cell of the hind wings, an ap-
parent thickening of the membrane on one
side of the vein, really a pouch, which is
made more conspicuous by its black color.
The eggs are exceedingly pretty objects,
somewhat in the shape of a blunted sugar-14
The Life of a Butterfly.
loaf, a little more than a millimetre — or
less than a twentieth of an inch — high,
amber green in color, the glistening sur-
face broken up by about twenty-two slight
longitudinal ridges, and the interspaces
traversed by straight and fine cross-lines,
while the summit of the egg is crowned
with a little rosette of ceils of extreme
delicacy.
The caterpillars are striking objects,-—
cylindrical, plump, naked worms, growing
to the length of nearly two inches, with
transverse bands of yellow and black. The
body tapers a little in front, and the head
is of much smaller diameter than the
middle of the body, with the face, which
is yellow, conspicuously marked with two
parallel black bows; from the top of the
second thoracic segment grows a pair of
long and slender flexible black filaments,
thrust upward, forward, and a little apart,
and which have a twitching motion when
the caterpillar is feeding or alarmed, and
move alternately forward in marching; aThe Milkweed Butterfly.
IS
similar, but shorter, more erect and in-
active pair of black filaments is also
found on the top of the eighth abdominal
segment. These filaments are scarcely
observable in the first stage of the cater-
pillar, which is then armed with several
series of simple, short hairs arising from
papillae, which are practically lost after-
ward.
The chrysalis is again a very striking
object, but simply on account of its color-
ing. It hangs by a slender little peduncle
of shining black, and is very compact and
stout, with plump, rounded form and only
the slightest projections, all of which are
rounded; the plumpness is in part due to
the shortness of the abdomen, the joints
of which, beyond the third, rapidly shorten
and contract, leaving at the hinder edge
of the latter a ridge which encircles the
body (except the wings) at its stoutest
part, and is made the more conspicuous
by bearing little tubercles set in a belt
of color, shining black in front, nacreous16	The Life of a Butterfly.
in the middle, and golden behind, while
the entire body of the chrysalis is of an
emerald green, except for a few gilt or
black dots on other definitely scattered
tubercles. It is over ah inch long, and
from eleven to twelve millimetres in di-
ameter at the stoutest part.
The first spring eggs are usually laid
near the base of the midrib of either
surface of the terminal or next to the
terminal leaves of the young plant of an
Asclepias, or milkweed, while they are
still erect or nearly erect. The under
surface seems to be preferred. Generally
but one egg will be found on a leaf, and
not often more than two or three on a
plant. Later they are also laid upon the
pedicel of the flower. The egg hatches
in four days or even slightly less, but is
sometimes delayed so as not to emerge
for five days or more.
The caterpillar feeds upon different
species of milkweed, Asclepias, although
“ it shows a wonderful dislike,” RileyThe Milkweed Butterfly.
17
remarks, ato the poke milkweed (A. phy-
tolaccoides Pursh.). . . . Larvae furnished
with this plant would wander about their
breeding-cages day after day, and would
eventually die rather than touch it” In
the north it generally appears to confine
itself to A. cornuti Dec., but has been
found on A. purpumscens Linn, and A.
incarnata Linn.; in the south and in
Missouri, it also feeds on the butterfly
weed, A. tuberosa Linn., A. amplexicaulis
Michx., A. tomentosa Ell., and A. curassa-
vica Linn., and has been taken in Cuba
by Dr. Gundlach on A. nivea Linn. It
has been discovered, too, on the dogbane,
Apocynum, A. androsaemifolium Linn., and
according to Coquillet will feed also on
Acerates.
The caterpillar eats voraciously, and
ordinarily matures rapidly. Sometimes,
however, it takes three or more weeks to
attain its growth. When observed toward
evening it will ordinarily be found quiet,
apparently resting for the night, planted18 The Life of a Butterfly.
on the under surface of the midrib of a
leaf, half way between the base and apex,
its head outward. From this it might ap-
pear that it fed only by day, but caged
specimens certainly eat at night, and I
have found it resting early in the morning
on the top of the leaves on a cloudy day.
It is almost always found near the top of
a plant, and when disturbed, so as to be
knocked off a leaf, the caterpillar coils
like a galley worm. Dewitz, writing of
the larva in Venezuela, says it spins a
thread on being seized, but I cannot
understand the statement; it spins less
thread than almost any caterpillar known
to me.
When preparing to moult it spins an
extra amount of silk upon the leaf for a
good foothold, and rests immovable for
many hours. The new skin has begun to
form beneath the old, and while still soft,
the new head-case is withdrawn from the
old into the region of the thoracic seg-
ments, till finally the old skin splits onThe Milkweed Butterfly.
19
the first segment behind the old head-
case, and the new caterpillar walks out
from its old skin; the old head-case falls
to the ground, and the old skin shrivels
to almost nothing, and is usually eaten up
by its former owner.
I once observed one of these caterpillars
while moulting; it had been stationary at
least twenty-four hours, and now first
began swaying its body from side to
side, — falling over so far that the thoracic
filament of the upper side became per-
pendicular, and then drawing itself forci-
bly back to an opposite position. The
muscular effort caused a considerable
indentation along the falling side of the
swaying larva at the point where the
white band widens, and at which muscles
are attached. The motion was repeated
about once in three seconds, and contin-
ued for nearly three-quarters of an hour;
now and then the larva would violently
shake its filaments or strain forward the
front of the thoracic segments, thus grad-20	The Life of a Butterfly.
ually detaching the old skin from the
new.
At last, after remaining quiet, as if to
gather strength for a final effort, it began
to make violent contortions, especially
about the thoracic region. These at first
seemed ineffectual, but suddenly the in-
tegument parted between the head and
body, and by the movements of the
larva, passed backward over the new
skin, slipping over the whole body at
once, and leaving a little empty pellicle
at the hinder extremity. The skin was
with difficulty removed from the fila-
ments, especially from one whose tip had
been bent in the former stage, and which
only parted after strong exertions; the
fresh filaments lay limp along the back
until they were gradually drawn forward,
the tip clinging to the moist body until
the last; but they did not regain their
full elasticity for some time. The re-
maining process scarcely lasted a minute;
the head, however, still remained attached,The Milkweed Butterfly.
21
and was only removed after repeated
lateral abrasions and violent efforts with
the front legs.
After these efforts the insect remained
quiet, resuming the same attitude, with
bent head, which it had taken before
moulting, — awaiting, undoubtedly, the
hardening of its integuments; and it was
nearly two hours before the colors of the
head became bright and fixed. The larva
now first devoured all the old pellicle,
except the head, and then moved off in
search of daintier diet.
The chrysalis usually hangs for about
twelve days, ranging in New England
generally from nine to fifteen ; but in the
South, according to Edwards, from five
to fifteen days ; in one case he reports
it as reduced to two days! On the other
hand, Gundlach says it hangs from eight
to twelve days in Cuba, and Dewitz gives
twelve days as the season in Venezuela.
I have known it to be extended in New
England to three weeks.22	The Life of a Butterfly.
“At last,” says Peak, “the golden spots
begin to lose their brilliancy and the
beautiful green disappears; the orange
wings of the imprisoned butterfly now
become visible through its temporary sar-
cophagus, which it bursts open on the fol-
lowing day, and the liberated insect soon
takes wing to join its comrades, select
its mate, and pass the happy hours of a
brief existence in revelling in the sweets
of the flowers among which it sprang
into being.”The Tongue and How it Works.
|NE of the most striking things
about butterflies, and one which
never fails to excite the admira-
tion of the observer, is the way in which
they feed. Every one has seen them at
the blossoms of clover, of milkweed, of
golden-rod, thistle, and phlox; and any
one who, approaching cautiously, has
then looked at them from near at hand,
has seen a wonderful apparatus like a
watch-spring suddenly uncoil from the
lower part of the head; and while the
tremulous excitement and eagerness of
its owner is shown by the rapid and
repeated uncoiling and half coiling again
of this organ, as well as by the shivering
palpitation of the wings, the tentative24 The Life of a Butterfly.
thrusts to probe the bottom of the gauzy
dish and the quiet satisfaction of the
butterfly when this is reached show that
it is through this slender, flexible thread
that the sweet fluids of the clover and
other blossoms are drawn into the body.
This flexible conduit, moreover, is the
most characteristic of all the structures
peculiar to the order of insects to which
butterflies belong, and may well merit
particular attention.
Notwithstanding its delicacy, and al-
though almost entirely concealed when
coiled, this “tongue” is frequently as
long as the entire body, and consists of
two lateral halves united down the
middle; each part is composed of an
immense number of short, transverse
rings, which are convex on the outer
surface, concave on the inner; and it is
by the union of these inner concavities
that a central tube is formed. The lateral
rings are also partially hollow, and were
therefore formerly supposed by some toHow the Tongue IVorks. 25
form sucking-tubes, in which case the
insect might be said to have two mouths,
for there would be two entrances to the
throat. This, however, is not the case,
the interior of each lateral half being
closed at the end and filled with nerves,
air-tubes, and muscles. The rings of
which it is composed are made up of
a great number of plates, united by the
more yielding part of the cuticle, allow-
ing of great freedom of motion. These
rings, at the points where the convex
and concave sides meet, are furnished
with a series of oblique, curving plates
or hooks, which, when the two lateral
halves are brought together, interlock in
the most complete manner, to form a
perfectly flexible yet impervious tube.
The external walls of the lateral tubes
are supplied with curious minute papillae
of greatly varying shape, size, and abund-
ance in different groups, but in general,
more highly organized and abundant in
the highest butterflies. In our Milkweed26
The Life of a Butterfly.
Butterfly, Burgess tells us, they are seated
on little circular plates, and are dotted
all over the outer wall, but especially
near the tip, and occur also, though in
much less number, within the central
canal. In this simple development they
must be regarded as organs of taste or
touch, as suggested by Fritz Muller; but
in some moths, as in the orange moth,
they become notched spines, and even act
as a saw or file, so that the tongue may
work its way through plant tissues in
search of juices. The muscles mentioned
as found in the interior of each lateral
half of this wonderful structure are ob-
liquely disposed, serving to coil the whole
into the watch-spring-like form in which
it is packed away when not in use.
Where was this curious and compli-
cated organ in the caterpillar? From
what, that is, did it take its rise? One
will have to look carefully to discover
it at all! The principal organs one will
find in the mouth of the caterpillar areHow the Tongue Works.	27
a pair of stout nipping jaws by which
it bites the leaves on which it feeds; but
just behind them, on either side of the
spinneret, will be seen little hemispherical
prominences, each with a pair of minute
appendages, the outer consisting of sev-
eral joints and closely resembling the
antennae, the inner and smaller of only
one; and it is this wee stiff thing, capable
only of the slightest motion of withdrawal
and protrusion on its cushiony base, that
becomes so enormously developed into
the complicated watch-spring tongue of
the butterfly, while the outer appendages
become in the perfect stage the feath-
ery-scaled side-pieces which protect the
tongue when rolled.
But now that we comprehend the struc-
ture of this wonderful piece of mechanism,
and can appreciate the change that has
been wrought in its development from an
utterly simple, almost microscopic joint,
do we understand any better its actual
use in extracting honey from flowers?28 The Life of a Butterfly.
Some have thought that the upward flow
was due to capillary motion; others to the
action of the so-called sucking stomach,
a sac-like expansion of the alimentary
canal just in advance of the true stomach;
others still that it is forced on by suc-
cessive undulations and contractions of
the tube itself.
The investigations, however, of one of
our own naturalists (who afterward dis-
tinguished himself in the construction of
the swiftest yachts in the world) has
shown the existence of a minute mus-
cular sac within the head, furnished with
a valve at its front extremity, where it
opens into the base of the tongue. When
the radiating muscles running from the
walls of the head to the periphery
of this sac are contracted, the sac is
expanded, and into the vacuum thus
produced the fluids into which the
tongue is plunged necessarily ascend.
On the relaxation of these muscles and
the squeezing of the sac by the otherHow the Tongue Works. 29
muscles which encircle it, the fluids now
in it, prevented by the valve from re-
treating the way they came, are forced
down the alimentary canal. When, then,
we see a butterfly busily engaged upon
a flower, it requires little imagination to
picture to one’s self a little force-pump
steadily at work within the dear creature’s
diminutive head, transferring the fluid
nectar from the base of the flower to the
greedy stomach.
;III.
The Course of its Life: one Phase of the
“Struggle for Existence” among Butter-
flies.
| HE course of life of the Milkweed
Butterfly is a simple one. Win-
ter is passed in the perfect or
“imago” state, but in what sort of places
is not known; in the extreme south it
remains on the wing, but that is not the
case further north, where the cold would
make it impossible. In the proper range
of the species (of which we shall have
somewhat to say later on) the warmth
of the early spring lures the hibernating
butterflies from any places of conceal-
ment they may have had, faded in color
but not so often ragged in attire. In
West Virginia, according to Edwards,
they may be seen on the blossoms ofThe Course of its Life.
3i
the wild plum the last of March, and
on lilacs and other flowering shrubs in
April. The tender leaves of the milk-
weed, on which the caterpillar feeds, are
just appearing above the ground, and on
these the females hasten to deposit their
eggs; they are usually laid on the under
surface of the terminal leaves while these
are still vertical. They hatch in four or
five days; later in the season, in the
south at least, in two or three days.
Immediately it has escaped from the
egg, the caterpillar completely devours
the shell, as if to leave no manifest token
to the wily and vigilant ichneumon-fly
that its former inhabitant was near at
hand, and then attacks the leaf on which
it was born, eating a slender hole often
entirely through it, and when it has done
feeding retires to the concealed side of
the leaf, — if it be still erect, to the inner,
that is the upper side; if extended hori-
zontally, to the lower surface. As soon
as appetite returns, — and it is a voracious32
The Life of a Butterfly.
feeder, —: it has its food ready at hand.
So it eats and rests, and rests and eats, at
night — to judge from a few observations
I have made — quite as well as by day,
except that it is more sluggish in cool
weather, and nights are cooler than the
day. In a day or two (the time de-
pending partly upon the weather), the
caterpillar makes its first moult, but its
habits remain much the same. Three
other similar moults follow before the
caterpillar has attained its growth, which
may be acquired in eleven days from
the egg, though it is usually longer than
that, and in cool weather may be greatly
extended.
Then comes the change to chrysalis,
to seek a good place for which the cater-
pillar usually leaves the plant (though I
have found the chrysalis hanging pendant
from the leaf) and seeks some such stable
place as the under side of a fence rail
or a jutting rock from which to suspend.
Mr. Edwards once found one on theThe Course of its Life.
33
under side of the T rail of a railway
track! Here it hangs for a variable
period, — two to fourteen days, according
to the season and temperature, and per-
haps the exposure; and the butterfly
again makes its appearance, the cycle of
changes complete.
How often this cycle may be repeated
during the year is doubtful and certainly
variable, the weather and latitude being
undoubtedly again important factors; it
has been a matter of some dispute. Riley,
making his observations in Missouri, con-
siders that there are but two broods
annually in that region. Edwards, in
West Virginia, believes there may be
four or five in his district, and one would
hardly look for more in the one than in
the other; but Edwards’ belief appears
to be founded upon the fact of finding
the insect in all its stages at nearly all
times the summer through, and upon his
observations of the time required for the
entire cycle, which is sometimes as short
as three weeks.
334
The Life of a Butterfly.
He leaves out of account entirely the
possibility that the parent butterfly may
lay eggs for a considerable length of time,
because he does not believe this to be
the case, and it certainly would be very
difficult to prove it to be or not to be
the case in a Southern district; but in a
more Northern region, where the number
of broods would be likely to be less than
at the South, it would not be so difficult,
and from what information I can gain I
am very strongly inclined to believe that
there are at the most two broods a year
north of the latitude, say, of Washington.
I shall on a future page explain this
more fully, in considering the distribution
of this butterfly, of which I believe there
is but one brood a year in most of New
England and northward. I should not
be surprised if it should turn out that
there were nowhere in the United States
more than three broods a year, and it
may even yet be proved that there are
but two.The Course of its Life.	3 5
In any case, this is one of the simplest
life-histories we have among our native
butterflies; and the dispute regarding it,
and the uncertainty that still surrounds it
and probably will surround it for a long
time, show how wide a field for observa-
tion is open to young naturalists. The
variations on this theme among our but-
terflies are endless. The cycle is inva-
riable; but winter may intervene at any
stage, usually at an identical stage in any
given species, though occasionally at two
different stages.
Relatively few butterflies, in New Eng-
land at least, winter as butterflies; such
are mostly confined to the Angle-wings
and their allies, of which Euvanessa an-
tiopa, the Mourning-cloak, is the best
known example. A considerable number
are dependent for their perpetuity upon
the power of resistance to cold possessed
by the egg; and then the egg is never
laid upon the leaf of the food plant,
which would fall to the ground in the36
The Life of a Butterfly.
autumn, but upon the twigs of the same
or upon some stable point near by if the
plant be an annual. The caterpillar then
usually hatches in the earliest spring and
feeds upon the tender leaves before they
are fairly unfolded. Such is the case with
some of the delicate Lycaenids, the Hair-
streaks, Blues, and Coppers; apparently
it is more common in this subfamily than
elsewhere, but it is known to occur in a
number of widely different butterflies, one
striking case being that of Parnassius, an
alpine genus.
What the condition of the embryo may
be in these instances is not at all known,
but it is certain that the cases are not
few in which the caterpillar hibernates
without tasting a morsel of food besides
its own egg-shell, much the same as if it
had wintered fully formed in the egg.
