Title: Florida Entomologist
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00098813/00049
 Material Information
Title: Florida Entomologist
Physical Description: Serial
Creator: Florida Entomological Society
Publisher: Florida Entomological Society
Place of Publication: Winter Haven, Fla.
Publication Date: 1997
Copyright Date: 1917
Subject: Florida Entomological Society
Entomology -- Periodicals
Insects -- Florida
Insects -- Florida -- Periodicals
Insects -- Periodicals
General Note: Eigenfactor: Florida Entomologist: http://www.bioone.org/doi/full/10.1653/024.092.0401
 Record Information
Bibliographic ID: UF00098813
Volume ID: VID00049
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: Open Access

Full Text

Behavioral Ecology Symposium '96: Sivinski


USDA, ARS, Center for Medical, Agricultural and Veterinary Entomology
Gainesville, Florida, 32604

Entomology has always looked outward and attempted to apply its knowledge for
the public good. In many ways we believe ourselves to belong to a "service science",
standing in relationship to Zoology as Engineering does to Physics or Education to
Psychology. A "pragmatic", medical or agricultural application is in the back or fore
front of many of our minds as we pursue our interests in ion exchange across mem
branes or the relationship between light intensity and pheromone emissions.
I would like to mention a neglected set of consumers of insect information, a grow
ing and urbanized population increasingly alienated from nature. One that only elec
tronically experiences the once familiar, but now rapidly disappearing or impossibly
remote "ice-age fauna" it evolved with. It is my belief that we are "innately" interested
in the things that have been important to us through our evolutionary history. There
is an appetite for watching animals, uncovering the patterns of their activity, the se
crets of their lives. This appetite was critical to predicting the times and places deer
could be hunted and where bear-wolves were likely to be hunting our ancestors (could
our love of horror films be due to the pleasure of honing ancient anti-predator skills?
"You damn fool! Don't go in that door!"). Many of us, myself included, spend freely to
fulfill an emotional design and catch (and then release) unneeded fish. However, I
would suggest that our appetites are not specific for the great mammals and birds of
the Pleistocene's prairies or any particular animals of any other place and time. And
what animals are better suited for contemporary "hands on" natural history than in
sects? The pleasures of discovery are much more available to an insect observer than
to a tourist watching a patch of elk hair disappear into a stand of pines.
Some of us already devote some of our energies to "public" education, and while I
can't know other's motives, it is my impression that much of it is done to explain our
"business". I would like to propose that we at least consider a change of heart; that we
grant as much respect to the fulfillment of our culture's emotional-spiritual needs as
we do to the patent of an attractant or the publication of scholarly work. The natural
historian, a person with a net, a flower press and a curiosity about the colors of beetles
and the poses of flies, should not strike us as eccentric but as profoundly purposeful.
The participants in this year's Behavioral Ecology Symposium would all admit to
being naturalists. In general, their topics concern themselves with "adaptive color
tion" defined in its broadest sense. I will address the often fantastic ornaments used
by flies to intimidate sexual rivals and woo mates. There will be a number of peculiar
curiosities discussed, obscure insects of no economic importance, some described by
bemused 19th century travelers and then forgotten. In light of the contemporary con
cerns of entomology, I would like to briefly defend "curiosities" and offer you a reason
to spend your time pondering insects that will never take a bite from a cabbage or in
ject a spirochete.
I perceive the sexual ornaments of flies to send a special message to human receive
ers. They bring to us news of intellectual liberation. By that I mean that their combi
nation of the marvelous and the mundane reminds us that the world is a "very strange
place." Rare curiosities are not trite, but points where that strangeness has come to
the surface-as we see the surface. In my studies I sometimes find myself falling into
a pitfall that Darwin warned against, that I base my hypotheses on what seems plau

120 Florida Entomologist 80(2) June, 1997

sible. Occasionally this model or that interpretation is dismissed, not on its merits,
but because it is too challenging to the imagination. It is easy to become overly skep
tical and stodgy. If I catch myself, I turn to a specimen of the truly bizarre Achias
(Diptera) I keep on my desk. Here is an animal I couldn't even make up! Achias is dis
cussed in the following, as are a number of other illuminating peculiarities. I hope
that in addition to its other merits this symposium can serve, like a Zen parable, as
an aspirin to treat a swollen and painful "common sense."