The doctors tell us we should never go to
bed hungry, but here are creatures which
go to their long winter's rest before they
have tasted any proper food; this isThe Course of its Life.
37
tolerably common among the Satyrids or
Meadow-browns.
Then there are others which hibernate
as partly grown or fully grown cater-
pillars; and these with the last — that is,
such as hibernate in some larval stage
or other — include a very considerable
proportion of our Northern butterflies,
perhaps even a majority of them. Among
them should be counted, perhaps, such
butterflies as our Dusky-wings of the gen-
era Thanaos and Pholisora, which in the
autumn prepare for the change to chrysa-
lis by making a sort of cocoon for the
purpose; these they never leave again
as caterpillars, but in them remain quite
unchanged until the early spring, when
after a short period in chrysalis they are
among the earliest butterflies of the
season. Finally, there is the chrysalis,
which really seems made to be the nor-
mal stage for hibernation for Lepidoptera,
and one which appears to be the nearly
universal condition among the moths, but38
The Life of a Butterfly.
which is vastly less common among the
butterflies, aspiring creatures whose lives
are as variegated as their broad wings.
With such a variety, all possible variety,
in the matter of hibernation, what wonder
that the lives of butterflies are varied?
Some pass a single cycle in the course
of the season, ending where they began,
some two or more; some among the
Northern insects appear to take two years
to complete the cycle, wintering succes-
sively in two different periods of life; or
they may take sometimes one and some-
times two years, as would appear from
recent observations in Canada; while to
cap the climax, there are not a few in-
stances, confined apparently to regions
where winter is a prime factor in life,
and probably resulting therefrom, where
a butterfly may be at the same time
single and double brooded, the progeny
of a single batch of eggs giving birth to
butterflies partly in the course of the
same season, partly in the season follow-The Course of its Life.	39
ing, the later butterflies flying with the de-
scendants of the earlier; that is, with their
nephews and nieces! This comes about
in a curious way, by the premature hiber-
nation from midsummer on of a certain
portion of the original brood at just that
particular time of life when their nephews
and nieces will hibernate. The tendency
to hibernate at that specific time of life
is thrown back into midsummer, when
there are no signs of provocation thereto.
The study of the influence of winter
life upon the histories of our butterflies
unfolds a curious and perplexing chapter.
We see what an element it has been in
the development of these creatures, how
they have always had to contend with
it, and in what manifold ways they have
made their struggle for existence. How
many may have succumbed!Its Vagrancy: a Lesson in Geographical
Distribution.
||ARDLY any other of our North
American insects equals our
Milkweed Butterfly in interest
when we consider its distribution. With
the exception of a few butterflies which
we possess in common with Europe, and
which may have had their origin there,
no other species is so wide-spread. There
is still some question whether certain
butterflies found in the American tropics
are to be regarded as, with this, varieties
of one species, or as distinct forms; but
without discussing a somewhat unprofit-
able question, we may better assume,
what is probably the case, that all belong
to a single species with local variations.Its Vagrancy.
4i
This granted, we find our butterfly to
be a distinctly American form, belonging
properly to the tropics. It is found all
over the West India Islands, as well as
on the mainland, and extends beyond the
tropics much farther in the north than in
the south, for its southernmost known
location in South America is the very
northern border of Patagonia, in latitude
40°, while in North America it extends
over almost every part of the United
States, no part of which lies within the
tropics, and even over a large part of
Canada, having actually been reported
from the southern borders of Hudson Bay,
and even from the Athabascan region in
the far northwest, nearly twenty degrees
farther from the tropics than in South
America.
Belonging as it does to a distinctively
tropical group of butterflies, it appears
here quite out of place. The semi-tropical
character of the southern United States,
somewhat more marked than in similar42
The Life of a Butterfly.
latitudes in South America, does not
make its appearance there surprising; but
farther north, beyond Philadelphia (which
would be latitudinally equivalent to its
farthest in South America), it clearly
appears like an interloper. Precisely here
comes in the interesting feature in its
distribution. We are accustomed to think
of the butterflies we see as born and bred
in the district where they are seen. It
is otherwise with birds; we know that
very many among them come and go
with great regularity and for great dis-
tances, though it is only within given
latitudinal boundaries that they will breed,
each in its own domain. It has hardly oc-
curred to entomologists to inquire whether
this may not sometimes be the case with
butterflies. The very fact that some birds
do and some do not move northward and
southward with the seasons, should sug-
gest the possibility that this may be the
case with other flying creatures.
Particularly should such a suggestionIts Vagrancy.
43
be heeded in those groups where great
flocks are known to occur and apparent
migrations take place. There are two
such groups represented among our but-
terflies, one of them that to which our
Milkweed Butterfly belongs; and there
seems to me nearly every reason to be-
lieve (what every recurring season of
observation serves only to strengthen)
that the natural northern limits of our
butterfly are not farther north (perhaps
much farther south) than about latitude
40° on the Atlantic border, and that the
eggs, caterpillars, and chrysalids of the
butterfly found much north of this are in
all cases the production of immigrants
from farther south, — no perfect butterfly
born at any considerable distance north
of this assumed limit ever laying an egg
the same season.
Indeed, it is probable — such is in my
belief the vagrant habit of the butterfly—
that a butterfly successfully hibernating
farther south may fly to some distance44
The Life of a Butterfly.
still beyond this before depositing all her
eggs \ and the progeny from these, though
born so far north, would again lay eggs.
But by far the greater part of the eggs
laid beyond the limit — almost the whole
of them — are laid by butterflies of the
second generation, born farther south, and
wandering northward in search of “fresh
fields and pastures new; ” and their pro-
geny will not lay eggs the same season.
This, let it be at once asserted, is a
theory only, made to explain the observed
facts, but appearing at present the best
way of explaining them. Let us state
some of these. We have already seen
that in West Virginia the butterflies ap-
pear as soon as the first shoots of milk-
weed arise above the ground in April.
The same is the case at least as far
north as Newburgh, New York; that is,
the first butterflies seen are those which
by their faded colors are believed to have
Jiibernated, and they lay their eggs on the
milkweed when it first comes out of theIts Vagrancy.
45
ground. At Newburgh the season is per-
haps three weeks or a month later than in
West Virginia, and at least as far north as
this, it would seem to be perfectly easy for
the hibernating butterfly to have flown
from even as far as Florida. It would
have had to fly no faster than a man walks,
counting only the daytime and deducting
half of that for bad weather.
Still farther north, where observations
have been quite as numerous as south
of it, a hibernated butterfly of this
species is an extraordinary rarity, and
the first butterflies of the season are those
which appear when the milkweed is al-
ready a foot or more above the ground,
synchronous with or slightly later than
the descendants of the hibernators farther
south, fresh and bright butterflies, cer-
tainly born the same season. According
to my observations in different parts of
New England, they are long-lived, laying
but a few eggs at a time, and probably
not depositing all until several weeks after46
The Life of a Butterfly.
birth. The butterflies are on the wing
the rest of the season, — first the parents,
next their children, the parents grad-
ually dying out.
There is no evidence whatever of a sec-
ond brood of caterpillars in New England;
that is, that these children ever lay eggs
the same season. In all double-brooded
butterflies known in New England, the
limits of their broods can be told, some-
times it is true rather vaguely. Still there
appears to be more definiteness here than
farther south, where the summer is longer;
and from observations made in numerous
places by scores of individuals, it appears
quite plain that our Milkweed Butterfly
passes in New England but once through
the entire cycle of its changes; and
though the theory here proposed finds
no known counterpart elsewhere, it ap-
pears to be gaining ground among ob-
servers. Of course it will require many
further observations to establish it.
Among butterflies it is only whenIts Vagrancy.
47
large numbers are concerned that their
movements attract attention. Very many
such observations are on record, but the
purpose of the movements has rarely
been even guessed at. With most but-
terflies which have more than one brood
in a year, and notably in our Milkweed
Butterfly, the number of individuals upon
the wing toward the end of the season
is far greater than at the beginning, and
it would therefore be easier to detect
massive movements in the later than in
the earlier part of the season. Now,
nobody has ever observed that the preva-
lent movement of our butterfly in the
earlier part of the season was northward,
as on the theory here upheld should be
the case, and this is undoubtedly its weak
point. Nevertheless I look confidently
to seeing this observed and proved at no
distant day, particularly by observers in
New York and Ohio.
The fact that the butterfly has been
seen on the shores of Hudson Bay and in48
The Life of a Butterfly.
the Athabascan region, beyond the limits
of the growth of its proper food plant,* is
another proof of its northward migration
in the warm season. But of the return
movement in the autumn there is plenty
of evidence. Apparently wherever it
abounds in any part of the United States,
certainly from Ontario to Florida and from
Massachusetts to Iowa and Missouri, the
butterfly shows a tendency in the autumn
to swarm in immense numbers, and at
night to settle down upon trees or shrubs,
so as to change the whole effect in color.
I will give one instance of the many that
have been published. Writing of Brigan-
tine Beach, New Jersey, in 1885, Dr. John
Hamilton, of Alleghany, Penn., says: —
“The multitude of this butterfly that as-
sembled here the first week in September is
* Asclepias is not found about Hudson Bay nor
north of the Saskatchewan, according to the Canadian
botanists; and although Apocynum does occur both in
the valley of the Moose River and in the Athabascan re-
gion, it is too exceptional a food plant to be considered.Its Vagrancy.
49
almost past belief. Millions is but feebly
expressive, — miles of them is no exaggeration.
On the island is a strip of ground from one
hundred and fifty to four hundred yards wide,
and about two and one half miles in length,
overgrown with Myrica cerifera (bayberry).
After three o’clock these butterflies, coming
from all directions, began to settle on the
bushes, and by evening every available twig
was occupied. To see such multitudes at rest,
all suspended from the lower sides of the
limbs, side by side, as is their well known
custom, was something well worth seeing. One
evening I travelled more than half the distance
of their encampment, and learned that it
extended the whole length and breadth of the
bushes. In the morning they gradually sepa-
rated, and did not appear unusually numerous
during the day, but in the afternoon they came
again as described. I found them on the
second, the day of my arrival, as related above;
and this was repeated daily till the sixth, the
forenoon of which was rather calm and sultry.
A storm of wind and rain came on about two
o’clock, p. m., continuing till midnight. The
4So
The Life of a Butterfly.
next afternoon few came to camp; the great
army had disappeared. But how? when?
where to? During the next few days they
appeared again in considerable numbers,—
about as they had been observed in former
Septembers, — but insignificant when compared
with those that preceded. The males and
females were about equal in numbers. Not a
single stalk of their food plant (Asclepias)
grows on the island.” *
This gathering of the clans is but the
first step in the southward movement,
which also has been observed in numer-
ous places. One of the earliest accounts
we owe to Riley, who states that P. B.
Sibley, of St. Joseph, Missouri, on Sep-
tember 19, 1868, saw “ millions of them,
filling the air to a height of three or four
hundred feet for several hours, flying
from north to south.”
I was myself fortunate enough to ob-
serve a movement of this sort during the
autumn of 1888; between nine and ten
* Can. Ent., xvii. 204.Its Vagrancy.	51
o'clock in the morning of September 2,
while sitting on the piazza of a house
facing the sea-shore at Hampton, N. H.,
and only a stone’s throw from the water,
a continuous stream of these butterflies
passed before me toward ,the southwest,
following the line of the sea-coast, with
the wind about northwest. There were
never less than three or four directly in
front of me, often a dozen or twenty. In
the hour that I watched them, I calcu-
lated that at least fifteen hundred passed
me, and without exception in the same
direction.
As flocks also occur in Florida, it seems
highly probable that the southern move-
ments may extend over the entire United
States, and that the northern limit of the
returning hibernators may be the limit
within which the butterfly is double-
brooded. As Riley, the first to suggest
an annual migration of this butterfly,
remarks: “ There is a southward migra-
tion late in the growing season in congre-52
The Life of a Butterfly.
gated masses, and a northward dispersion
early in the season through isolated
individuals.,,
Another fact in support of this theory
is that no one has ever found a butterfly
of this species in hibernation, while there
is hardly a single one of our butterflies
which hibernates in the perfect stage of
which there is not some record of its
having been found; and yet our Milkweed
Butterfly is nearly twice as large as any
of these, and would therefore be more
readily noticed; and if it hibernates,
would naturally seek just such spots as
are sought by them. It is by no means
improbable, then, that in the winter
the species does not exist north of, say,
latitude 310.
In further support of the theory may
be mentioned its observed remarkable
powers of flight, and particularly its com-
mercial extension in recent years, which
forms a chapter in its history quite ex-
ceptional in a butterfly; for within thirtyIts Vagrancy.
53
years, or a little more, it has begun to
invade so many regions of the world as
to make one think at first blush that it
may some day vie with Vanessa cardni
in cosmopolitan character. The facts
concerning its exotic distribution, so far
as I have been able to gather them, are
as follows.
It first reached the Hawaiian Islands,
fully two thousand miles distant from
America, some time not far from 1845 to
1850. At any rate we have the direct
statement of Rev. Luther H. Gulick, who
was born upon the islands, that in 1852,
after eleven years* absence, he returned
to the islands, and his brother drew his
attention to the fact that Asclepias had
been introduced during his absence, and
had already become a troublesome weed;
that his brother had noticed that wher-
ever the milkweed appeared, there also
the Milkweed Butterfly made its advent,
a butterfly unknown until after the milk-
weed had been introduced.54
The Life of a Butterfly.
We next find it in 1857 as far away
as the Island of Ponape, in the Caroline
Islands of Micronesia, a distance of
another two thousand miles or so from
the Hawaiian Islands. This fact we also
owe to Dr. Gulick’s personal testimony.
He was for some time a resident of
Ponape, and the butterfly was first seen
by him in the year mentioned, not long
after he had discovered several young
milkweeds, which had sprung up in earth
in which various other plants had been
brought from the Hawaiian Islands in a
Wardian case. The plants were brought
in a missionary vessel which sailed from
Honolulu, and on its way to Ponape
touched only at Apaiang of the Gilbert
Islands and Ebon of the Marshall Group,
both low coral atolls, and at Kusaie, which
is of basaltic formation and richly clothed
with verdure, but where the butterfly did
not then occur. It is evidently impossible
that in a voyage consisting in the whole
of fifty-four days, the insect in any stageIts Vagrancy.
55
or stages could have been transported in
the Wardian case itself, for it easily un-
dergoes all its transformations in warm
regions in a month or five weeks at
most.
If the butterflies were introduced at
that time, as there is every reason to
believe from Dr. Gulick’s accounts, there
seems no other supposition possible than
that an impregnated female flew into the
hold of the vessel while lading at Hono-
lulu, and was carried perforce to Ponape;
or possibly a pair of butterflies. It would
certainly be absurd to suppose that a
gravid female could have flown over two
or three thousand miles of ocean, and in
addition have appeared on Ponape Island
almost simultaneously with a few plants
of Asclepias. As the butterflies live
through the entire winter and then lay
eggs in the spring, there is nothing in
any way really surprising in Dr. Gulick's
statements, unless it be impossible for an
impregnated female to live in enforced56
The Life of a Butterfly.
hibernation a couple of months without
laying; when it would be necessary to
suppose a pair to have been transported,
which would of course be more strange.
Granting our explanation to be just, it
is highly probable that it was from this
single ancestor (or pair) that the swarms
which have now spread over the entire
South Seas, in many of which it is the
commonest butterfly known, have sprung.
Our knowledge of the period and extent
of this later distribution we owe largely
to Professor Semper, who states that the
butterfly was first seen in 1863 by Captain
Rachan, one of the numerous collectors of
the Museum Godefroy, on the islands of
the Tonga or Friendly group, again nearly
another two thousand miles from Ponape.
The first specimen actually obtained was
secured in 1866 on Niuafau, one of the
islands of this group; and in the same
year caterpillars were discovered on As-
clepias curassavicay a plant now spread
quite as far as the butterfly.Its Vagrancy.
57
We now begin to be able to record in
part the rapidity of its spread, for it was
first seen in Tutuilla, one of the islands of
the neighboring Samoan group, in 1867;
but upon Upolu and Savaii, islands of
the same Samoan group, distant at the
nearest some fifty miles, not until 1869.
Yet in Upolu it became one of the
commonest butterflies in 1870. It was
not until 1868 that it was discovered at
Tongatabu, one of the southern of the
Tonga Islands, but in the same year it
was seen in the open sea five hundred
nautical miles to the southeast. In 1869
it had appeared at Rarotonga, one of the
Hervey Islands, five hundred miles or
more away. In 1870 to 1872 it was found
on Huahine and Tahiti of the Society
Islands, again five hundred miles or more
distant.
So far the account of Professor Semper.
But Mr. James J. Walker, who sailed in
the South Seas in 1883, and nearly
everywhere found this one of the com-The Life of a Butterfly.
58
monest butterflies, states that he was
informed at the Marquesas Islands (which
lie to the northeast of the Society Isl-
ands, again at the distance of some five
hundred miles), by a Roman Catholic
missionary residing there forty years, that
he distinctly remembered seeing the first
specimen there about the year i860; it
should be noted that the Marquesas Isl-
ands are nearly as distant in a southeast-
erly direction from the Hawaiian Islands
as the Carolines are to the southwest.
Mr. Walker also found the butterfly on
the Hervey and Society Islands, and at
Oparo, one of the Andaman group, in
28° south latitude, though it had not
then reached Pitcairn Island, which lies
much farther east and somewhat farther
north.
These statistics indicate its movements
from the Caroline Islands, in an easterly
and southeasterly direction, but it has
also left its marks by the way, in a south-
ward extension from this route of travel.Its Vagrancy.