Florida Entomologist 80(2)


Department of Entomology and Nematology
University of Florida, Gainesville


Research at institutions of higher education could be restored to at least a shadow
of its original role through publication in a manner appropriate for immediate class
room use, with questions that pique and direct the interests and activities of students.
Studies on basic natural history may be good candidates for such publication and an
example is drawn from fireflies: Two woodland species show directional orientation in
their pupation sites on the trunks of trees; one uses southerly exposure and the other
occurs on the north side of smaller trees, and much lower on the trunks. These con
trasting positions have different thermal consequences, as demonstrated with a phys
ical model, which possibly have a role in reducing interspecific sexual contact or prey

Key Words: fireflies, behavior, life history, orientation, ecology


La investigaci6n en instituciones de educaci6n avanzada podria ser restaurada
parcialmente a su rol original a traves de publicaciones, de manera tal que las mismas
puedan ser usadas para ensenar, con preguntas que atraigan el interns de estudiantes
y que se relacionen con sus actividades. Los studios de historic natural basica pue
den ser buenos candidates para ese tipo de publicaciones, y un ejemplo del mismo se
puede obtener con luci6rnagas: Dos species de luci6rnagas muestran diferencias en
la ubicaci6n de sus pupas en los troncos de los arboles; una especie las ubica expuestas
hacia el sur y la otra usa el lado norte de arboles mas pequenos y en la zona mas baja
del tronco. Estas posiciones contrastantes tienen diferentes consecuencias t6rmicas,
como se demuestra con un modelo fisico, las cuales podrian tener un papel en reducir
el contact sexual o la competencia por alimento entire las dos species.

June, 1997

Behavioral Ecology Symposium '96: Lloyd

In times past it went without question that the connection between research and
teaching was that professors who did basic research maintained their intellectual in
terest in scholarship and passed on to their students an inquisitive attitude and love
of the pursuit of knowledge as the essence of life and a life-sustaining spirit. Students
thus became living repositories of what was then acknowledged to be a civilizing Ideal
of western culture. An academician of the time translated the expression "publish or
perish" as meaning that if he did not publish he had mentally perished, and in doing
so was failing in his professional responsibilities to his students and his civilization.
Over the past 30 years this fundamental understanding and connection has been
eroded and forgotten, and a great deal of what is now done as "scholarly publication"
has little direct bearing on a "civilizing education."
The essence of scholarly research is discovery and originality. In my experience,
good students find it more interesting to actively participate in doing something that
relates to discovery than to see someone else do it on TV. It is worth exploring to de
termine whether some primary publications in science could be written directly for
the classroom, rather than for the narrow and generally disinterested "readership" of
a scientific journal, even leaving some obvious refinements for students to manage.
Original research papers could be used as texts, and beginning students have direct
contact with researchers themselves-who could speak directly to them in their pa
pers, and then perhaps personally through the internet, thus achieving a quasi-oral
tradition of wide dimensions! Students would use an original publication as a source
of information and to stimulate their imaginations for initiating their own school-time
and life-time pass-time research. What once might have been a scarcely read, esoteric
and expensive "contribution to . ." could be an informative introduction and back
ground with suggestions and questions for personal projects and class discussion.
Though it pains me to admit it, fans of electronic publication may be the first to see
the desirability and simplicity of doing this.
There is another twist to this notion. Since I have chased fireflies for about a third
of a century, I am often asked by citizens and reporters, by letter and phone, "what is
happening to the fireflies, I don't see them anymore?" Only people who once knew and
pursued fireflies can ask such a question, because those who have never known them
cannot miss them. Similarly, might not students who learn by reading and doing orig
inal research and see it in connection with their personal education, understand and
care more about what we have long considered to be the intellectual values and
strengths of an enlightened civilization? The irony, the flip side of this is that here I
address this notion to many who have never seen a firefly.
Obviously, some research subjects lend themselves to such instruction better than
others, because of technical complexity and expense, but there are many available
sources of inspiration. As John Sivinski has pointed out, one unfailing repository of
observations and ideas worth developing are the anecdotes, sketches, and specula
tions that insect naturalists accumulate. From my search for new sources and angles,
I would add that many taxonomists especially know what is lost to lab-bound and ur
ban biologists, because of their solitary hours of collecting and observing their quarry
in the field, which are as basic field investigations, typically followed by solo hours of
contemplation as they curate their specimens. I have found that much of what can be
done with firefly taxonomy and behavior can be used almost immediately in the class
room. It should be as a personal goal and measure of scholarly accomplishment and
fulfillment to see the development of some significant area of insect research begun
and developed by undergraduate students in a teaching/research connection. Think of
the satisfaction that graduates would enjoy when they subsequently saw their own
studies used in a general entomology text.

Florida Entomologist 80(2)

For several years I have taught a general biology course entitled Biology and Nat
ural History With Fireflies in which every class meeting is a field trip or lab and in
volves some research-related activity. Instead of giving oral lectures, I write the
students letters; instead of laboratory and field exercises with recipes and empty lines
to write on, I give them a background text on a subject, the material and equipment
they may want to use, and directions so they can do some things they will find inter
testing. English, religion, architecture, microbiology, German literature, journalism,
pre med., and animal science majors, to mention a few of the represented fields, expe
rience first hand the basics of biological research, including the design of empirical
studies and the gathering of data, the use of statistical analysis, and the value of mod
els and theoretical perspective. During class meetings students are only required to
be focused and interested, and try to accomplish what they recognize with increasing
skill as sound biology.
As an example, the "Letter" below provides the introduction and background for a
number of field studies that students can make in winter in a flood plain forest in
Gainesville, about two miles from the indoor classroom. The Letter is modified for use
here. Scientifically, this Letter is the first publication of the outlines of a seemingly
simple but perhaps very complex element of firefly biology. The Letter omits statisti
cal descriptions and analyses, which are a field/lab experience themselves, but illus
trates the observations and raises questions that students anywhere in the
geographic range of the species can discuss and independently or jointly pursue in the
lab and woods (Fig. 1). More than this, when students begin to address specific ques
tions about this apparently simple behavior of mere beetle larvae, they discover that
it is potentially so complex that it may never be completely understood, and for them
this itself is encouragement to continue, to enjoy the study, and sometimes to see such
biology as also of the arts and humanities.


Dear Fireflyers, When fireflies and light are mentioned in the same breath, one re
flexively thinks bioluminescence, and of the use that fireflies and taxonomists have
made of pulses of living light for species recognition, that behavioral ecologists have
made of firefly flashes for studying mate competition and mate choice, and finally, of
the use that biochemists, cell biologists, and physicians now make of bioluminescence
chemistry for enzyme analysis, cell physiology, exobiology (extraterrestrial life
searches), and medical diagnoses. Our knowledge of firefly flash communication in na
ture began with the incidental observations of a chemist, Frank McDermott, who
went to the field to observe fireflies out of an interest in the mechanism of their lumi
nosity, but stayed to discover that some lightningbug species can be distinguished by
their flashed mating signals. What I will tell here began with a taxonomist's interest
in getting a photograph, and became an enigma in the realm of what some might call
environmental physiology. It is about a connection that some fireflies have with light
other than through their remarkable ability to generate it.
The larvae of one species may use sunlight to hasten or perhaps, maybe, even to
manipulate their pupal duration and adult eclosion time ("date"). Pyractomena fire
flies, and perhaps all of the fireflies in their tribe (Cratomorphini), unlike other
lampyrids that do it in hidden chambers underground, climb up on vegetation to pu
pate. Aerial pupation was reported by Francis Williams near the beginning of the
passing century and observed in some detail by Lawrent Buschman, who examined
this behavior in the marsh-inhabiting species Pyractomena lucifera (Melsheimer).
Aerial pupation would seem to be a reasonable adaptation for larvae that live on

June, 1997

Behavioral Ecology Symposium '96: Lloyd

emergent vegetation over water and hunt the aquatic snails below, or that could have
their habitat submerged by the flood water of a creek or river spilled out of its banks
onto adjacent flood plain. Pyractomena borealis (Randall) pupae hang on tree trunks,
by means of laterally projecting points that extend into their cast larval skins they
previously glued to the trunk by the tail-end. At eclosion, the pearly-white, general
adults walk a few centimeters leaving behind the larval and pupal skins and dangling
tracheal linings, and remain motionless until their cuticle has tanned. Sometimes
adult males are found waiting next to or on top of pupae (female only?; Fig. 2).
In the winter of 1982-83 I visited the flood plain forest along Possum Creek in
Gainesville to get photographs of pupating Pyractomena borealis, whose adults I had
seen flying and flashing there in considerable numbers the previous March. I found
one, then several, then numbers of them, and it soon became obvious that they did not
occur randomly over the tree trunks. Sometimes pupae occurred together, sometimes
alongside vines or in crevices, and occasionally below twig bases. They used trees of
several species and bark textures, usually anchoring themselves between knee and
basketball-rim height. I returned again and again for more photographs, notes, and
measurements of pupation locations. Then, larvae and pupae of another woodland

...,.*............ .0- i "
-.:.-.--------:- --.------- ....
--.----------------. J _-I o...L.......

^- ------------
:::V .-.-.-.- .-.-.------

-- ------------- ----jo limbicollis--
-------------:.'-.-.-.-.-- .'-.-: .'-- oO :v':v..:.-..v-----------------! ii

,--.---.--.-."-. ..... ... ... bor.alis............ .....
---- - --------------------

Fig. 1. Locations of specimen label records for P. borealis and P. limbicollis from
several North American collections. Woodland Pyractomena species in addition to
these two probably also pupate up on the trunks of trees or shrubs.


124 Florida Entomologist 80(2) June, 1997

Fig. 2 Male borealis with a borealis pupa, sex unknown.