59
For it has reached Waigiou, New Britain,
New Ireland, New Guinea, the Louisiade
Islands, every part of the Solomon and
New Hebrides groups, the Duke of York
Island, the Loyalty and Fiji Islands, New
Caledonia, Norfolk Island, the northern
island of New Zealand, the entire east-
ern coast of Australia, from Cape York
southward even as far as Hobart Town
in Tasmania. It reached Lord Howe’s
Islands in 1870; Clarence River, on the
opposite coast of Australia, in 1871; Mel-
bourne in 1872, and has now extended
even to Celebes, and according to Kirby,
to Java.
It thus appears that it now possesses
a territory in the Pacific Ocean of at
least uo° of longitude and 65° of latitude.
But this is by no means all; it has
moved also in some strange way in the
opposite direction from the American
continent. It has long been known at
Bermuda as one of the extremely few
butterflies to be found on that island.6o
The Life of a Butterfly.
Specimens now in the collection of
Messrs. Godman and Salvin were taken
in 1864 on the islands of Fayal and
Flores, but it seems not to have been
since recorded at the Azores. In 1877,
however, it made its appearance on the
continent of Europe at La Vendee, on the
Atlantic coast of France, and a number
of instances of its capture in England
have since been signalized. These in-
stances are so numerous, and recorded
for so many different years, that it would
seem highly probable that the butter-
fly has been endeavoring to maintain a
foothold ever since 1876, when the first
instance of its occurrence was recorded.
The first specimen was found at Neath,
in South Wales, in September ; a second
one in Sussex in the same month, and a
third at Hayward’s Heath in October. In
i877 one was taken at Poole Harbor. It
did not appear again till 1881, when a
specimen was taken in Kent in September.
Again in 1884, one was taken in the IsleIts Vagrancy.
61
of Wight. In August and September,
1885, nine specimens were taken in the
counties of Dorset, Devon, Cornwall, and
the Isle of Wight. It was again taken in
1886 in the south of England, in Guern-
sey, at Gibraltar, and in Portugal.
I have spoken of this extension of its
natural region as one due to commercial
agencies, because it would seem that the
distance to which the insect has been
carried must be due to something more
than its very remarkable powers of flight.
The fact that the butterfly has been seen
flying at sea an immense distance from
land is a sufficient proof of the latter,
and we should be far from questioning
its power to compass with no very great
difficulty one-half the extreme distances
to which we know it has been carried
without power of alighting. But that this
should occur with a female heavy with
eggs, or two individuals together, male
and female (and no other supposition
would permit us to understand its subse-62	The Life of a Butterfly.
quent propagation in the regions visited)
is past credence, — more especially as we
have in the instance of its spread from
the Hawaiian Islands to the Caroline
group an almost certain proof of the
method of its transport, through artificial
aid. The alighting of one of these but-
terflies laden with fertile eggs upon some
part of a vessel or within its hold would
be a strange but by no means impossible
occurrence; and this is all that is neces-
sary to explain its transport over the
wider regions.
That having once established itself in
one of the Micronesian Islands, it could
easily spread over the whole of Polynesia
through the insect’s natural powers of
flight, will hardly be questioned. But
that this has taken place not only within
(historic times, but within the last twenty
or thirty years, as has been shown by
Semper, makes it almost certain that its
first introduction to the South Seas was
by artificial means; for if it could haveIts Vagrancy.
63
been brought about solely by the powers
of flight of the insect, aided by the nat-
ural currents of the air, it would have
happened long ago ; and the facts that the
insect has been able to establish itself
wherever it chose when it got a foothold,
and that until a very recent period it
has not so established itself, are sufficient
proofs that commercial agencies, so much
more abundant in later times than for-
merly, have been the great means of in-
troducing these butterflies to the islands
of the Pacific.	It is highly probable
that it owed its first introduction to the
Hawaiian Islands to similar agencies, and
that its appearance in Europe is due to
the same cause.
This is all I have now to say of the
theory here advanced, that the butterflies
of this species found annually in the
northern United States and northward
must be regarded as a wave of migrants
from the south which have come north
like the birds to breed, after which their64
The Life of a Butterfly.
young return, or if they fail to do so,
perish. What after all is this but an
exaggerated case of the perpetual flux
which has been going on with all our
Northern butterflies ever since at least the
glacial epoch? There were certainly no
butterflies here when the country was
flooded with ice, and all that have since
appeared are but immigrants from the
South; and at the northern limits of their
distribution there have no doubt ever been
a continual struggle and an ebb and flow
of life as a severer winter or an exces-
sive drought destroyed the species in toto
over a certain range, replenished later by
migrants from the border.
In short, the present distribution of life
is the result of the constant attempt of
every species to extend its territory, to
gain more room to live, the outcome of
the balance of forces which have held all
in restraint. And though our Milkweed
Butterfly presents phenomena in its dis-
tribution of an apparently exceptionalIts Vagrancy.
65
kind, showing it to be an unusual vagrant,
it is probable that any species we might
select for discussion would in the last
analysis serve equally well to show how
the struggle for room to live in has been
but one phase of the ever-present struggle
for existence.
5V.
The Critical Periods of its Life.
OW-A-DAYS it is not necessary
to insist upon “the struggle for
existencepopular literature has
made it a household phrase. Every crea-
ture, all its life long, may be said to be
subject to the bondage of danger and
fear; and every creature is furnished with
structural or physiological resources to
meet danger. Still, there are in the lives
of the lower animals, as well as of man,
certain critical periods when more than
at others they are liable to destruction.
Although through most of its life pro-
tected to a certain extent, as we shall see,
by its noxious qualities, I see no reason
to suppose that in its egg state the Milk-
weed Butterfly is not subject to the same
decimation to which I believe all or nearlyThe Critical Periods of its Life.	67
all butterflies are doomed. From my
experience (not very extended) in observ-
ing eggs of butterflies laid in the free air,
I should indeed judge that decimation
was rather a feeble term to apply to the
destruction of egg life. I have in fact
seen spiders at their meals upon them;
and Mr. W. H. Edwards, speaking of the
Violet-tip, Polygonia interrogationis, once
wrote: —
“ When it is considered how many eggs are
laid, and that so short a time intervenes be-
tween the egg and the imago, it is surprising
how few butterflies of this species are the
result. From eggs that were laid on my vines
in July and August, amounting I am sure to
many hundreds, very few larvae were hatched.
. . . The eggs are destroyed by spiders and vari-
ous insects by wholesale. I have had the con-
tents of one of my [breeding] kegs swept away
in a night, leaving not a trace of shell behind.”
Not a few eggs, too, are eaten dry by
minute parasitic Hymenoptera so small
as to look like mere grains of dust. Lit-68
The Life of a Butterfly.
tie attention has been paid to these by
naturalists, but from my fortune with
eggs I have found in the open air and
attempted to rear, a good tenth must be
ruined in this way; they will even invade
the muslin net under which eggs are laid
by butterflies in captivity, and pierce
them. Among the dozen and a half of
our butterflies whose eggs have been
actually found so pierced, the Milkweed
Butterfly must be counted. If, then, any
possible noxious qualities in the egg are
of no avail against that speck of dust
the Trichogramma, spiders and crickets
are not likely to eye it askance. It is
highly probable with this butterfly, as it
is certain with many others, that the egg
is the most perilous period of its life, one
whose mortality will show the greatest
percentage of loss to the species.
To counterbalance this, the egg state
is usually the shortest of any of the four
the saved remnant are to pass. In the
Milkweed Butterfly, its duration does notThe Critical Periods of its Life. 69
exceed three or four days. It is very
rarely less than that in any butterfly,
even in the hottest season or region, but
is sometimes prolonged considerably; and
curiously enough, this is particularly the
case in those same species in which the
larval stages are also prolonged, as if a
certain inertness belonged to the whole
life. In view of the dangers to the egg,
the fact we have already mentioned,
that some species winter solely in the
egg state, is not a little surprising. It
should be noted, however, that with almost
no exception, the species of temperate
regions known to winter in the egg belong
to the gossamer-winged butterflies, the
Lycaeninae, the shells of whose eggs are
as a rule of exceptional thickness; and
the same is true (or the eggs are very
coarsely ribbed) in every known case
outside this group, with a single published
exception, which I believe erroneous.*
* Coenonympha pamphilns, said by Assmuss to winter
in western Russia in the egg. This butterfly is double-70
The Life of a Butterfly.
Small as it is, the newly hatched cater-
pillar has almost equal peril, and from the
same larger enemies — spiders, crickets,
and bugs — which have threatened the
egg. Edwards, in the passage quoted,
adds: “ And in the same way I have
lost scores of small larvae.” It is highly
probable, as suggested in a previous chap-
ter, that this is the reason why the young
caterpillar ordinarily devours nearly or
quite all the shell it has left the moment
it enters upon free life; for by removing
this indication of its proximity, the cater-
pillar indulging in such a propensity is the
more likely to live to leave descendants,
and thus a habit may become fixed. But
it is perhaps in greater peril, because to
obtain its food it must move, and a
brooded in England, — the first brood appearing, ac-
cording to Buckler, in May, impossible to have come
from a wintering egg; indeed one instance is on record
of pupation on April 5. It is double or triple brooded
in Silesia, according to Prittwitz, and the caterpillar is
full grown at the beginning of May. It must surely
hibernate as a caterpillar, as Schwartz claims.The Critical Periods of its Life.	71
moving object, especially to insect vision,
is more apparent than one at rest It
satisfies its appetite therefore at the risk
of its life, and after its meal remains
rigidly still.
It is probable, however, that it is at its
next stage, after once moulting, that the
caterpillar’s danger is at its greatest. For
with its growth in size, it has now become
a mark for a new class of enemies, the
ichneumon flies and tachinid flies, which
nourish their young within its body.
Our Milkweed Butterfly is certainly more
exempt from danger from them than is
perhaps any other of our native species,
and was long thought to be totally free
from them, but we already know of two
kinds of four-winged ichneumon flies —
a larger and a smaller — and one kind
of two-winged tachinid flies which prey
upon it
I believe that it is at this early stage in
their life that caterpillars are ordinarily
stung, giving the worms hatched within72
The Life of a Butterfly.
time to prey their life long upon the
nourishment meant for the bodies of their
hosts. How often caterpillars taken at
large and brought to the breeding cage to
rear yield these concealed villains, only
those who have tried it realize; nor how
many such harpies may be carried within
one caterpillar and the caterpillar still
thrive, — a peripatetic banqueting hall for
unbidden guests!
When larger the caterpillars have the
birds, and if they be low feeders, the
reptiles to contend with; but though one
of our ornithologists says he has found
the caterpillars of our commonest yellow
butterfly in the stomachs of as many as
twenty different kinds of our native birds,
it is not probable that the destruction by
birds is nearly as great as by insect foes.
Against the birds they have only their
form and color to oppose, through which
by their resemblance to their surroundings
or to some uneatable object they may
escape detection or be left undisturbed;The Critical Periods of its Life. 73
and this resemblance is often more strik-
ing than one would believe, seeing them
shut up in a naturalist’s box or prepared
for the cabinet. To gain any conception
of it one must see them, if he can, in their
natural haunts, and he will discover either
how hard they are to find or what grotesque
or inanimate things they appear to be.
This is not at all the case with the
subject of our sketch, for the plump cater-
pillar of the Milkweed Butterfly, though
when at rest it usually conceals itself
beneath a leaf to which it clings upside
down, is a most conspicuous object, with
its transverse belts of black upon a
white and yellow or orange ground; but
in this case, and probably in a few others,
such as the almost equally conspicuous
caterpillars of our Swallow-tails, we are
brought into that class which owes its
immunity to attack to the nauseous quali-
ties of its members, which these “ warning
colors ” by their very conspicuousness
appear to proclaim.7 4
The Life of a Butterfly.
One would think, at first, that one
danger to caterpillars might be in finding
proper food supply, especially if it were
known how very particular many kinds of
caterpillars are, — starving to death unless
they can find the single species on which
they feed. Doubtless this is an occasional
cause of death. I once found several
eggs of Anthocharis genutia on a slender
Arabis which would not have yielded
enough nutriment for one caterpillar, and
after an hour or more of search — and I
was very eager to find more eggs — could
not find another stalk of that or any
similar cruciferous plant (on which alone
the caterpillar is supposed to feed) within
an area of several acres. But such cases
are rare, and it is to the parent the credit
is to be given, who with careful search
almost invariably selects the very choicest
spots for the progeny that shall spring
from her eggs.
A more obvious danger is to be found
in the necessity for those repeated slough-The Critical Periods of its Life. 75
ings of the skin which its growth neces-
sitates; at such times the caterpillar is
completely helpless, generally incapable
of more than a feeble wriggle, and
continues in this condition for hours,
sometimes for days at a time. Sometimes
it seeks special concealment for the
change, but in a great many cases, strange
as it seems, ecdysis occurs in plain sight
of all, upon the surface of a leaf; it has
then only its immobility to protect it.
A similar critical period is found during
the change to chrysalis, which generally
occupies a day or two; this frequently
occurs under protection of some con-
cealment, though this is less often the
case among the brush-footed butterflies,
to which our milkweed species belongs,
than with others. But when the change
has taken place and the at first flabby
chrysalis has gained its form and color, the
matter at most of but a few hours, there
is rarely an instance where resemblance
or color-harmony with its surroundings76
The Life of a Butterfly.
is not a very effective protection; and in
many the angularities of the surface, like
the spines of many caterpillars, may be
an additional source of safety. It would
seem, indeed, that having fairly reached
the chrysalis state, there was relatively
little danger until it was upon the wing.
It is true that many chrysalids give forth
not the butterfly, but two-winged or four-
winged parasites, but these are a legacy
from the caterpillar and not an independ-
ent production; these insect parasites are
not known to sting the chrysalis.*
It is usually but a meagre remnant
that passes through all these stages to
* I had supposed this to be absolutely true, but
Howard, in his excellent account of the biology of one
family of these parasites (Proc. U. S. nat. mus., xiv.
569), says: “The pupa [of Lepidopteraj itself is sel-
dom attacked, yet certain of the Pteromalines which
preferably oviposit in larvae about to transform will
also lay their eggs in just-formed pupae. The same is
the case with certain members of the genus Chalcis,
particularly those parasitic upon diurnal Lepidoptera,
and I am not sure that C. flavipes does not oviposit by
preference on the fresh chrysalids of Chlorippe clyton
and Agraulis vanillaeThe Critical Periods of its Life.	77
finally attain the winged state. Probably
in the Milkweed Butterfly the remnant is
exceptionally large, but with the mass of
our butterflies it seems extremely doubtful
whether five per cent of the embryos
which are deposited in their egg-shell
ever attain their majority; and when these
have gotten them wings and fly away,
new dangers interpose. There must be
more or less delay about laying the eggs;
in many species the eggs are not ripe to
lay for several days, in some apparently
not for weeks; suitable places must be
chosen, and in nearly all instances the
eggs must be laid one by one, and are
laid on different plants; all of this of
course takes time, and only in sunshine
can they or do they so occupy themselves.
Meanwhile there come the storms which
destroy them in large numbers, and the
birds which snatch them up; especially
do the more sluggish females suffer, heavy
with their burden of eggs. It seems quite
safe to say that not one half the possible78
The Life of a Butterfly.
eggs of a given brood of butterflies ever
get laid at all.
And so it comes about that for every
pair of butterflies a given season, with its
potentiality of a brood of hundreds, the
next season witnesses but still one pair.
The balance of Nature is kept up. Yet
now and again there are fluctuations, beats
of the biologic pendulum. One year vast
numbers of one kind are seen, which
was scarce or almost not to be found the
very year before. This is probably to be
explained by the fact that their scarcity
the first year meant paucity of food
to their parasites, which consequently
starved, and were not present in sufficient
force the second year to make their usual
inroad upon their numbers. Just so does
the entire animal world show a struggle
between herbivore and carnivore for the
occupation of the ground. “ Increase and
multiply ” is the primal law of each.VI.
A Favored Race: Mimicry and Protective
Resemblance.
ILEY tells us that in Missouri
the tachinid parasites of the
Milkweed Butterfly, its most
meddling insect foe, are sometimes so
active that not one in fifty of the
caterpillars escape them; yet it will, I
think, be agreed by all observers that as
compared with the caterpillars of other
butterflies it enjoys a very considerable
immunity. Is it possible that this is
simply because it is not attacked by
birds or toads? Its “warning” colors, at
any rate, are more likely to be the result
of immunity from the attack of vertebrate
than of insect foes, and the former are of
a relatively greater importance in theSo
The Life of a Butterfly.
tropics, which it must not be forgotten
is the real home of this insect, than in the
temperate region.
Nor should it be overlooked that the
three most striking insects which — in New
England at least—ordinarily accompany
the caterpillar upon the milkweed, the
caterpillar of the moth, Euchaetes egle>
the longicorn beetle, Tetraopes tetraoph-
thalmuSy and the chrysomelid, Doryphora
clavicollis, are equally conspicuous and
have similar coloring, a contrast of black
and orange in broad markings. Have
they, perchance, gained these colors by
unconscious mimicry? Why should four
such striking insects with similar colors
preponderate upon the milkweed?
However this may be,—whether or no
the gayly banded caterpillar and the gold-
and-black-dotted pea-green chrysalis of
the Milkweed Butterfly are specially pro-
tected from vertebrate foes, there can
be no doubt that this is the case with the
perfect butterfly. Indeed all the membersA Favored Race.