-. -

species, Pyractomena limbicollis Green, began to appear up on trees and in many re-

spects this species was as a foil for P borealis, providing a useful and informative and
certainly puzzling contrast.

P. borealis pupae show a surprising directional orientation in their choice of pupa-
tion sites on the trees. In a sample of 240 pupae during three winters, the mean direc-
tion sittis on the triis. In a sample of 240 pupai during thrii winttirs, the mtian dirit.

Behavioral Ecology Symposium '96: Lloyd 125

70 I I

60- I Pyractomena borealis
I n = 240 pupae
50 I( (@ 30 intervals)

40 -

4 -I I I

o I

10 I

1-30 6t-900 121-150* 181-210' 241-270' 301-330P
31-6W 91-120P 151-180 221-240" 271-300* 331-360
intervals Position (true*) of Pupae On Trees ,* = compass-20

Fig. 3. Directional orientation on tree trunks of P borealis pupae during three win
ters, at the Possum Creek-Hog Town Creek flood plain site.

tion was southerly, that is, about 180 true (= compass -2; Fig. 3). But sunlight is
more than illumination and a suitable directional cue for orientation-if indeed the
larvae are using sunlight for orientation-because it warms what it shines upon. By
choosing a pupation site at or near the south side of trees in January, when ambient
temperature may be low for many days and even drop below freezing, P borealis pu
pae raise their body temperature during pupal development by several degrees, pre
sumably decreasing the duration of pupation. One potentially dangerous thermal
consequence of the sun-exposing behavior of P borealis is that they must be able to
survive extreme temperature changes over a very short period of time; on a clear and
sunny winter day the temperature of a dark-barked tree may reach over 90 F (32 C)
at three in the afternoon, and by midnight drop well below freezing (32 F, 0 C). One
wonders how they manage this!
Pupation up on trees has another conspicuous variable that has thermal conse
quences. Were the adaptive significance of aerial pupation merely the avoidance of ris
ing flood water, we might expect their vertical distribution on the trees to be rather
limited, with pupal distribution clumped around some height-perhaps just above a
residual high-water mark left by previous flooding, possibly cueing upon chemical res
idues left by the water, or algal growth encouraged by flood borne nutrients. Not so;
the vertical distribution has considerable spread (Fig. 4). Height may have thermal
significance because (1) in winter the ground below may be a heat sink and have a ten
dency to hold lower-trunk temperatures down, and (2) with increasing altitude there
is less shading from sunlight by the trunks, branches and leafless twigs of adjacent

Florida Entomologist 80(2)

June, 1997

4 -
61-70 Pyractomena borealis
601-650 n =240

0 5 10 15 20 25 30 35 40 45 50
Number of Pupae Observed

Fig. 4. The height of P borealis pupae on tree trunks.

trees. Obviously then, vertical as well as circumferential positioning on a tree could
potentially be used by larvae for manipulating the timing of their metamorphoses.
And, there are other possible though more subtle influences on the thermal relations
of these pupae. For example, larvae use different species of trees, species that vary in
the smoothness of their bark and in the water content of their wood, and these are
probably not independent in their effects.
The bark on beech trees is smooth and presents few cliffs and side-directing chan
nels; the bark on oak is rough, with the crevices seemingly the equivalent of four story
buildings and presenting an obstacle course for short-legged, prostrate larvae. I com
paratively ranked the bark of each tree that larvae selected for the energy and time I
expected would be required to climb over (up) them. Beech and sugarberry were typ
ically toward the least expensive end of the ranking, and red maple and oak were at
the most expensive end. In consideration of the difficulty of climbing, one would ex
pect that pupae might be found higher on smooth than on rough trees, and perhaps
there would be fewer of them. This is what I observed. Trees with smoother bark had
more, and species with coarser bark had fewer pupae and they were not as high on the
trees (Fig. 5).
Because trunks of different tree species vary in their water content, in sunshine a
tree with more water will take longer to warm up, and remain warm longer into a cool
ing winter evening. Tree-water will also dampen temperature changes, preventing
rapid extremes-only two pupae were found on dead (dried out?) trunks. Bark coarse
ness and thickness could have an influence through the insulation it places between
a hanging pupa and the warm water held in the tissues of the trees. On the other
hand, rough bark and its crevices provide protective and perhaps thermally amplified
niches that provide dead air pockets and radiating walls.

Behavioral Ecology Symposium '96: Lloyd

Bark Score 1 1.5 2 2.5 3 4
b 690

M 2510
450 .
: _390
Z 210
n -Scale 0 15 30 0 8 16 0 3 6 0 2 4 0 3 6 0 1 2
Number 101 69 17 8 12 3
Mean 261 237 239 171 154 103

Fig. 5. The height of P borealis pupae on trees with different bark roughness.