8t
of this group of insects the world over,
so far as known, have a more or less
rank, in our species a carroty, odor; and
some of them contain a pungent fluid
which will exude under pressure and stain
the skin. They are evidently malodorous
or nauseous to birds, and their easy
movements on the wing are an added
proof that they are rarely or never then
attacked.
One observer in Brazil, watching a pair
of puff birds catching butterflies to feed
their young, noticed that the butterflies
of this group were never attacked, though
flying lazily about in great numbers.
Their abundance wherever found — and
they frequently swarm — is another silent
evidence of their safety on this score;
and since the statement was first made,
thirty years ago, that they were probably
proof against attack on the ground of
their unpalatableness, it has, I believe,
never been denied by any observer, and
has been so many times verified and con-82
The Life of a Butterfly.
firmed that there is now general consent
to the proposition.
But this is only the beginning of the
matter. Here again we meet with the
old story of “ the struggle for existence/'
Life is such a warfare that artifice and
deceit are to be looked for on every
hand. Given a favored race, it will be a
gain to their unfavored persecuted friends
to resemble one of them, and so a prize —
the prize of a better hold on life — is
offered for a forgery. It was an as-
tonishing fact that was brought to light
by Bates, — that a group of butterflies
occurred in Brazil of vivid coloring and
slow and easy flight, which were the
constant subjects of mimicry by butterflies
of quite a different type, normally white
and tolerably uniform in color, but which
had so changed their livery, and even the
form of their wings, as closely to resemble
the objects they mimicked in brilliancy
of color and variegation, and even in
mode of flight.A Favored Race.
83
Some, says he, “ show a minute and
palpably intentional likeness which is per-
fectly staggering.” Indeed, the likeness
proved so close that even after he became
aware of the mimicry his practised eye
was often deceived. Or if he wandered
to a new locality, where occurred a new
set of Ithomyiae (the most numerously
represented among the mimicked genera),
the Leptalides (the mimickers) would
vary with them so as to preserve the
mockery, band for band and spot for
spot. It is now known that wherever
these protected butterflies are found, the
world over, they are accompanied by a
set of mimicking forms of some totally
different group. We have in our own
country two such instances of a con-
spicuous kind, one species of Basilarchia
(Nymphalinae) mimicking Anosia plexip-
pus (Euploeinae), and another species
of Basilarchia mimicking Tasitia berenice
(Euploeinae).
The fact of a resemblance so close that84
'The Life of a Butterfly.
it is to all appearances a “ palpably in-
tentional likeness ” is impossible to ques-
tion. But how explain it? How can a
butterfly change its appearance to such
a degree? “ Can the Ethiopian change
his skin or the leopard his spots ?”
The answer, as Bates clearly saw, was
to be looked for in the same direction
as when accounting for the assumption
by animals of the color of their surround-
ings. Both are produced in the same
way, and have the same cause and end.
It is only by keeping in view this tolera-
bly obvious truth that we can explain all
the freaks of mimicry. “The specific,
mimetic analogies,” says Bates, “ are
adaptations, — phenomena of precisely
the same nature as those in which insects
. . . are assimilated in superficial ap-
pearance to the vegetable or inorganic
substance on which or amongst which
they live.”
To gain an idea, then, of the processes
by which the “ staggering ” examples ofA Favored Race.
85
mimicry are produced, we must look first
at the simplest forms of protective resem-
blance. Go to the seashore and observe
the grasshoppers among the beach grass.
They fly up at your approach, whiz off a
rod or so, and alight. Can you see them?
They are colored so nearly like the sands
they live upon that detection of one at
rest is almost impossible unless one has
seen it alight. On yonder grassy bluff,
a stone's throw away, you will find none
of them, but other kinds equally, or
almost equally, lost to sight by their
harmony with their surroundings. What
chance of life for either if they suddenly
changed places ? They would be so con-
spicuous that every passing bird or other
insectivorous creature would sight them.
Of course these protective colors have
been gained by very slow steps. Every
grasshopper that lived by preference
among the sands was liable to be eaten.
In the long run just those would be eaten
which were most easily seen. One which86
The Life of a Butterfly.
varied in coloring in never so small a
degree, so as to be less easily seen than
his brother, would live to perpetuate his
kind, and his brother would come to an
untimely end; the progeny would show
the fortunate variation, and be more likely
to be spared to transmit in increased
volume the probability of the happy
coloring.
Given, then, a brood of grasshoppers
that find their preferred food in sandy
spots, and unless other and more powerful
forces act upon them, it must result, from
their liability to be eaten by creatures
fond of grasshoppers, that in time they will
resemble in coloring the sand on which
they live, — it is impossible that they
should not. Any creature not specially
protected by nauseousness or habit or
special device of some sort must in the
very nature of things, if it is to live at
all, have some other protection; and that
afforded by color and pattern is by far
the most common. The world is madeA Favored Race.	87
up of eaters and eaten, of devices to catch
and devices to avoid being caught.
We may apply the same reasoning to
two kinds of butterflies subject naturally
to the same class of enemies; that is,
living in the same region and flying at
the same time. If one has the slightest
advantage over the other in the fight for
life, by being, for instance, distasteful to
one class of common enemies, so that
these forbear to attack it after experiment
or by instinct (the result of ancestral ex-
periments), and there be among the less
favored flock, here and there, an individual
which, under circumstances favoring it,
such as distance or shadow, may more
often than its fellows be mistaken by the
enemy for one of its distasteful neighbors
through its possession of a little more
than usual of a certain tint on a part of
the wing, a little larger spot here, or more
of the semblance of a band there, — how
small soever this difference may be, it
must, by the very laws of natural selection,88
The Life of a Butterfly.
be cherished, perpetuated, increased, by
slow but sure steps. Nor is there any
limit to its increase except its absolute
deception of the enemy. So long as there
is the slightest advantage in variation in
a definite possible direction, the struggle
for existence will compel that variation.
Knowing what we now know of the laws
of life, mimicry of favored races might
even have been predicted.
It would seem, then, to be plain that all
cases of protective coloring and mimetic
form come under one and the same law,
and have been produced by the same
means (the survival of the best mocker),
whether the object imitated be animal,
vegetable, or mineral. The actual out-
come is, indeed, vastly more surprising
in some cases than in others, — in some
“ perfectly staggering,” as Bates says; yet
though there be to all appearances a
“ palpably intentional likeness,” there is
found to be no intention in the case so
far as mocker and mocked are concerned,A Favored Race.
89
but the result of a natural selection against
which neither could even strive, and of
which neither was ever conscious.
The process has been a long one, so
that in the case of parastatic mimicry, as
that form which involves the copying of
one’s fellows might be termed (or if
one prefers an English term, neighborly
mimicry), we may readily presume far less
difference between mocker and mocked
when the mimicry between them first
began than now exists between the
mocked and the normal relatives of the
mocker. It is argued, indeed, with great
show of reason, that as the resemblance
grew stronger the birds became more
sharpsighted, which reflected again on the
mimicry, and that thus the final departure
from the normal type was intensified; but
this assumption is not necessary.
It is to be presumed that the actual
colors found in a mimicking butterfly
are, with rare exceptions, such as existed
somewhere in the ancestral form. In the90
The Life of a Butterfly.
case of our own mimicking Basilarchia,
for example, whose orange ground tint
is so totally at variance with the general
color of the other normal members of
the group, it will be observed that all
the normal species possess some orange.
Without this as a precedent fact, such
perfect mimicry might perhaps never have
arisen. Individuals among the normal
species vary somewhat in this particular,
so that it is easy to suppose that some
of the original B. arcliippus, with more
orange than usual, may have escaped
capture, on occasion, from this cause.
From such a small beginning, such as one
may now see every year in B. astyanax,
sprang doubtless the whole story, and at
last we have a butterfly which has for a
ground color of both surfaces of the wings
an orange which is the exact counterpart
of that of Anosia plexippus; by reason
of which, in all probability, it enjoys a
freedom from molestation comparable to
that attributed to A. plexippus, so that itA Favored Race.
9i
ventures more into the open country than
its allies, and thus gains a wider pasturage
and surer subsistence.
It is not necessary for our purpose to
enter here into further details about the
various forms and phases of mimicry;*
it is sufficient to point out that in some
instances only one sex, and that the
female, departs from the livery of its kind
to mimic that of its pattern; that in
others a normal male of the mimicking
group may have several forms of female,
one mimicking one, another another, of
the protected type; that some butterflies
of the protected group are mimicked by
others within the same group; and that
butterflies of other than the protected
group are the subjects of mimicry.
Of this last we have one of the most
* Those wishing to pursue this matter further are
referred to my paper on the subject in the Atlantic
Monthly for February, 1889, printed also in my “Butter-
flies of the Eastern United States,” as Excursus xxiii.
pp. 710--720, to which latter will be found appended a
bibliography of the subject.92
The Life of a Butterfly,
curious cases in our own country. One
of our fritillaries, Semnopsyche diana, is
remarkable for the great difference in
coloring between the male and the female:
the male would be taken at once for a
fritillary from the general tone of color-
ing both above and below; but at first
glance one would say the female was a
Basilarchia (a butterfly of quite a different
tribe), from its close resemblance to B.
astyanax. Indeed, when first seen alive
by Edwards, then on the hunt for the
species and ignorant of the female, he
actually thought it a species of Basilarchia
near B. astyanax. Its range in the South,
too, is completely covered by that of
B. astyanax.
The fact of mimicry in this instance
must be regarded as unquestionable; but
the strangeness is the greater that in our
country it is just in this genus Basilarchia,
here mimicked, and in no other, that
striking mimicry of the “ favored tribe ”
occurs. It is a case somewhat parallel toA Favored Race.
93
mimicry within the favored tribe, but the
more striking that the cases mentioned
stand alone among all the American
genera. Mimicry is far more common
(even relatively) in the tropics than in
the temperate regions; so, too, are, in-
sectivorous animals. And it should be
remarked that if we are correct in our
opinion that our Milkweed Butterfly is,
in the strictest view of the matter, an
autochthone in our country only in the
extreme South, then it is easier to explain
the mimicry of Basilarchia, for there
would not seem to be any occasion for
mimicry of B. archippus in the higher
North. I have myself never seen a but-
terfly (but often a moth) pursued or
snatched by a bird, though I once came
across the evident signs of capture in the
scattered wings of Euphoeades troilus on
a damp roadside where I had just missed
its coursing back and forth. Mimicry in
the perfect stage is only needed where
their vertebrate foes are numerous and
pressing.94
The Life of a Butterfly.
When we take a general view of mimicry
as exhibited by one butterfly for another,
how strange it seems ! and what an inter-
esting illustration it is of the adaptability
and pliancy of natural forces that for the
evident protection of one species in the
struggle for existence so exact and beau-
tiful a resemblance should be brought
about! Consider for a moment that the
subjects of mimicry are at the final stage
of life; they have already passed through
nearly all the dangers to which the
species, as a species, is subjected, — so
rudely subjected that they are a mere
remnant of those brought into the world
with them. During the early period of
their life they were exposed to vastly
more dangers than they can now experi-
ence. At times they were absolutely
helpless, without the power of movement.
They are now endowed with powers of
flight sufficient to thwart the purpose of
many a foe; yet it is in just this period
that these special and extraordinary pro-A Favored Race.
95
visions for their safety — and for the
accomplishment, so far as the species is
concerned, of the end of their life — are
given them. All this has been brought
about for the sole purpose of prolonging
their aerial life for the exceedingly few
days which are necessary for pairing and
the deposition of eggs. The more we
contemplate so strange and perfect a
provision, and the means by which it is
accomplished, the more are we impressed
with the capabilities of natural selection,
and begin to comprehend how powerful
an element it has been in the development
of the varied world of beauty about us.VIL
Scent-Scales: a Question of Sexual Selection.
ERHAPS it will be remembered
„that in describing the Milkweed
Butterfly in our first chapter, the
male was said to be distinguished by the
presence of a small but conspicuous black
pouch or pocket, like a swelling or blister,
on one of the veins of the hind wings. It
is so heavily clothed with scales that we
hardly notice that it is a pouch, open by
a narrow slit on the upper surface of the
wing on the side farthest from the vein
to which it is attached. What does it
contain? And what is it for?
These questions open to us one of the
most curious, one of the most interesting,
and yet one of the least studied chapters
in the life and structure of butterflies.
The interior of the pouch is filled withScent-Scales.
97
scales attached to the membrane, of which
there are two kinds, differing in form and
in setting: one kind is long oval, with
uniform margins, and which differs from
the scales exposed on the surface of the
wing only in being slenderer and in not
being toothed or notched at the tip; the
other kind, however, can scarcely be
called scales, for they are more properly
rods, seated in the centre of little raised
rings which are quite absent from the
base of the other scales, whether within
or without the pouch. The pouch is
evidently for their protection, and we
naturally ask what they are for, that they
should be so carefully guarded.
To answer this inquiry, we shall have
to go further afield, and see what other
butterflies may have comparable with
these; and we may well be astonished at
the revelation that will be forthcoming,
for we shall discover these singular scales
in every direction and in every form,
but usually concealed in major part or
798
The Life of a Butterfly.
entirely, either by the other scales or by
folds of the wing membrane.
Sometimes they differ but little from
ordinary scales, or may be lopsided; at
others, they assume the form of a battle-
dore and are studded with beads; in many
they terminate in a longer or shorter
fringe of the most excessive tenuity, quite
invisible without a tolerably strong power
of the microscope; very generally they
are exceedingly slender and even then
may be fringed at the tip; or they may
assume the form of hairs shaped like a
shepherd’s crook, or a straight rod with
a whip-lash at the end; or a twisted
ribbon or a chain of links; or they may
be spatulate or even fan-shaped at the
extremity. The range, indeed, of their
form is strikingly at variance with their
concealment and microscopic size, and
they are infinitely more varied than the
exposed scales, which are almost always
toothed or combed at the tip, which is
rarely or never the case with these.Scent-Scales.
99
They are sometimes scattered more or
less sparingly over the upper surface of
the wings, all but their tips concealed
even from the microscope by the over-
lapping of the ordinary scales; but they
are generally gathered more or less
abundantly into clusters on either surface
of the wings, ordinarily on the upper, in
definite patches visible as a sort of cloud-
ing to the naked eye, or along veins
which they then appear to thicken. In
nearly all, if not all, of these cases the
visibility to the eye does not, however,
result from the multitude of the special
scales, but from the exceptional develop-
ment and varying color of the attendant
normal scales, which are either larger,
darker, or more opalescent than their
neighbors, or are raised or turned at an
angle which catches the eye by the gen-
eral effect; it is much as if the special
scales we are investigating were playing
hide-and-go-seek among the others, cry-
ing “ here we are ” by signs, but still
hiding from view.IOO
The Life of a Butterfly.
In other cases these patches are con-
cealed by being placed where the front
and hind wings overlap; or, as in the
case of our Milkweed Butterfly, a special
membranous pouch is developed for their
benefit. In one great division of the
Skippers the membrane of the front
margin of the fore wing is expanded and
folded over upon itself, but so deftly that
it is often hard to see that this is the
case; and in many of the allied Swallow-
tails, the plaited part of the hind wings
next the abdomen will be found put to
service in a similar manner; while the
space within, in both these instances, is
so crammed with these special scales that
when the folds are opened the scales may
be removed en masse as a bunch of silken
floss.
The study of these scales has only just
begun, and the above gives only a faint
notion of the wonderful variety in form
and curious disposition so far seen; but
we have learned one thing at least whichScent-Scales.
IOI
gives a clew to their meaning and pur-
pose. They are absolutely peculiar to the
male sex, just as much so as that whole
class of structures known as secondary
sexual characteristics, upon the presence
of which Darwin bases his theory of
sexual selection, — the one prime feature
of his philosophy not subscribed to by
his colleague in the introduction of the
modern doctrine of evolution, Wallace.
On account of this restriction to the
male sex they have been termed andro-
conia, or male-scales; and Wallace may
well employ them as an argument upon
his side, for while they will not yield the
palm to other secondary sexual features
in variety and beauty, they are so placed
that with the exception of such cases as
the folds of the hind wings of Swallow-tails
(the wider expansion of which may open
the fold and expose the androconia), they
are concealed either absolutely or all but
their tips, where at the most nothing but
a fringe or an edge is exposed, and the102
The Life of a Butterfly.
C 0
) I
size of this so infinitesimally small that
the wildest adherent of the hypothesis
of clear vision by insects would hardly
regard vision as a possible factor in thp
a r J ■ • Vv**7 .-<' < *■■ ■ *'	- ,	; V'
case.	r .	/
S t. v-.. ^ ^	A f	•
The patches of accessory scales might
be regarded differently; but these are not
always present, and an explanation must
cover both cases, whence the presence of
absolutely concealed androconia in as great
a variety of form and position as that
of other secondary sexual characteristics
must be regarded as a fatal blow to
. /
Darwin’s view of sexual selection. Other-
wise there should be no need of his
efforts to show how the male displays
these particular charms to his consort at
the breeding season.
But we know more about these andro-
conia. About fifteen years ago, Fritz
Muller, a naturalist who by his researches
in various fields has done much to bring
new evidence in support of the theory of
evolution, astonished naturalists with aScent-Scales.
103
long list of odors emitted by butterflies
and moths; and among the sources of
these odors he claimed various patches
of scales, which are nothing more nor
less than those where androconia lie
concealed, as he has pointed out in a
series of papers where these are figured
and termed scent-scales.
This view of their function has been
found more plausible since minute canals
have been traced through certain andro-
conia, canals which in some cases continue
through and open at the extremities of
the delicate threads which form their
fringe; and also since Weismann has
clearly shown that there is a living tissue
in the wings which would allow of the
production of an odor through local active
scent-glands. Moreover the number of
instances of aromatic butterflies in which
the odor is referable to the androconia
is constantly increasing, although as this
odor is lost after death, observations on
this head can only be made in the field.104 Tbe Life of a Butterfly.
A few instances may suffice, all of which,
it must be remembered, are confined to
male butterflies.
Antirrhea emits a strong odor from a
patch of scales near that part of the
hind wings, which is covered by the fore
wings, and which is further protected by
a curving mane of pale buff hairs; a faint
odor “ resembling that of sable fresh
from the furrier’s shop ” is given out in
Stichophthalma by a patch on the hind
wings accompanied by an erectile wisp
of hairs; a black spot of scales near
the base of the under side of the fore
wings in Didonis gives out a musk-like
odor.
In our own fauna we have a striking
instance of this odor in the scent emitted
by the scales clustered along the median
nervules of the upper surface of the fore
wing in Argynnis atlantis, — scales which
have a distinct odor of sandal-wood, so
strong that it is hardly possible to handle
living specimens without recognizing it,Scent-Scales.	105
and which I have known to be retained
for many weeks after death, when the
insect had been enclosed at capture in a
paper envelope. This is the more remark-
able because I have never detected the
same or any odor in the two allied spe-
cies of Argynnis of New England, which
nevertheless possess precisely the same
scales, and in the same position. Finally,
we have the instance of our Milkweed
Butterfly; the scales found in the little
pouch which we are considering emit a
slightly honeyed odor over and above the
carroty smell which all the scales possess.
All these are instances from a single
family of butterflies.
In other families may be mentioned
our common blue butterfly of the spring,
Cyaniris pseudargiolus, which has an ex-
ceedingly delicate odor, which I can only
describe as that of newly-stirred earth
in the spring, or of crushed violet stems;
our large Callidryas eubnle, which has
a slight violet odor;	where the106 The Life of a Butterfly.
upper side of the wings, especially if
parted after having been closed some
time, gives a rather faint but “ very deli-
cious perfumeour white butterfly,
Pieris oleracea, where a faint odor of
syringa blossoms may be detected; the
European P. napi, said by different writers
to have the odor of thyme, verbena,
orange, or balsam; Papilio polydamas,
which, according to Muller, has “ two
sets of males emitting equally strong
but quite different odors,” — a case of
diosmism!
These instances, with the many others
known, clearly prove that very many but-
terflies emit odors, apparently in most
cases agreeable to us, which it is in
the highest degree probable are emitted
through microscopic canals, which course
through microscopic scales to micro-
scopic glands at their base within the
wing membranes ; that while these are in
some cases easily perceptible by man, he
is unable to detect any odor whatever inScent-Scales.
107
their very next neighbors, which possess
the very same apparatus. The only ex-
planation of this which appears plausible
is that odors exist imperceptible by us
but perceptible by them, an explanation
which requires us to attribute to these
insects an exceedingly delicate and high
perception of odors.
This requirement will hardly appear to
present the remotest difficulty to persons
who have witnessed what is termed the
“assembling” of moths; that is, the habit
of the males of many species, notably
of the Bombycidae, to collect in swarms
around a female that has been disclosed
from its cocoon entirely out of sight of
and often at a great distance (sometimes
miles) from her visitors, — in a dwelling,
for instance, in the middle of a great city.
Instances of this sort are numerous and
well known, and compel us to admit a
power or a delicacy of sense perception in
the direction of smell which opens the
door to many mysteries in the under-108 The Life of a Butterfly.
standing of the lives of these humble
creatures; they do not so much “walk
by sight” as fly by smell!
This, then, is the probable function of
all these androconia, as we know it to be
the property of some, — a property which
seems founded in and not accessory to
their structure; we may easily believe
that the concealment of the androconia
is to prevent the too rapid dissipation
of the odor, to offer the opportunity for
its sudden flush, — in other words, to
place it under some control.
The exquisite and varied form of the
androconia is another matter. They dif-
fer marvellously from ordinary scales in
the variety of their form and exquisite
structure. Ordinary scales seem made on
a single pattern; they are minute enough,
but they are huge as compared with most
androconia. Why such delicate and ex-
quisite patterns on such a microscopic
scale? Who is to see and enjoy them?
Assuredly not the butterflies themselves;Scent-Scales.
109
they may profit, indeed, by their function,
and no doubt natural selection has per-
fected that to an extraordinary degree,
well beyond our ken; but the androconia
must practically be quite invisible to
them. Is there not here, then, a beauty
of form and of structure which is an
end in itself, of no possible profit to the
possessor?
[Readers wishing to look into this subject more fully
will find it discussed in detail in my “ Butterflies of the
Eastern United States” (see particularly Excursuses
xvi. xxxi. xl. lxvii. and Ixx. and plates 43-51), where
also references will be found to the literature of the
subject.]VIII.
Insect Vision. Does Sight or Smell most
control the action of Butterflies ?
N the last chapter we pointed out
the existence in many male
butterflies of structures of extra-
ordinary variety, and considering their
microscopic size, of remarkable complex-
ity, which are so hidden from sight that
they can rarely or never be seen by their
neighbors, even if we credit the latter
with exceptional powers of vision. A
lurking feeling may yet exist in many
minds that they may, after all, be thus
endowed; and we therefore depart a
moment from the general method em-
ployed in the present work of setting
forth the life of a single butterfly, toInsect Vision.
in
inquire whether it may or may not be
true that insects, our Milkweed Butterfly
of course included, have remarkable
powers of sight, for then we shall be
better able to interpret many of their
lines of conduct.
Vision in insects with compound eyes
has been a subject of discussion for very
many years, and the views of naturalists
regarding it have differed from time to
time very considerably. The question
has recently received much new light
from the anatomical investigations of
Patten and others, and the experimental
researches of Forel and Plateau. The
compound eyes of butterflies, as of other
insects, are made up externally of a num-
ber of adjoining hexagonal facets, each
separate facet being the exposed portion
of a crystalline lens which is followed
beneath b.y a slender tube containing,
first, a terminal body, the crystalline cone
or retinidium, — formerly looked upon as
a second refracting medium, but by Patten112 The Life of a Butterfly.
regarded as a retinal body sensitive to
the light, — and second, of a collection
of rods.
The office of this so-called crystalline
cone is the principal point in dispute.
The later researches regard it as a recep-
tacle for the termination of the nervous
elements, and as performing no office in
either modifying or destroying the image;
while the old view regarded it as having
similar properties to the crystalline lens
of our own eye. Now if the later view
be regarded as correct, which there seems
every reason to believe, the form and
nature of this receptacle is such, as
Plateau points out, that an image may
be formed at any point within its depth;
but at the same time all the sensitive
points of the cone in advance of or behind
an image will be illuminated and will also
to some degree be excited by the same
object, so that whatever image is formed
can in no way be seen as distinct, but
only as entirely confused, much as hap-Insect Vision.
113
pens in the human eye when the image
is focussed beyond the retina.
This theoretical view can be perfectly
well subjected to experiment; and this
has been done in the most thorough man-
ner by the Belgian naturalist, Plateau.
At first his experiments were made almost
entirely by placing the insect desired to
be experimented upon at one end of a
closed compartment at whose other end
were two distinct openings to the light,
— one simple and large enough for the
escape of the insect, the other much larger
but covered with a trellis or grating, so
that while the actual superficial area of the
open spaces might be the same in each
case, admitting the same quantity of light,
the trellised opening would appear greater.
At the same time one could modify at
will the amount of light which would
enter either of these two different classes
of openings. The insects almost invari-
ably flew to the trellis. Numerous experi-
ments were made by Plateau upon this
8ii4
The Life of a Butterfly.
basis, resulting in his concluding that
insects with compound eyes did not
well distinguish between two illuminated
openings, being sometimes led astray by
the excess of luminosity, sometimes by
the apparent excess of surface. In gen-
eral, he thought they could not dis-
tinguish the form of objects, or only in
a vague way.
Objections were raised that in these
cases it was hot objects, but luminous
openings, the power of seeing the form
of which was tested; and also that the
judgment of the insect was brought into
service under unnatural conditions, so that
the experiments proved nothing very
definite or decisive regarding the actual
power of vision on its part. Thereupon
Plateau devised a new method by which
experiments could be made, obviating all
these objections, and has carried out
these experiments even more carefully
and extensively than the previous, insti-
tuting, moreover, comparative investiga-Insect Vision.
US
tions with the vision of vertebrates under
precisely similar circumstances.
To do this he constructed what he has
termed a labyrinth, — a table covered with
a large number of vertical barriers placed
in concentric series in such positions that
the creature must take a very circuitous
course among them to escape. Of course
only crawling insects could be experi-
mented on, but the structure of the eyes
is essentially the same in all. Care was
taken that the coloration of the surface
should be of a neutral tint, and the
vertical barriers were painted of various
colors, — white, brown, or black; also that
the animal should begin its movements
without excitement, by the quiet lifting
of a glass cover under which it had been
placed, near the centre of the labyrinth,
when it would find itself surrounded
by walls between which were abundant
passages.
Where the vision was really good, the
animal would be expected to move in aii 6	The Life of a Butterfly.
serpentine course between the obstacles,
never striking them, usually moving
toward the opening which was largest
or nearest. When, however, the power
of vision was to some extent defective,
so that the animal seemed to be aware
of the existence of an obstacle only
when it had almost reached it, then the
movements should be made in a series
of zigzags which would change direction
somewhat sharply, shortly before the
barriers were reached. Or if the vision
were absolutely defective, the creature
would be likely to find its way only by
first striking the objects and then moving
around them.
The result of his experiments proved
that vertebrates had complete vision,
directing their movements with ease
without striking any of the obstacles,
moving in the nearest path; while insects
acted in all cases as if they had a
veil before their eyes, — their change of
direction before reaching the barriersInsect Vision.
ii 7
being such as to indicate that it was
only when they reached the shadow of
the obstacles before them, when they
could distinguish some difference in the
intensity of light, that they turned aside
to avoid such obstacles.
The result of these experiments has
been so uniform and so clear that how-
ever the actual anatomical structure of
the eyes of insects may be regarded,
there can be little doubt that their vision
is so extremely imperfect that they per-
ceive sharp images of nothing immobile,
and therefore do not distinguish the
precise form of objects, though they can
readily distinguish objects in motion.
Indeed, they are particularly keen in this
sort of vision, so that their sight must
be best while upon the wing; for in this
case it is the same as if they were at
rest while all the objects about them were
moving. So too, they can distinguish
masses of color, but not pattern, except
in the vaguest way. The comparative118 The Life of a Butterfly.
experiments revealed an enormous differ-
ence in the behavior of vertebrates and
insects under precisely identical circum-
stances, — the former acting as if they
possessed human vision, the latter as
if they could distinguish the form and
boundaries of objects in a most imperfect
way, at the best.
“A flying insect,” says Plateau at the con-
clusion of his various experiments, “has a
very lively perception of light and shadow, so
that without distinguishing as we should do all
the details of its route, it knows how to avoid
all masses, such as the trunks of trees, bushes,
rocks, walls, etc., and passes them at a con-
venient distance. Caught from any cause in
a mass of shrubbery or any other group of
vegetation, it takes advantage, when it wishes
to pass on, of those passages through which
the greatest amount of light filters, or as
between two equal in this respect, of such as
seems to it to offer the most room. If the
wind move the leaves, these openings may
oscillate, but thanks to its good perception ofInsect Vision.
119
movements, the insect can then see them
better. In flying, the insect moves in undula-
tions so as to follow the direction of the
displacements and to traverse the openings
without striking.
“When its mode of sustenance necessitates
a visit to certain flowers, it moves toward them
either with certainty, in being guided by its
sense of smell only, if its power in this direction
is well developed, or by chance if its olfactory
powers are relatively slight. Incapable of dis-
tinguishing by their forms flowers of the same
color, it goes directly toward the colored spots
which to it mean corollas or inflorescence,
turns, hesitates, and does not decide what to do
until the distance has become sufficiently slight
to enable it to determine by the odor whether
or not it has found what it seeks.
“The sense of smell only, or this combined
with visibility of movements, assures the meet-
ing of the sexes; and finally it is the perception
of movements which warns of the approach of
an enemy and permits escape in time.
“This brief statement is sufficient to show
how an insect with facetted eyes, though it120
The Life of a Butterfly.
only has a confused visual perception of objects
at rest, frequently acts in a manner to suggest
to one who does not closely analyze the
phenomena that the eyesight of these creatures
is as distinct as that of the vertebrates.”
It may be added that in his experi-
ments Plateau found that the family of
butterflies called Lycaenidae, of which
our tiny blue butterflies may be taken as
a type, were disturbed by a moving object
only when at half the distance at which
one of the Nymphalidae (like our Milk-
weed Butterfly) would take alarm. Now,
if we examine the eyes of these two
groups we find a singular difference which
may account for this: each of the eyes of
the Nymphalidae covers the continuous
surface, roughly speaking, of half a
sphere; in the Lycaenidae the posterior
half of this half-sphere entirely lacks any
facets; indeed their ocellar globe has
relatively the smallest visual surface of
any butterflies; this fact at once explains
their tremulous movement before alight-Insect Vision.
121
ing on an object; they seem to be
forever uncertain just where it is best
for them to settle.
Soon after the publication of Plateau's
observations, I was interested in watch-
ing at leisure the movements of a couple
of wasps searching for flies along the roof
of a tent in which I lay upon my back
in my camp among the Rocky Mountains.
There were a dozen or twenty flies in the
tent, which when not in motion were
alighted on a rope which stretched loosely
just below the ridgepole. The wasps
were in incessant motion, and in the
course of the morning were seen to cap-
ture only three or four flies, the flies
usually being able to dodge them when-
ever an attack upon them was made. I
was unable to see that a wasp accelerated
its motion in the least when approaching
the flies, or directed its flight immediately
upon them, until within two or three
inches of its intended victim; and as it
often passed one by at no greater dis-122
The Life of a Butterfly.
tance than this without any attempt at
capture, the impression was strong that
the wasp’s distinct vision while in flight
did not exceed this distance.
But what was most surprising was the
great number of mistakes made by the
wasps. Every slight stain or defect in
the canvas or minute shadow upon it was
repeatedly attacked by the wasps as if
they supposed it to be a suitable object
for food. There seemed to be no power
on their part of distinguishing between
a spot of color upon the canvas having
no elevation whatever and an object or
body resting upon it. Several times the
shadow made by a fly alighted upon the
outside of the tent was pounced upon by
the wasps on the inside, and such objects,
mere shadows or stains, were repeatedly
attacked by the same wasp over and over
again, often with only half a minute’s
interval or even less than that. I cannot
now recollect exactly the estimate I made
at the time (but failed to record) of theInsect Vision.
123
relative number of attacks upon false
objects to those upon proper victims, but
I am under the impression that the mis-
takes were to the correct judgments as
twenty or thirty to one. These obser-
vations lasted two or three hours, and
certainly secern to show that the vision
of these insects even when in flight is
exceedingly defective, judged by our own
standards.
Now let us apply these conclusions
(the imperfect and vague power of vision
and the keen sense of smell) to two of
the principal acts in a butterfly's life,
having to do with the perpetuation of
its kind, — the meeting of the sexes, and
the deposition of the eggs for the next
brood.
The presence of scent-producing andro-
conia is supplemented by other known
scent-organs in other parts of the body,
principally in the abdomen, numerous
instances of which have been collected by
Fritz Muller, Haase, and others; indeed124
The Life of a Butterfly.
the literature of the subject is becoming
somewhat extensive. Now the mere pres-
ence of scent organs confined to one sex
is in itself ample evidence that the odors
are perceptible to the opposite sex, and
we have in certain specific structures in
the enlarged antennal club §f butterflies
what are plainly sense-organs supplied
with nerve endings; and inasmuch as
there is no structure found in them which
could subserve the purpose of hearing,
or indeed of any other of the senses
known to us excepting that of smell, it
is the belief of physiologists that here
are situated the organs of smell in
butterflies.
The under surface of the antennae of
butterflies is invariably naked to a greater
or less degree; and more plainly in some
joints than in others, little dimples can be
readily seen. It is in these little pits that
are situated the organs of smell; each
consists of a sac-like cavity, the opening
into which is often protected by cuticularInsect Vision.
125
processes, and at the bottom of which in
the hypoderm is situated a fusiform body
with a delicate conical ending extended
free into the centre of the sac, its other
extremity being in direct continuation of
a nervous thread.
The assembling of male moths around
a concealed female already alluded to,
and the strange gathering of the males of
our Southern butterfly, Apostrapliia chari-
thonia, around the female chrysalis just
before and at its eclosion, so graphically
described in Mr. Edwards’s sumptuous
work on American butterflies, are indubi-
table witnesses to the importance of the
sense of smell in effecting the union of
the sexes. That this sense is endowed
with a keen discriminating quality will
hardly be questioned by those familiar
with the immense abundance of indi-
viduals of closely allied species in any of
the warmer parts of the world, and the
excessive rarity of known hybrids. It is
difficult for us, with our dull powers, to126
The Life of a Butterfly.
appreciate it; but granted, it is amply
sufficient to explain the proper union of
the sexes in the heterogeneous multitude
of butterflies which swarm within the
tropics; and when we remember that
in a very large number of cases it is
impossible for the trained naturalist to
discriminate between closely allied but-
terflies in the field (and often exceedingly
difficult in the laboratory) one must be
led to the conclusion that the imperfect
vision of these creatures can prove in no
way sufficient for the continued existence
of the species, but that the sense of smell,
brought to a high degree of perfection in
the struggle for existence, alone has the
power of saving them from extermination,
by enabling each kind to choose its
proper mate.