' I ' I m . I . I . . I ;* 0
100 200 300 400 500 600 70
Minutes Elapsed Since Sunrise sunset



a 20

C 16

4 12



S-(-C N (CO) ----A--
-0-0-0- w (co)
S (C0) *

-D--D-- E (C)

air (C') ------***

0 1 O sun intensity (light meter units) -0-0-0-


Fig. 6. The basic physical model of a tree with pupae. The tree was a photographic
chemical jug filled with dry sand, painted flat black up to the sand level; the model
fireflies were 1 cm clay spheres, painted black, each with a thermocouple inside.

Florida Entomologist 80(2)

Wet vs Dry Jugs (south clay pupae) U u
Note wet jug less extreme & variable, sunset

/0 100 200 300 400 500 600 700 800
sunrise Minutes Elaspsed Since Sunrise 26 Febmary 1988

Fig. 7. The comparison of temperatures of model pupae at north and south posi
tions on a dry-sand jug and a wet-sand jug; a physical model examining the influence
of tree water content on pupal temperature.

Questions of water content and heat storage can be explored with a simple physi
cal model. I made artificial tree trunks of plastic jugs used to store photographic dark
room chemicals, and hung them in the sun on cool winter days. Each bottle had a 1 cm
clay sphere with a thermocouple inside, at each of four directions (N, S, E, W); spheres
were painted flat (i.e., not enamel) black and held against the surface of their jug with
an elastic band around the jugs and passing over the thermocouple wires. Jugs were
of two "trunk" sizes, some contained dry sand and some water-saturated sand, some
were hung near the ground and others more than a meter above the ground. Results
were generally as expected. Figure 6 shows the temperatures recorded from the basic
physical model, a large dry-sand jug, on a cold winter day, with air temperature for
comparison, and also sunlight intensity as measured with a photographic exposure
(visible light) meter.
Note that the temperature/time courses of clay spheres (model pupae) on different
sides of a tree are not the same: the S (south) clay sphere (black dots) warmed more
and climbed from freezing to nearly 28 C; the N sphere (open triangles) closely fol
lowed air temperature; and that a brief shading at 460 min. affected the S and W
spheres but the E and N spheres scarcely if at all. Many comparisons among such
spheres andjugs are possible; Figure 7 shows temperature/time plots for N and S clay
pupae on wet and dry jugs, with the moderating effect and thermal gain from "tree
water." However, one photographed pupa was discovered to be conspicuously arched
out away from the tree, suggesting that it should not be presumed that pupae fas
tened to trees have no control over their body temperature; perhaps they press

June, 1997

Behavioral Ecology Symposium '96: Lloyd


n=73 150
borelis pupae:
I per tree

B C D 30

n=51 s n 28 t w n-63
borealis larvae borealis pupae & larva: limbhiollis pupae: all
that did not remaL L tree in protctd sites

O Males,n=13
* Females, n=34



0 '30 60' 90 1'2o' IS0' 1210'24' 270' 330
Azimuth On Tree (degrees, True N)
nrdpoins. 30W, iatrvals
4 --- ..

bomalis, n=220
|0 limbicollis, a60


0 20 40 60 80 100 120 140



l Males, n-13
U Females, n=47



' soo'30 50 70 90'no' 1131501o'TC o1 o'232o
Height On Tree (CM)
midpoMs. 20 cm iHervals

0 borealis. n-220
limbicollis, n=61

3, 9-5 15-5 21.5 27- 33.5 39. 45.5
DBH (tree diameter. CM)
midpoint, 6 cm interv.1l

Fig. 8. Graphs illustrating data that are pertinent to some basic questions about
P borealis pupation biology, and the remarkably contrasting behavior of P limbicollis.
(A) Azimuth positions of solitary P borealis pupae that presumably were not influ
enced by others; (B) Azimuths of P borealis larvae that did not remain in position,
showing that they abandoned what would seem to be a good angle-though they may
have moved to fine-tune their positioning(?); (C) positions of P borealis larvae and pu
pae situated in sheltered locations showing that the shelters did not have highly de
viant azimuths; (D) The north-easterly azimuth orientation of P limbicollis pupae; (E,
F) Azimuth and height positions of male and female P borealis. (G) Heights of pupal
positions of both species; (H) Trunk diameters (DBH, diameter breast height) of pu
pation trees of both species.

against a warm tree to warm up, or arch out away to cool down by increasing air in
sulation and circulation between them and their too-warm tree.
The behavior of these juvenile fireflies raises many questions that students can ap
proach. Do larvae actually manipulate with some precision their thermal gains from