The same thing appears in the ques-
tion of food for the young caterpillar just
hatched; this must be near at hand, or
it will perish. Yet caterpillars are very
choice of their diet, some extraordinarilyInsect Vision.
127
so. That of our Milkweed Butterfly, for
instance, will eat nothing but plants of
the genus Asclepias, and occasionally
Acerates and Apocynum, — one belong-
ing to the same order, the other its
next-door neighbor; even in Asclepias
it has its favorites.
But there are some butterflies whose
food in early life is even confined, so far
as known, to a single species of plant;
none are indiscriminate feeders, and the
great majority are restricted to a very
few sorts. Moreover, closely allied
butterflies may feed on widely different
kinds of plants, as for example, our com-
mon yellow butterfly, Eurymus philodice,
which lives on Leguminosae, and Legu-
minosae alone, and principally on clover;
while its near neighbor, E. interior, hard
to distinguish from it, has been found to
die rather than touch clover or a dozen
other kinds of Leguminosae, but feeds
instead on Vaccinium, one of the Eri-
caceae, botanically far removed.128
The Life of a Butterfly.
How, then, does the parent infallibly*
discover the plant upon which it should
lay its eggs? This is an act of instinct,
one will say.	But is this any expla-
nation? We wish to know how the
instinct acts. A parent butterfly that in
its caterpillar life has been nourished upon
willow, has no means in the winged con-
dition of tasting the willow to recognize
it, its organs for obtaining food being
suited only for liquid nourishment. Nor
can it be by sight.
It is true that butterflies are attracted
by flowers through their means of vision.
Interesting stories are told of their being
deceived by painted or artificial flowers.
But in these cases there is no reason to
suppose that it is anything but the tint in
* The cases are exceedingly few, may almost be
counted upon one’s fingers, in which the eggs of butter-
flies are known to have been laid in free nature on
anything but a plant on which the caterpillar will feed ;
and in those few cases the proper food-plant was near
enough for the caterpillar easily to reach it in its wan-
derings, — within a few feet, for instance.Insect Vision.
129
mass that attracts them to the coveted
spot. Pray how differs the green of one
plant from that of all others? In the case
of those which feed upon a considerable
number of plants they must have all the
powers a trained botanist has at command
if it be by the form of the leaves, when
nearly every leaf they might properly
choose could be almost exactly dupli-
cated by another as good as poison to
them. No! Anatomy and experiment
both teach us that butterflies have no
vision sufficiently clear for any such
powers of distinction as are required of
them in selecting special food plants for
their young, which yet they discover in
an unerring manner.
There remains, then, apparently nothing
but smell. Now we know that many
plants are odorous quite apart from their
flowers, and that the leaves and other
parts of plants, as well as the flowers, are
furnished abundantly with special struc-
tures, like nectar glands, glandular hairs,
9130 The Life of a Butterfly.
and other glandular devices, from some
of which odors are exhaled perceptible
to our dull senses, — perhaps from many
others, perceptible to keener organs; and
if one with this in mind will but watch
the movements of a mother butterfly seek-
ing a spot whereon to lay her eggs, he
will not fail to recognize that many of
these actions seem particularly in keep-
ing with the notion that she is at work
scenting the various plants that bear a
general resemblance in their aspect to
the plant which she seeks (many, indeed,
which have no such general appearance),
settling or half-settling in a dozen differ-
ent places in the near vicinity of the
plant, reaching it by nearer and nearer
approaches, and finally settling with satis-
faction at the desired spot. To such an
observer it will seem tolerably clear that
it is to the sense of smell that butter-
flies owe their recognition of botanical
species.
The consequences of the conclusions atInsect Vision.
131
which we have arrived are far-reaching.
If vision plays such a secondary part in
the life of insects, then the whole struc-
ture upon which the theory of sexual
selection in insects has been based is at
fault, and it supports the objections to it
which Wallace has brought on other
grounds. It shows that the recognition
of the food plant by the mother, which
does not and cannot taste it, must be by
some other sense than that of sight. It
shows that the very high development
of scent scales of varied patterns and
character among male butterflies, indi-
cating a direct sexual use, can only be
understood if we regard the greater
variety and brilliancy of the colors of
butterflies as contrasted with moths to
have no sexual significance whatever.
Brilliant masculine colors may possibly
have arisen in birds through sexual
selection, but such an origin is impossible
in butterflies with their vague vision ; and
as the males cannot be attractive to their132 The Life of a Butterfly.
mates by seductive colors, they resort to
odors, and vie with each other in the pro-
duction of sweet-smelling garments. It
becomes clear that the exquisite beauty
and variety in the butterfly world is not
recognizable by themselves, and forms no
element in their lives.The Fore Legs of the Male and the Hanging of
the Chrysalis: a Lesson in Classification.
ACTS, I once heard Agassiz say
in one of his lectures, “ are stupid
things, until they are brought
under some general law.” It has no
special interest or significance, in itself
considered, to know that our Milkweed
Butterfly when a chrysalis is suspended
by its tail. It is curious, of course; but it
has no real meaning in itself alone, — it
is then a stupid, barren fact. The study
of the structure or transformations of any
creature has more than a merely transi-
tory interest only when comparisons can
be made with the parts and changes of
other creatures, and we can ask what ar§134 The Life of a Butterfly.
the differences between them and what
the causes thereof. It is of far greater
interest to compare the morphologically
identical but physiologically different
human hand and foot than the fore and
hind foot of, say, a fox, both morpho-
logically and physiologically the same.
So when we come to any repeated struc-
ture in an animal, and find it repeated in
a different way, we inquire at once into
its meaning and are unsatisfied until this
be discovered.
Now there is one point of this sort that
would particularly pique the curiosity of
an inquirer if he were to study the struc-
ture of our Milkweed Butterfly; for he
would soon learn that while the creature,
like all insects, has three pairs of legs, it
walks or rests on only two, and keeps
the front pair folded against the breast,
the ends hanging down. And if he were
to examine this pair more closely, he
would find that not only are all the parts
abbreviated, but that the joints of theA Lesson in Classification.	135
tarsi are all run together, particularly in
the male, and that the claws, being use-
less, are entirely wanting ; in short, the
whole leg, though flexible, is atrophied,
and of no actual use at all.
What is the next step the inquirer
would be likely to take? Why, naturally,
to see whether other butterflies were so
made. Suppose him next, then, to catch
a Swallow-tail Butterfly, as a larger sort
easy to examine. Well! here, with slight
variations in length, all six legs are pre-
cisely alike, excepting that those in front
have a little flabellate appendage to the
tibia which is not found on the other
legs or on the fore legs of the Milkweed
Butterfly. Instead of being deficient,
they are redundant; hypertrophy re-
places atrophy, and the leg is used like
the others. If he have the spirit of the
naturalist, nothing will now satisfy our
inquirer until he has examined all the
butterflies he can lay his hands upon,—
the more so as he will find an extraor-The Life of a Butterfly.
136
dinary diversity among them, and yet
among similar kinds great similarity.
It is by just such comparisons as these,
applied to every part, — the subdivisions
of the trunk, the structure of the antennae,
the eyes, the palpi, the tongue, the wings,
and the legs, — that we learn the relation-
ships of butterflies, and group them into
their various families, subfamilies, tribes,
and genera. And we have to study as
well the structure of the insect in its
earlier stages, — the egg, its form and
sculpture, the caterpillar, its shape and
clothing, and the chrysalis with its special
mode of suspension, its form, and the
relations of its sheathed limbs. Even
the habits bear their testimony, and the
structure of the internal organs as well.
The proper classification of butterflies
thus involves not only the carrying in the
mind an endless array of detailed facts
drawn from every part of the body of
the different creatures in their several
stages of existence, but also the properA Lesson in Classification.	137
valuation of the facts, — whether they are
of considerable importance to the life of
the animal, whether they may be traced
through long series, whether they are
correlated with others which are equally
or almost equally extensive, whether they
may be related to a past phase of exist-
ence now of no concern, and so on. A
successful naturalist, then, is one who
with the largest knowledge combines the
keenest insight into valuations, and can
most surely divine the blood relationships
of the present and the past; for only so
far as our classifications give us an insight
into blood relationships do they have any
value whatever.
To apply these remarks in more definite
fashion, let us proceed to the inquiry
whether the structure of the fore legs
of the perfect butterfly or the mode of
suspension of the chrysalis gives us any
probable clew to relationships among
these creatures. To do this in the brief-
est and simplest way it will be convenient133
The Life of a Butterfly.
to premise that butterflies are generally
regarded as divisible into from four to
six great families, — the Hesperidae, or
Skippers; the Papilionidae, or Typical
Butterflies (divided into two subfamilies,
the Pierinae, sometimes regarded as a
distinct “ family,” and the Papilioninae
proper); the Lycaenidae, or Gossamer
Wings (divided into two subfamilies, the
Lycaeninae proper and the Lemoniinae,
sometimes, perhaps generally, regarded as
a separate “ family ”); and the Nymphali-
dae, or Brush-footed Butterflies (divided
into a number of subfamilies, many of
which have been regarded as “ families ”
by different authors).
The “ family ” or “ subfamily ” value of
a group is of relatively little importance
as compared to determining who are its
neighbors, and in all the varying cases
given above there is no question in this
particular on the part of any one. No
one would venture to separate by the
interpolation of another group the Lycae-A Lesson in Classification.	139
ninae from the Lemoniinae on the one
hand or the Pierinae from the Papilioninae
on the other. By universal consent they
must accompany each other, whether they
are regarded as families or subfamilies.
There has, however, been a considerable
difference of opinion regarding the relative
position of the four great groups among
themselves. Will the structure of the
legs throw any light upon this question?
Let us see.
In the Hesperidae the fore legs, like
those of the moths (which form the other
and lower extreme of the series of Lepi-
doptera, to which butterflies belong),
differ in no respect from the other pairs,
excepting that the hind tibiae are usually
, furnished with a pair of spurs at the
middle as well as at the tip, and the fore
tibiae bear near the middle a peculiar
flabellate appendage, the use of which is
unknown, but which, morphologically, is
unquestionably a spur.
In the subfamily Papilioninae, the mid-140 The Life of a Butterfly.
die pair of spurs of the hind legs is
altogether lost, but the fore-tibial appen-
dages remain, and the fore leg is
otherwise entirely similar in character
to the other legs. Next, in the closely
allied subfamily, Pierinae, the fore-tibial
appendages disappear, but the fore legs
still remain identically like the other
pairs.
As soon, however, as we have reached
the Lycaenidae, we notice signs of an
abortion of the fore legs, but only in
one sex, the male. In the subfamily
Lycaeninae, while the fore leg of the
female does not differ from the other
legs, that of the male begins to lose a
part of its armature and to become abbre-
viated : the tarsal spurs are denuded of
scales, and both the tibial and tarsal
spines are diminished in number; the
paronychia and pads at the tip are
invariably absent; and the claws are rep-
resented by an apical spine or spines
differing from the other spines at most inA Lesson in Classification.	141
size. In the subfamily Lemoniinae the
change has become much greater; for
with scarcely an exception, the fore leg of
the male has become very much smaller
than in the female, and while each part is
reduced in size, the tarsus is represented
by a diminished number of joints, totally
devoid, as is also the tibia, of any arma-
ture whatever, but clothed abundantly
with long scales and hairs. There is
here also sometimes a faint indication of
change in the female, the spines of the
tarsus being less abundant than on the
other legs.
In the Nymphalidae, the change affects
both sexes; not, however, in the lowest
subfamily, the Libytheinae, which, on
this account, many authors who have
given special attention to the structure
of the legs have classed with the Lemo*
niinae. But in all other Nymphalidae,
as in our Milkweed Butterfly, we have for
the first time both sexes fully represented
in the atrophy of the fore legs, and the142
The Life of a Butterfly.
abortion is also carried to a far greater
extent. They are also frequently fur-
nished, especially in the male sex, with
a spreading brush of long hairs, which
gives them a peculiar lappet-like appear-
ance, on account of which they have been
called Brush-footed Butterflies. They are
quite useless, and in some are reduced
to the extremest degree. There is in
butterflies no other structural feature of so
great functional importance which differs
so greatly in the different families, and
follows as here a progressive course.
Changing our point of view, let us next
consider the different ways in which
chrysalids of butterflies are suspended
or cared for. It is well known that as a
general rule moths undergo their trans-
formations to chrysalis within a cocoon,
spun by the caterpillar, or in a cell
moulded beneath the surface of the
ground. The same is true of the lowest
family of butterflies, the Hesperidae,
which usually make such a cocoon withinA Lesson in Classification.	143
a rolled-up leaf or cluster of leaves, and
hence had given them by Boisduval the
term Involuti, or enrolled. It was not
noted by him, nor has it been, as far as
we are aware, by any author, though
figured by many, that within this cocoon
they generally spin a pair of shrouds,
into the middle of one of which they
plunge their cremaster, while by the
other they support the middle of the
body.
Now, remove this outer cocoon and
leave the shrouds, and one has, with only
such changes as are absolutely required
by the lack of the encircling cocoon, the
character of the support of the chrysalis
of the Papilionidae; namely, a button of
silk attached to the object from which the
chrysalis hangs, and a loose girt around
the middle of the body. In the Lycae-
nidae, we pass simply to a still closer
attachment of these fastenings, so that
the rounded chrysalis appears almost
glued to the surface to which it is144
The Life of a Butterfly.
attached; and these two families, the
Lycaenidae and the Papilionidae, were
classed by Boisduval under his Succincti,
or girt.
In the Nymphalidae, by the loss of
the median girt, the chrysalis hangs sus-
pended by its hinder end, and forms the
group termed by Boisduval Suspensi or
Penduli, — that is, hung, — which he and
his followers interpose between the Invo-
luti and the Succincti. Yet we have here
a regular progression from the cocoon of
the moths to the almost total absence
of the use of any silk for the quiescent
period of life. The few exceptions to
this rule seem to be entirely explanable
as instances of reversion. Thus the only
case among the higher butterflies where
a cocoon, properly speaking, is made, is
in the subfamilies most closely allied to
the Hesperidae, among the groups of
Parnassini and Anthocharini,* and again
* Bar asserts that the same is true in some South
American Lemoniinae.A Lesson in Classification.	145
in exceedingly feeble instances, where the
necessities appear to be overwhelmingly
great, among some of the Nymphalidae,
which have lost even the last remnant of
the cocoon of moths, — namely, in some
of the Satyrinae, which lack cremastral
hooks and undergo their transformations
ordinarily in the rudest form of a cell
which they can construct at the surface
of the ground, by the mere movements
of the body and the spinning of one or
two threads of silk.
Moreover, note this:	an additional
feature appears in the structure of the
chrysalis of a large number of the Nym-
phalidae, which would seem to indicate
that they inherit the mark of the “ suc-
cinct ” condition of their ancestors, in the
straight ventral surface of the entire
chrysalis, — a feature absolutely without
value in its present suspended condition,
but full of meaning, since it is one
necessarily common to all the close-
bound members of the Succincti, the
10146 The Life of a Butterfly.
Lycaenidae. That is to say, the straight
ventral surface of many hanging Nym-
phalidae indicates that this group has
passed through and come out of a “ suc-
cinct ” mode of suspension.
Now it will be observed that these
two series, drawn from such independent,
important, and different sources as the
development or atrophy of the fore legs
of the butterfly and the mode of sus-
pension of the chrysalis, are entirely
parallel. Indeed, taken together they
indicate, with a certainty somewhat rare
in the study of animals so nearly allied,
the general interrelationship of the fami-
lies concerned, which a study of other
details of their structure only the more
strongly confirms, — the Papilionidae, for
example, presenting very many more
points of contact with the Hesperidae
than can be shown to exist between
the Nymphalidae and the Hesperidae.
Each of these families contains groups
of lesser extent founded on structuralA Lesson in Classification.	147
or developmental characteristics of lesser
value, which it is the aim of the syste-
matic naturalist to set forth in terms
which shall be, as nearly as possible, a
transcript of the teachings of Nature her-
self. Just so far as he seeks to make it
his own system, will he fail; just so far
as he strives to have Nature reveal herself,
will he succeed. He must not regard
popular opinion, but seek his sup-
port solely from Nature. The investigator
in this field must be modest, patient,
reverent.X.
What is its Proper Name ? A Brief Chapter
on Nomenclature as used by Naturalists.
j|UCH outcry is made against the
terminology of natural history.
But if any one can point out
a more sensible method than that be-
queathed to us by Linne and his followers,
none will be more ready to welcome it
than the naturalists. In this, as in every-
thing else, the most serviceable method
will obtain the most service. Indeed, one
sometimes fancies that the critics forget
what names are for. They merely stand
for things, and convenience is their only
hope of life. They must be brief at any
rate, whether they have any meaning in
themselves or not; and they must follow
certain rules, or we shall stand in such a
wilderness that no one can find his way,What is its Proper Name? 149
and the very object for which the names
are meant will be obscured or lost.
We have seen that there are various
groups of butterflies of greater or less
value, and that each contains other groups,
until we reach the species and variety.
Varieties, indeed, were not much consid-
ered in Linne’s time, and it is only
since Darwin’s day that they have gained
much significance. As the outcome of
the disputes and arguments about names,
men have pretty generally agreed to give
to each different kind of animal a double
name, — the first indicative of the ultimate
structural group, or genus, to which it
belongs; the second a specific name,
agreeing in gender with the former, which
has or is supposed to have a Latin form;
to which has sometimes been added in
later times, when needed, a varietal name.