^ 3014

Florida Entomologist 80(2)

azimuth and height?-how about thermal conditions in pockets between the ridges of
a muscle tree (Carolina beech)? Would a larva select a pupation site 15 from a "pre
cise target position" or "ideal directional site," if other pupae or a sheltering vine were
positioned there? Could a P borealis juvenile be expected to integrate all or some of
the variables noted or discussed, to control the moment when it, as an adult enters the
competitive reproductive environment? Would a male-to-be larva that was late get
ting to a tree accelerate its development? Of course it would be absurd to ask whether
a larva could control its gender by adjusting its developmental temperature.
Fundamental to comparing observations and sets of observations, and of interest
to the mathematically-minded, note the problem of calculating statistical descriptions
such as mean positions and amount of spread in circular data, that is, of angular po
sitions around a tree-consider this: the average position of a pupa 5 west of north
and another 5 east of north, is half of 355' + 005 and thus 180', which is true south!
Nor is it simple and straightforward to compare the means and deviations (spread) of
samples to determine the likelihood that they are "identical" (drawn from the same
population). Were my samples properly made?-my data show that more larvae
climbed smooth-barked trees (Fig. 5), but were there more smooth trees in the woods;
but, perhaps it is not relative abundance that should be considered, but rather the
identity of nearest neighbors to trees actually climbed, because individual larvae may
not move far in the days or weeks before pupation. If you are interested in physics or
photo-journalism, can you suggest a better method of measuring insolation (solar ra
diation), or a way to see infrared patterns on and among the trunks of the trees that
might be available to tree-seeking larvae?
Figure 8 illustrates data that bear on several questions: do azimuths of solitary P
borealis pupae show the same directionality? (Fig. 8A); did hanging larvae that sub
sequently moved, have the same near-southern azimuth? (Fig. 8B)-this question of
course relates to the (proximate) mechanism of orientation; do solitary larvae and pu
pae that occur in protected sites deviate appreciably from an approximate southern
azimuth? (Fig. 8C).
On several occasions I found adult P borealis males attending pupae (Fig. 2). This
raises questions related to mate finding and competition: are males able to recognize
female pupae?; would guarding a sexually unidentified pupa have a better long run
payoff than searching with a signal light at night, and would this probability and pay
off change through the mating season?; might males accelerate their eclosion to ap
pear earlier in the season to be ahead of and be waiting for unfertilized (high value)
females? This last speculation presently finds no support in the azimuth and height
data, assuming that accelerating males would show different pupation azimuths and
heights than females (Fig. 8E and F). Perhaps P borealis fireflies in north central
Florida accelerate their seasonal appearance to avoid predaceous Photuris species,
which pupate in the soil and thus are stuck in a cold cellar.
The pupation behavior of the smaller species P limbicollis stands in such contrast
to that of P borealis that it reinforces the suspicion that there really is something sig
nificant occurring in P borealis, providing both encouragement to proceed and an
other firefly subject for a comparative study. In my sample, P limbicollis pupated
toward the north (Fig. 8D) and much lower on smaller trees (Fig. 8G and H)-being
low down on the north side of small trees would result in a cooler-than-air tempera
ture regime.
The adult season of P limbicollis is about three weeks later than that of P borealis,
and limbicollis adults appear with a versatile firefly predator belonging to the Pho
turis versicolor complex. The (sexual) flash pattern of P limbicollis males is virtually
identical with one flash pattern emitted by the males of this Photuris, an instance of
the pattern-matching phenomenon seen in males of many Photuris species. What

June, 1997

Behavioral Ecology Symposium '96: Lloyd

would P limbicollis gain by synchronizing with a pattern-mimicking predator, or is
limbicollis manipulating its adult season to avoid a critical seasonal overlap with its
congener P borealis? If this is the case, is the avoided overlap that with mate-seeking
adults or with first instar larvae that must find soft-bodied and perhaps only minute
gastropod prey in the same forest litter?
These fireflies clearly present sufficient questions with respect to proximate mech
anisms and ultimate consequences, to provide fireflyers many years of intriguing "off
season" field work. Find quiet and mysterious trails.


I thank Robin Goodson, John Sivinski, and Steve Wing for reading and insightful
comments on the manuscript, several classes of firefly students for fresh views of fire
flies and research, and Flora MacColl for additional editorial assistance. Journal Se
ries No. R-05567.


ANDERSON, CHARLES H., AND JOHN D. MURRAY, (Eds.). 1971. The Professors: Work
and life styles among academicians. Schenkman Pub. Co., Inc., Cambridge
BOOTH, WAYNE C. 1988. The vocation of a teacher: rhetorical occasions 1967-1988,
353 pp. University of Chicago Press, Chicago.
BUSCHMAN, LAWRENT L. 1984. Biology of the firefly Pyractomena lucifera (Coleoptera:
Lampyridae). Florida Entomol. 67: 529-542.
GREEN, JOHN WAGONER 1957. Revision of the Nearctic Species of Pyractomena (Co
leoptera: Lampyridae). Wasmann Journal Biology 15: 237-284.
HEINRICH, BERND. (Ed). 1981. Insect thermoregulation. Wiley, New York.
LLOYD, J. E. 1997. Firefly Mating Ecology, Selection, and Evolution, Chapter 10 [In]
The Evolution of Mating Systems in Insects and Arachnids. (Ed) J. Choe and B.
Crespi. Cambridge Univ. Press. United Kingdom.
OLEKSA, JAMES. 1996. Fireflies: beauty and beyond. Fireflyer Companion 1 (2): 23-24.
WILLIAMS, FRANCIS X. 1917. Notes on the life-history of some North American Lampy
ridae. Journal of the New York Entomological Society 25: 11 33.
ZAR, JERROLD H. 1984. Biostatistical Analysis. 718 pp., [especially Chapters 24 and
25], Prentice-Hall, Inc., Englewood Cliffs, New Jersey.

Florida Entomologist 80(2)


Entomology & Nematology Department, University of Florida
Gainesville, Florida 32611


Iridescence, in both the visible and ultraviolet (UV) spectra, is produced by various
means and may serve several functions in different animals. In insects, such colors
are often considered as anti-predator adaptations, either crypsis or aposematism, or
a means of thermoregulation. A less explored alternative is social signaling. Irides
cent colors are particularly useful in this context because they are brightest from cer
tain directions and body orientation could be employed to direct a visual signal to
particular receivers. In phanaeine dung beetles the head and prothorasic shield re
flect a visible-light and UV iridescence that is best seen from a position facing the in
sect. The less iridescent male horn is silhouetted against the prothorasic shield. Since
horn size is indicative of male size, such a display may be directed to sexual competi
tors in agonistic interactions. Broad and reflective prothorasic surfaces on males
might also be preferred by females choosing a mate, who will cooperate in future
brood care, since they would make infestations of kleptoparasitic flies more obvious.

Key Words: Scarabaeidae, mate choice, intrasexual selection, ultraviolet reflectance,


La iridiscencia, en ambos espectros, visible y ultavioleta (UV), es producida de di
versas maneras y puede ejercer diversas funciones en diferentes species animals. En
insects, dichos colors generalmente son considerados como adaptaciones biol6gicas
para protegerse de sus depredadores por mecanismos cripticos o de aposematismo, o
como una forma de termoregulaci6n. Otra alternative, menos estudiada, es la iridiscen
cia como un medio de comunicaci6n social. Los colors iridiscentes son particularmente
tiles en este context porque son demasiado brillantes desde ciertas direcciones y la
orientaci6n corporal pudiera ser empleada para dirigir una senal visual a receptores
particulares. En los escarabajos de esti6rcol (Phanaeine), la cabeza y la coraza protora-
cica reflejan una luz visible y una iridiscencia ultraviolet que se observa mejor desde
una posici6n de frente al insecto. El cuerno de los machos, un poco menos iridiscente,
forma una silueta contra la coraza protoracica. Si consideramos que el tamano del
cuerno del macho refleja el tamano corporal, este mecanismo pudiera ser dirigido a
competidores sexuales en interacciones agonistas. Las superficies protoracicas anchas
y reflejantes presents en los machos, pudieran tambien ser preferidas por hembras eli
giendo su pareja sexual, quienes cooperaran en el cuidado future de su progenie, puesto
que pudieran hacer mas obvias las infestaciones de moscas cleptoparasitas.

Iridescence is found in many organisms, but among terrestrial animals it is most
highly developed in two groups, birds and insects. Perhaps not coincidentally, these
classes also exhibit well developed visual systems, and protean body coverings. The two
groups frequently interact; birds are among the principal predators of insects, and irides
cent species of both are largely diurnal, suggesting that these colors are used in interspe
cific and/or intraspecific communication. In this paper, the mechanics of iridescence are
briefly described, as are some of the different structures that cause iridescence. Different