To all higher groups single names of a
Latin plural form are given; and attempts
have been made, with partial success, to
have each group of the same category150 The Life of a Butterfly.
bear a similar termination, families to end
in -idae, subfamilies in -inae, etc.
Our Milkweed Butterfly, for instance
(which requires no varietal name), is
Anosia plexippus, and it belongs to the
subfamily Euploeinae (named from one
of its principal genera, Euploea) and the
family Nymphalidae (named after one
of its oldest genera, Nymphalis). As a
member of the Nymphalidae it shares
with the others of its family a certain
number of characteristics which are not
found combined, and rarely found at all,
in any member of the other families. As
one of the Euploeinae it has certain pecu-
liarities in common with other genera of
the same subfamily, not found combined,
generally not found at all, in the other
subfamilies of Nymphalidae. And as a
member of the genus Anosia it has cer-
tain structural characteristics in every, or
nearly every, stage of life which separate
it from the other genera of Euploeinae.
It bears within itself, then, charactersWhat is its Proper Name? 151
which belong distinctively to family, sub-
family, genus, and species, — and it is
the province of the naturalist to discrimi-
nate these, — and the terminology of the
nomenclature is necessary to speak con-
cisely about them and discuss their
relationships. Without such a makeshift
and tool small progress would be possible.
Imagine mechanics without names for
their tools, engineers without a nomen-
clature for the parts of their machines,
and you will feebly fancy the naturalist
without his terminology.
I have said our butterfly’s name was
Anosia plexippus. Yet if one were to look
in the last catalogue of American butter-
flies, he would not find this name, but in
the place where it should stand, the words
Danais arckippus. This is certainly per-
plexing to a beginner; and the perplexity
is unavoidable as long as men will not
think alike and act alike. A name once
rooted by usage for a few years only, even
if it can unquestionably be shown that152 The Life of a Butterfly.
such usage is in contempt of accepted
rules, may have a stronger chance of life,
such is the natural conservatism of man-
kind, than one clearly correct • Especially
is this the case where the mistake has
been made in some important work by
an “ authority,” and the error corrected
by some unknown but patient worker
in a less conspicuous volume. In this
way the name is apt to vary in different
countries.
Now, both names, archippus and plex-
ippus, have been employed and in
“ usage ” from very early times, but it
has been shown by several naturalists in
different parts of the world that the
use of the specific name archippus per-
petuates an error; and while in other
countries plexippus is now commonly
employed, in America, thanks to most
of our catalogue-makers, archippus still
often retains its usurped place, probably
because Boisduval and LeConte used it
half a century ago. In many cases theWhat is its Proper Name?	153
question is one of priority of usage, pure
and simple, and then the older name in
time gains place, but often only after a
struggle which is sometimes ludicrous;
proved priority of baptism is the only
safe guide, the only chance for real
stability.
As to the differing generic name, that
is on another basis. Naturalists are not
agreed and probably never will agree
about the precise limits of “ genera,”
some dividing a given collection of
species into several genera, some regard-
ing them as one or two only. Now the
old genus Danais, as looked at by those
who use this name for our butterfly,
comprises a very large number of very
diverse forms, which the latest students
of these insects have shown should be
segregated into a number of groups ; and
Anosia is the one of these smaller groups
into which our species falls.
The use of the generic name, then,
indicates to some extent the views of the154 The Life of a Butterfly.
person employing one or the other term.
Certainly the more closely we study but-
terflies (which in the past have been
studied largely upon superficial aspects
and by an examination of the perfect
insect alone) the more shall we be
inclined to look with favor on structural
groups of lesser extent, as enabling us,
first, to formulate statements regarding
their distribution and relationships which
will throw new light on many dark spots
in science; and second, to correlate their
differences with those of other groups of
insects where corresponding minor groups
are accepted, because the structure of
these latter insects is not concealed by
a coating of scales, and their study not
beset with so many practical difficulties.
There is still a further reason for
such a preference. With the beginning
of systematic zoology, all butterflies were
grouped in a single genus, Papilio; study
had then progressed only so far as to
seize upon the most salient and obviousWhat is its Proper Name ?	155
points in their structure. As the study
became deeper and more searching, and
new forms of very diverse character
swelled by degrees and at last immensely
the number of species (even in the last
edition of his Systema Naturae, a cen-
tury and a quarter ago, Linne recognized
but about three hundred kinds, while
now we know at least ten thousand), it
became necessary to subdivide them into
a greater and greater number of groups,
subordinate to one another, and families,
subfamilies, and tribes were introduced to
distinguish and class them. The “genus”
has been more and more restricted with
the growth of our knowledge, and the
tendency has all along been to have
it represent the structural group of nar-
rowest limitations, while the study of
the structure of butterflies in all their
stages has constantly narrowed these
limitations.
To object, then, to the employment
of a “ generic ” term for the narrowest156 The Life of a Butterfly.
groups, when they can be shown to have
a real foundation, is practically to put
a check upon the study of their struc-
ture, which it is the very first business
of the systematic zoologist to encourage
and indeed to insist upon. No point is
so minute as not to merit investigation
and fair consideration. The researches
of Darwin surely carry this lesson.XI.
Some Points not touched upon, and some
Things we do not yet sufficiently know. —
A Suggestion for Future Study.
LTHOUGH in the preceding
chapters we have given a tolera-
bly full history of our Milkweed
Butterfly, and have by this means illus-
trated in a few instances the structure and
transformations of other butterflies, there
are still many points about Anosia plexip-
pus which have not even been alluded to,
while our references to others have been
so meagre that whole classes of phenom-
ena, some of the highest interest, have
been left untouched.
Thus, to confine ourselves for the mo-
ment to our principal subject, not a word
has been said either about the growth of158 The Life of a Butterfly.
the creature within the egg or about the
internal structure of the caterpillar, chry-
salis, and butterfly, — by what processes
both the internal and the external changes
are brought about, and in what way the
new organs, such as the wings of the
perfect insect, originate, present as they
are within the body of the caterpillar
from its earliest life. These topics are
full of interest both in themselves and in
the further comparisons with the same
processes in other butterflies or in still
other allied creatures. Our present know-
ledge upon these subjects, however, is
more limited than that of the topics
already discussed, and they will not yet so
readily lend themselves to popular inter-
est ; and therefore, though it renders our
little book less complete and symmetrical,
they are not touched upon at all.
There is, however, one other topic
which, having been somewhat widely
studied, especially in this country, and
not yet more than mentioned here, mayA Suggestion for Future Study.	159
be briefly stated. I refer to the remark-
able differences between the earliest and
latest stages of one and the same cat-
erpillar. This is less marked in our
Milkweed Butterfly than in most of our
butterflies, especially than is common
among other members of the same family,
which when adult have bodies and even
sometimes heads bristly with compound
spines, and altogether lack such a formi-
dable appearance in their tender youth.
The full-grown caterpillar of our Milk-
weed Butterfly is, as we have seen, a
gayly striped yellow, black, and green,
naked or nearly naked worm four or five
centimetres long, with a black-banded yel-
low head and two pairs of black flexible
filaments on the body. When it first
emerges from the egg, however, it is as
a pale green cylindrical worm, about
three millimetres long, with a black head,
and on its body, scarcely to be seen with-
out a microscope, are several rows of
minute papillae, one papilla to a segment160 The Life of a Butterfly.
in each row, each surmounted by a short
simple hair about half as long as one of
the segments. Careful examination will
also show on each side of the second
thoracic and eighth abdominal segments
a somewhat similar but larger and hair-
less papilla. These last persist through
life, and as we have seen, grow to more
importance as filaments; all the others
disappear completely with the first change
of skin.
What the meaning may be of this
infantile clothing, which in one form or
another is present in the earliest stages
of all butterfly caterpillars, and is alto-
gether lost in the subsequent growth,
usually with the first change of skin and
completely, but which sometimes per-
sists through one or more subsequent
changes, more or less modified, is not
perfectly clear. In a large number these
hairs, more or less specialized, are plainly
the outlets of glands lying at their base,
secreting fluids which often may be seenA Suggestion for Future Study.	161
as droplets at the tips of the hairs; and
it is possible that in these cases the
glandular secretion serves as a protec-
tion to the little creature through its
odor or offensive taste.
But this is not the case with all, of
which our Milkweed caterpillar is an
instance; and it is altogether probable,
from the universality of this form of
clothing and its general simplicity, that
it has a phyletic significance yet to be
discovered, — an indication of the nature of
the clothing of the primitive caterpillar of
all butterflies, or better, all lepidopterous
insects, and that the glandular structure
of some is a secondary feature brought
about through natural selection. Here is
a highly interesting and important chapter
in the phylogeny of insects which will
well repay the conscientious student.
Let us now turn for a brief moment to
some of the phenomena witnessed in
other butterflies, but which cannot be
illustrated from the butterfly we have
ii162
The Life of a Butterfly.
chosen for special consideration. We
will mention but one among many.
Among the subjects of general philo-
sophical interest which the study of
animals during the Darwinian epoch has
brought to notice, few have excited more
attention and interest than the existence
in a vast number of animals of two or
more distinct forms in the same species.
That this is very commonly true of the
two sexes goes without saying; but
besides this it often happens that one
sex may appear under two distinct guises,
or that alternate broods of the same
animals may differ so much from each
other as in many cases to deceive the
most acute naturalist into the very reason-
able belief that they are distinct species.
Much attention has been given to this
subject among the butterflies, and we
have in our native species a consider-
able number of instances in illustration.
Indeed a large proportion of our but-
terflies show, in some peculiarities ofA Suggestion for Future Study. 163
the scales of the male sex and their
arrangement into special patches, a ready
distinction from the opposite sex. There
are also a very large number which differ
from the opposite sex in the general
color or .pattern of the upper or lower
surface of the wings. Curiously enough,
when we consider how very generally
the under surface of the hind wing is
variegated in butterflies, we rarely find
in this place any distinction between the
sexes. It is largely confined, at any rate
with the butterflies of the temperate zone,
to the massive coloring of the upper
surface; and here, whenever one of the
sexes departs from the typical coloring
of the group to which it belongs, in
order to assume a livery distinct from its
mate, it is almost always the female, at
least among our own butterflies, which
is thus distinguished.
But besides that form of dimorphism
which simply intensifies the distinctions
between the sexes, we have in some cases164 The Life of a Butterfly.
a double dimorphism, so to speak, which
not only distinguishes one sex from the
other, but divides the members of one
of the sexes into two distinct groups,
one of which more nearly resembles the
opposite sex, while the other may depart
widely from it. There are cases, such
as our common yellow butterfly, familiar
to every out-door entomologist, wherein
the female, instead of being of nearly
the same color as the male, with some
distinctions in the marginal bands, is
of so very pallid a hue as to strike
the observer at once. The contrary is
true as regards the female of the Tiger
Swallow-tail, Jasoniades glaucus, and the
male of the Spring Azure, Cyaniris
pseudargiolus, in which, in certain parts
of the country, a dark form of the sex,
with nearly uniform brown upper surface,
is to be found.
But dimorphism by no means stops
here, for we have in some of our other
butterflies quite as striking or even moreA Suggestion for Future Study. 165
striking peculiarities. Take for instance
the case of one of our Angle-wings,
Polygonia interrogationis. Here is an
insect where there are two very distinct
forms in each sex, and in each of which
the sexes are readily distinguished by
the coloration of the wings; they differ
in the brightness and variegation of the
lower surface of both wings and the ob-
scurity of the upper surface of the hind
pair; that is to say, there are four sets
of individuals, which may be separated
quite as readily as most closely allied
butterflies, and more so than a great
many acknowleged species of the best-
studied faunas. Nor is this by any means
the whole statement of the case; these
two types, bred from eggs laid by
the same parent, differ not only in the
markings of the wings, but also in their
form and in the structure of the male
clasping organs; in fact, we have two
sets, permanently distinct from each other,
and to which we cannot apply the name166	The Life of a Butterfly.
of species only because we know them
to have the same parent.
Now butterflies seek their own kind
for mating, and nothing more is needed
to establish these forms as good species
than that each should persistently seek
its own kind. Indeed, one can hardly
help surmising that they already do so
to a considerable extent, and that this is
an instance of an almost formed species,
beyond which it is almost impossible to
go without becoming one. Between this
condition and that of species in which
the sexes do not differ and there is
very little variation, there is almost every
grade of difference, so that we may fairly
imagine that we know one means by
which species are originated. Here, per-
haps, if anywhere among butterflies, we
ought to suspect that physiological selec-
tion, the province of which is so well
insisted upon by Romanes, is beginning
to play its part in the formation or rather
the differentiation of species; since be-A Suggestion for Future Study.	167
sides the colorational marks which may
enable the sexes to choose their mates
with discrimination, we have the first
steps toward those changes in the organs
ancillary to generation which everywhere
mark absolutely distinct forms, and are
safeguards against admixture.
More striking and perhaps more con-
fusing than these examples are those
where the dimorphism is seasonal; that
is, where the butterflies of the first brood
differ, and sometimes to an extraordinary
extent, from those of succeeding broods
the same season; or even where the
earliest appearing members of a spring
brood may be separated by coloring,
pattern, or form of wings from the later
emerging individuals of the same brood.
Perhaps the most striking instance that
we have is in the often quoted case of the
Zebra Swallow-tail, Iphiclides ajax> whose
changes have been so thoroughly worked
out by Mr. W. H. Edwards. Here each
form appears at a different season of the168	The Life of a Butterfly.
year; marcellus is the early spring type,
telamonides the late spring, and ajax the
summer and autumn type. Nearly all
the butterflies which, in West Virginia,
emerge from the chrysalis before the
middle of April, are marcellus; between
that and the end of May, telamonides;
after this, ajax. The first two, however
do not represent distinct broods, for
telamonides is not the direct conseasonal
produce of marcellus, but both are made
up of butterflies which have wintered as
chrysalids, — those which disclose their
inmates earliest producing marcellus, the
others telamonides; while all butterflies
produced from eggs of the same season —
and there are several successive broods
— belong to ajax. These forms differ in
the length of the long tails upon the hind
wings, in the clothing of the front of the
head, in the extent of the blood-red
spots upon the hind wings, and in other
markings, and before their relation was
known were regarded by all naturalists as
distinct species.A Suggestion for Future Study.	169
Geographical variation of course occurs
in butterflies, as in other animals, but
when this is combined at once with
seasonal dimorphism, sexual dimorphism,
and ordinary simple dimorphism, one
becomes almost hopelessly involved in
trying to disentangle the threads. One
such instance is found in our Spring
Azure, Cyaniris psen da rgio lus.
This species, in which the males and
females differ considerably in the mark-
ings of the upper surface, is spread over
almost the entire North American con-
tinent In the extreme North, from
Labrador to Alaska, it is single-brooded,
but appears in two forms, lucia and vio-
lacea, differing largely in the heaviness
of the markings of the under surface.
Whether one form flies before the other
does not appear from any observations
on the spot, as these are too meagre; but
there is certainly nothing to show that
there is any difference in this respect from
what we find farther south, To examine170 The Life of a Butterfly.
this we must confine our attention to the
eastern half of the continent. Not far
from latitude 450 north, two new phe-
nomena appear. The butterfly becomes
double-brooded and trimorphic; and the
third form, with still lighter markings, of
which the second brood is exclusively
composed, appears also as a member of
the first brood, — the three forms succeed-
ing each other at least within a month, in
the order of the heaviness of the dark
markings of the under surface; namely,
lucia, violacea, neglecta. Confining our
attention for the present to the first brood,
observations would seem to show that in
the northern part of the belt of its
trimorphism, the form neglecta is com-
paratively rare, but that in proceeding
farther and farther south it becomes pro-
portionally more and more numerous,
until, as about Albany, N. Y., it has alto-
gether usurped in numerical importance
the place formerly occupied by lucia,
which entirely disappears at about theA Suggestion for Future Study. 171
latitude of 41 °, except (probably) in
mountainous districts. At the same time
the second brood, although apparently
not more the product of neglecta than of
violacea, becomes more abundant.
We now reach another belt of country,
in which we find the butterfly again
dimorphic in the first generation, — vio-
lacea and neglecta, in the order of their
appearance and the summer generation
as before. But we have not far to pass,
say to 38° or 390 north, before we reach
a new condition, in which the first form
of the first generation becomes sexually
dimorphic, — the males appearing under
two guises, one blue above, the normal
violacea; the other dark brown, vio-
lacea-nigra; and this apparently continues
as the condition of things as far toward
the Gulf as the species extends. There
can be little doubt that this succession of
changes in passing from north southward
is modified and interfered with to a con-
siderable extent by the Alleghanies, and172
The Life of a Butterfly.
that on their flanks, in very near vicinity,
we may find at least some examples of
all these forms.
In the extreme West, on the Pacific
slope, we have a new form, piasus, most
resembling neglecta, which so far as
observations have gone appears to be sin-
gle-brooded in the North, double-brooded
in the South, and to show no difference
between the broods, — as sharp a contrast
as could well be found to the character
of the species elsewhere ; and it is the
more strange since in Arizona {though
it should be noted, among the mountains)
an ashen-tinted form of violacea appears,
to which Edwards has given the varietal
name cinerea. The form piasus occurs
as far north as Central California. What
is found immediately to the north of that
is not well known. At Vancouver, how-
ever, lucia and violacea are met with, and
violacea at least in Oregon; and it would
seem as if in the intervening area not
only lucia but also violacea must disap-A Suggestion for Future Study. 173
pear, and leave only the more weakly
marked Pacific representative of neglecta
as the remnant of the polymorphism of
the first brood, and which does not
become digoneutic until left in possession
of the field
Enough has been said, even in the few
instances given, to show that we find
among butterflies topics enough of inter-
est to any one. Yet it should be pointed
out that it is only within recent years that
these varying phases of life and form
have begun to be studied, and that new
problems arise with every increase of
knowledge, with every forward step in
investigation, so that no one may com-
plain that the ground is too trodden, or
that new elements of interest may not
arise by honest search. Surely the know-
ledge yet to be gained from the study
of the life histories of butterflies will
prove far beyond anything we are yet
acquainted with, just as the problems
discussed the past twenty-five years are174 Bhe Life °f a Butterfly.
of far wider interest than those which
went before.
Even in the case of Anosia plexippus, no
one can say that further study is not
required, for although this interesting but-
terfly is one of our best-known species,
there are several points in dispute regard-
ing it, and many features in its history
which need further investigation. The
movements of the butterfly in the spring
and in the autumn will require systematic
and concerted observation over a wide
extent of territory before their satisfactory
solution can be expected. Where swarms
and bevies occur, they should be carefully
observed from day to day and hour to
hour, to study the movements and intent
of the throng.
The whole question of the regular or
irregular migration of butterflies can be
studied better with this species than with
any other in the world, because there is
none so subject to congregational move-
ments which occurs where so manyA Suggestion for Future Study. 175
intelligent observers are stationed. If, as
I think I have shown it to be probable,
there is over the entire extent of the
country inhabited by it, at least east of
the Rocky Mountains and north of the
Gulf States, a periodic movement of the
butterfly, to the south in the season which
corresponds to the end of September in
New England, and to the north in the
time of the first (and in the middle belt,
of the second) season of egg-laying, then
observers ought to note at these periods
the general direction of movement — not
over a few feet or rods, but as far as the
eye can follow them — of as many speci-
mens as possible, tabulate them, and
publish the results. By this means I
believe a periodic movement could finally
be as well established as the annual
migrations of birds; to this work every
one can contribute who knows the butter-
fly by sight.
Then we need many more careful obser-
vations on the immunity supposed to be176	The Life of a Butterfly.
enjoyed by this butterfly in its various
stages, — an immunity certainly not per-
fect, and the exact nature and extent
of which will reward only patient and
conscientious field work. It would be
interesting to know the relative willing-
ness with which insectivorous birds would
devour the caterpillar and that of the
Black Swallow-tail, Papilio polyxenes,
which feeds on parsley, etc., and bears
some resemblance to it. The life history,
and particularly the number of broods
in a season, should be worked out inde-
pendently in many places, and for several
seasons in each, to determine questions
in which writers are at variance. For this
the condition and abundance of the but-
terfly should be observed from week to
week, and with it the relative numbers of
caterpillars in each stage, tabulating all
the facts that can be obtained, including
observations of the ovaries of such
females as are captured.
The following would seem to be anA Suggestion for Future Study. 177
interesting and valuable experiment for
some one having access to a greenhouse
empty in summer to try: place together
males and females reared from July cat-
erpillars in the northern half of New
England, without other admixture, in
such a greenhouse, covering the open
windows with netting to prevent escape,
and placing an abundance of Asclepiads
in bloom and in young shoots within. If
the butterflies will breed in confinement,
then the females should lay eggs, if, as
some believe, there are in this district two
broods of butterflies from July on; they
should not lay eggs if, as I believe, there
is but a single generation. Coupled with
this there should be similar experiments
farther south, where there is more than
one brood, to see whether they will breed
at all in such confined quarters.
There are still further inquiries that
should be made, if we wish to pursue the
study of this insect beyond our own
borders; for what has been written above178 The Life of a Butterfly.
regarding its distribution in recent years
opens the question as to the progress it
is making in overspreading the globe, and
what effect its introduction may have
upon the other butterfly inhabitants of
the regions it invades. In New England
our newly introduced cabbage pest, Pieris
rapae, has practically exterminated our
native Pieris oleracea, except in wild
regions, apparently by getting a few days
start of it in the season and monopolizing
the best feeding-grounds. Then, as the
Milkweed Butterfly has been shown to
vary considerably when individuals from
North and South America and the West
Indies are compared, the origin of the
Pacific hordes may perhaps be traced by
seeing to what type they correspond.
The variation itself has been in no way
properly studied, and it remains to inquire
whether the wanderers in the Pacific are
going to show any departure from the
type.
In considering the exotic distributionA Suggestion for Future Study. 179
of the Milkweed Butterfly, it should,
moreover, not be forgotten that in reach-
ing the Old World it may be said to
have reached its own, for Anosia does
not belong to the exclusively New World
type of Euploeinae, Ithomyini, but is
one of the exceedingly few New World
genera of the otherwise Old World type,
Limnaini. The work, then, that still re-
mains to be done on this one insect
might well engage a lifetime, and we
are the first to show how insufficient a
presentation of the subject the present
volume affords!EXPLANATION OF THE PLATES.
(all the plates illustrate anosia plexippus.)
Plate I.
The Male Butterfly ; natural size.
Plate II.
Figs. 1-5. Front view of the head of the cater-
pillar, at each of its five stages; the line
beneath each indicates the width.
Fig. 6. Caterpillar, last stage; natural size.
Fig. 7. Egg; enlarged about 17 diameters.
Fig. 8. Caterpillar, first stage; enlarged about
14 diameters.
Fig. 9. Longitudinal section through the female
butterfly to show the internal anatomy: —
/, tongue; p, palpus; a, antenna; pr, pro-
thorax; mes, mesothorax; met, metathorax;
ps, pharyngeal sac; b, brain; sog, suboeso-
phageal ganglion; 1-2, blended first and second
ganglia of the caterpillar; 3-4, blended third
and fourth of same; /, /, /, the three legs; ac,
aortal chamber; dv, dorsal vessel; oe, oeso-
phagus ; res, reservoir for air or food; st,
stomach ; mv, malpighian vessels ; i, intestine;Explanation of the Plates.	181
c, colon; r, rectum ; cp, copulatory pouch ; 0,
oviduct; #£*, accessory glands ; sp, spermotheca;
ovaries (not fully developed); nc, nervous
cord. Enlarged 3 diameters.
Plate III.
Fig. 1. Side view of the appendages of the butter-
fly a little enlarged, with some separate parts
further enlarged : —p, palpus ; f fore tibia and
tarsus, male and female; /*', tip of fore tarsus
of female; in', m", side and dorsal view of last
joint of middle tarsus of male.
Fig. 2. Side view of the denuded extremity of the
abdomen of the male butterfly, showing the
pencil of extensile hairs in place. Enlarged.
Fig. 3. Greatly enlarged view of part of dorsal
surface of hind wing of the male butterfly,
showing the distribution of the scales on and
about the lowest median nervule and the
adjoining pouch.
Fig. 4. Cross section through the same, showing
the nervule and the pouch, the latter the
broader.
Plate IV.
Figs. 1-6. Scales from the hind wing of the male
butterfly: 1-3, from the edge of the pouch ;
4, from the vein next the pouch ; 5-6, from the
interior floor of the pouch. Greatly enlarged.
Fig. 7. Rear view of chrysalis ; natural size.
Fig. 8. Longitudinal section of the head to show
the pharyngeal sac: — mx, left maxilla (the
right removed) ; mjl, floor of mouth cavity or182 Explanation of the Plates.
pharyngeal sac; oe, oesophagus ; ov, oral valve;
sd, salivary duct; dm, fin, dorsal and frontal
muscles which open the sac. Above the sac
are seen the cut ends of the transversely encir-
cling muscles which close the sac. Enlarged
about 20 diameters.
Fig. 9* Side view of chrysalis; natural size.
Fig. io. Cross section of the spiral tongue, the
anterior portion uppermost, to show the mode
in which the two halves unite to form a central
canal through which the fluid food ascends: —
c, central canal; tr, trachea; n, nerve ; m, m\
muscles of one side. Enlarged about 125
diameters.
[Plate I. is from the original by C. V. Riley. Plate II.
fig. 9, plate III. fig. 2, and plate IV. figs. 8 and 10 are from
the originals by Edward Burgess.]PlatePlatePlate 3.
4Plate 4
10INDEX
Acerates as food plant, 17,127.
Agassiz, L., quoted, 133.
Agraulis vanillae, 76.
Androconia, 101.
Angle-wings, 35, 165.
Artosia plexippus: see Milk-
weed Butterfly.
Antennae, structure of, 124.
Anthocharini, 144.
Anthocharis genutia, 74.
Antirrhea, 104.
Apocynum as food plant, 17,
48, 127.
Apostraphia charithoma, 125.
Arabis as food plant, 74.
Argynnis atlantis, 104.
Aromatic butterflies, 104.
Artificial flowers attracting
butterflies, 128.
Asclepias as food plant, 16,
127.
Asmuss, E. P., quoted, 69.
Assembling of Lepidoptera,
107.
Atrophy of legs, 13, 134.
Balance of Nature, 78.
Bar, C., quoted, 144.
Basilarchia, 83,92; B. archip-
pus, 90, 93; B. astyanax,
90, 92.
Bates, H. W., quoted, 82,
83, 84.
Birds vs, insects, 72.
Black Swallow-tail, 176.
Blues, 36.
Bombycidae, assembling of,
107.
Broods, single and double,
38, 167.
Brush-footed butterflies, 138.
Buckler, W., quoted, 70.
Burgess, E., quoted, 26.
Butterfly, Milkweed; see
Milkweed Butterfly.
Callidryas eubule, 105.
Caterpillar, danger to, 70;
early stage of, 70; infan-
tile clothing, 160; in hiber-
nation, 37; juvenile, 159;
milkweed, 14.
Chalcis, 76.
Chlorippe clyton, 76.
Chrysalis, dangers of, 75, 76;
milkweed, 15; modes of
suspension, 142.
Classification, 133.
Clover as food plant, 127.
Coenonympha pamphilus,
69.
Colors, warning, 73, 79.
Commercial agencies in dis-
tribution, 61.
Conspicuous caterpillars, 73,
80.
Coppers, 36.
Cyaniris pseudargiolus, 105,
164, 169; C. p. lucia, 170;
C. p. neglecta, 170; C. p.
piasus, 172; C. p. violacea,
170; C. p. violacea-cinerea,
172; C. p. violacea-nigra,
171.
Cycles of life, 38.Danais archippus, 151.
Darwin, C., quoted, 101.
Desiderata, 174.
Dewitz, H., quoted, 18, 21.
Didonis, 104.
Dimorphism, 162; double,
164; seasonal, 167; sexual,
162.
Diosmism, 106.
Distribution, geographical,
4°.
Doryphora clavicollis, 80.
Dusky-wings, 37.
Edwards, W. H., quoted,
21, 3°> 33’ 67, 70,125, 167.
Egg, destruction of, 67; in
hibernation, 35; of Milk-
weed Butterfly, 13.
Emigration of butterflies, 43.
Ericaceae as food plants, 127.
Euchaetes egle, 80.
Euphoeades troilus, 93.
Euploea, 150.
Euploeinae, 83, 150, 179.
Eurymus interior, 127; E.
philodice, 127.
Euvanessa antiopa, 35.
Experiments, suggestions of,
177; in vision, 113.
Eyes of butterflies, structure
of, III.
Families of butterflies, 138,
150.
Favored butterflies, 82.
Food plant of Milkweed But-
terfly, 16; selected by
smell, 130.
Fore legs, structure of, 13,
134-
Forel, A., quoted, in.
Fritillaries, 92.
Genera, limitations of, 153.
Glandular hairs, 160; scales,
I03‘
Gossamer-wings, 69, 138.
Groups in butterflies, 138.
Gulick, L. H., quoted, 53.
Gundlach, J., quoted, 17,
21.
Haase, E., quoted, 123.
Habits of caterpillar, 17.
Hair-streaks, 36.
Hamilton, J quoted, 48.
Hesperidae, 138, 139, 142.
Hibernation, 35; of butter-
fly) 35 5 °f caterpillar, 37;
of chrysalis, 37; of egg,
35*
Howard, L. O., quoted, 76.
Ichneumon-flies, 31, 71, 76.
Imago: see Milkweed But-
terfly Butterfly.
Immigration in butterflies, 42.
Infantile clothing of cater-
pillars, 160.
Insect vision, no.
Instinct, 128.
Involuti, 143.
Iphiclides ajax, 167; I. a.
ajax, 168; I. a. marcellus,
168; I. a. telamonides, 168.
Ithomyiae, 83.
Ithomyini, 179.
Jasoniades glaucus, 164.
Juvenile caterpillars, 159.
Larva: see Caterpillar.
Legs, atrophy of, 13, 134;
gradation in structure of,
137.
Leguminosae as food plants,
I27-.
Lemonnnae, 138, 141, 144.
Leptalides, 83.
Libytheinae, 141.
Life history of butterflies, 30.
Limnaini, 179.
Lycaenidae, 138, 140, 143.
Lycaenids, 36.
Lycaeninae, 69, 138, 140.Index,
185
Malodorous butterflies, 81.
Meadow-browns, 37.
Melete, 105.
Migration of butterflies, 42,
174-
Milkweed Butterfly, 1; broods,
33; cycles, 33; desiderata,
174; distribution, 40; found
in Asia, 59; in Europe, 60;
in high north, 47 ; in Pacific
Islands, 54; life history, 30;
parasites, 68, 71; protec-
tion, 80; suggestions for
experiments, 177.
Milkweed Butterfly Butterfly,
12; atrophy of fore legs,
13, 134; broods, 33; emi-
gration, 43; fore legs, 13,
134; hibernation, 30; im-
migration, 43; legs, 13,
134; life, 30; migration,
43; mimicked, 83,	90;
odor, 81, 105; pouch in
wing, 13; scales of male,
97; seasons, 30; swarming,
48; tongue, 23; vagrancy,
40; variations require study,
178; wanderings abroad,
53; warning colors, 83;
wing-pouch, 13.
Milkweed Butterfly Caterpil-
lar , 14, 31; conspicuous
colors, 73; early stage, 15,
159; food plant, 16; habits,
17, 31; juvenile, 15, 159;
moulting, 18; mouth-parts,
26; warning colors, 73.
Milkweed Butterfly Chrysa-
lis, 15; duration, 21; sus-
pension, 32.
Milkweed Butterfly Egg, 13;
where laid, 16, 31.
Mimicry, 79, 89.
Mistakes of vision, 122.
Moulting of caterpillar, 19.
Mourning-cloak, 35.
Mouth-parts, 23.
Muller, F., quoted, 26,102,
106.
Names, 148.
Nauseous butterflies, 81.
Nomenclature, its necessity,
148.
Nymphalidae, 138, 141, 144,
i45j i5°-
Nymphalis, 150.
Odor of butterflies, 81, 104;
of plants, 129^
Papilio, 154; P. polydamas,
106; P. polyxenes, 176.
Papilionidae, 138, 143.
Papilioninae, 138, 139.
Papillae of tongue, 25.
Parasites, 67, 71, 76.
Parastatic mimicry, 89.
Parnassini, 144.
Parnassius, 36.
Patten, W., quoted, in.
Peale, T. R., quoted, 22.
Penduli, 144.
Pholisora, 37.
Phylogeny of butterflies, 146,
161.
Physiological selection, 166.
Pierinae, 138, 140.
Pieris napi, 106 ; P. oleracea,
106, 178; P. rapae, 178.
Plateau, F., quoted, in,
112, 113, ti8.
Polygonia interrogationis, 67,
l65- .	.
Pouch in wing, 13.
Prittwitz, O. von, quoted,
70.
Protective resemblance, 79.
Pteromalines, 76.
Pupa: see Chrysalis.
Reptiles as foes, 72.
Riley, C. V., quoted, 16,
33, 5°, 5T> 79-
Romanes, G. J., quoted,
166.
Satyrids, 37, 145.
Scale patches, 163.Index.
186
Scales, 97.
Scent in butterflies, 81, 129.
Scent-scales, 96.
Schwarz, C., quoted, 70.
Seasonal dimorphism, 167.
Secondary sexual characters,
101.
Secretory glands, 103, 160.
Selection, sexual, 96, 101,
Semnopsyche diana, 92.
Semper, C., quoted, 56.
Sexual characteristics, 13; di-
morphism, 162; selection,
96, 101, 131.
Skippers, 100, 138.
Smell, its function in butter-
flies, no.
Spring azure, 164, 169.
Stichophthalma, 104.
“ Struggle for existence,” 30,
65, 82.
Subdivisions of butterflies,
138, 150.
Succmcti, 144.
Suggestions for experiments,
177.
Suspensi, 144.
Suspension of chrysalids, 142.
Swallow-tails, 73, 100, 135.
Swarming in butterflies, 4 7.
Systematic zoology, 154.
Tachinid flies, 71.
Tasitia berenice, 83.
Tetraopes tetraophthalmus,
80.
Thanaos, 37.
Tiger Swallow-tail, 164.
Tongue, 23.
Tribes, 155.
Typical butterflies, 138.
Vaccinium as food plant, 127.
Variation in butterflies, 162.
Violet-tip, 67.
Vision, no; mistakes of, 122.
Walker, J. J., quoted, 57.
Wallace, A. R., quoted,
101.
Wanderings of Milkweed
Butterfly, 53.
Warning colors, 73.
Wasps, mistakes of vision in,
122.
Weismann, A., quoted, 103.
Wings, origin of, 158.
Zebra Swallow-tail, 167.