June, 1997

Behavioral Ecology Symposium '96: Vulinec

hypotheses are then proposed for the evolution of iridescent coloration, and each hypoth
esis is considered in relation to iridescence in dung beetles (Coleoptera: Scarabaeidae).
Animal coloration often correlates with a species visual capabilities. Mammals are
typically colored with shades of brown and black, the hues of melanin. Most apparently
do not see color, or do not respond to color stimuli. Primates are an exception, and not
only respond to color but are often brightly colored themselves, e.g., the faces and rumps
of mandrills, (Mandrillus sphin). Humans generally see wavelengths between 400 and
790 nm (referred to from this point as the "visible" spectrum; Endler 1990). Birds per
ceive not only the colors visible to humans, but also detect ultraviolet colors with wave
lengths shorter than 400 nm (Goldsmith 1980; Parrish et al. 1984). Bees perceive
wavelengths from the UV range up to 650 nm, but do not distinguish orange (600-650
nm) from yellow (550-600 nm), or blue (400-480 nm) from violet (380-400). Other in
sects' visual systems vary. For example, the absorbance maxima of the photo pigments
in some moth eyes are 345, 440 and 520, while the peaks in Heliconius butterflies are
350, 460 and 550 (Endler 1990). These differences may be due to differences in available
light and other components of their respective environments (e. g., Lall et al. 1980).
Browns, reds, and yellows in animals are almost always formed by pigmentation,
and can be washed out of the underlying structure with solvents. Blue and green col
ors are usually structural and cannot be permanently changed unless the structure it
self is crushed.
Structural colors can be produced in two ways. One is through diffusion, i.e., the
scattering of short wave colors, blue and violet, by submicroscopic particles, that re
sults in Tyndall blue. The blues of the sky and human eyes are formed this way (Si
mon 1971), as are some blues on butterfly scales (Huxley 1976).
The other structural means of color production is through interference, which
causes the brilliant changing hues common in iridescent insects. Thin films, such as
oil on water, reflect some incoming light from their shiny top surfaces. The rest of the
light enters the film and is refracted by the film's greater density compared to air. This
light then slows as it passes through the film, and when it reaches the lower surface,
is reflected back. When it rejoins the light reflected off the upper surface, it has been
traveling slower, and is thus out of phase with the reflected beam of light. If the phase
difference between the two beams equals one full wavelength, or a multiple thereof,
the color of that particular wavelength will be reinforced. If the amplitudes (crests
and troughs) of the two light beams are equal, reinforcement will be strongest, and
the color purest. All other wavelengths are either weakened, if they are out of phase,
or eliminated if the crest of one beam meets the trough of the other. If the angle of in
cident light is changed, a different color will appear. Changing the width of the film
will also select for different wavelengths, and thus different colors (Simon 1971).
Thin films are not the only way to obtain interference colors. Thin slits arranged
equidistant from each other, called diffraction gratings, also cause iridescence (Hinton
1973). Additionally, a structure called a space lattice, where minute particles sus
pended in a medium are arranged in layers stacked on top of each other, produces iri
descent reflections. The microscopic structure of iridescent bird feathers are made up
of stacks of melanin rods within layers of keratin, creating a space lattice (Simon 1971).



Of all soft body coverings bird feathers are the most strikingly iridescent. Many
birds are largely iridescent, such as the Resplendent Quetzal (Pharomachrus
mocino). Others are dull, but exhibit patches of iridescent feathers. These patches, in

Florida Entomologist 80(2)

otherwise dull-colored birds, also strongly reflect ultraviolet wavelengths (Radwan
1993). As will be seen in insects, iridescent patches are highly directional, appearing
brightest from particular angles of view.

Butterfly scales

In butterflies, several types of iridescent scales have been described. The average
lepidopteran wing has rows of alternate long and short scales. The longer are cover
scales, which arch over and hide the short, ground scales. In iridescent patches, the
cover scales are specialized, but the ground scales are usually undifferentiated. Iri
descence may arise from lamellarr thin-film iridescent" scales, "microrib thin-film ir
descent" scales, laminarr thin-film iridescent" scales, or "diffraction lattice" scales,
whose interiors are filled with crystals of a cubic lattice that produces a diffraction
color (Ghiradella 1985). These various ways of producing iridescence can be readily
modified, as demonstrated by ultraviolet reflectance in Colias, where the lamellar
thin-film color is inherited at a single locus (Silberglied & Taylor 1973). Because of
this plasticity and that several scale types can be found in taxa without any particular
correlation to phylogenetic associations, iridescence in butterflies probably evolves in
response to selection (Ghiradella 1985).
Some butterflies have intense UV reflection caused by interference, and produced
in the same manner as visible iridescence. This ultraviolet reflectance can overlay vis
ible colors (Silberglied 1979). Additionally, most iridescent scales also contain mela
nin, which absorbs much of the light not reflected by the iridescence and enhances the
brilliance of the color (Nijhout 1991). The intensity of most colored surfaces varies lin
early with the angle between the light source, the reflecting surface, and the observer
(Endler 1990). With interference colors, reflectance at a given wavelength "cuts on"
and "cuts off" more abruptly, and the peak wavelength (that is, the color), shifts with
changes in the angle (Silberglied 1979). Additionally, at certain angles the reflected
light will be highly polarized. When flying Colias eurytheme and other species are ob
served through a UV viewing device, they resemble flashing beacons (Silberglied
1979). Crane (1954) writes: "With every wingbeat, a flying Morhpo butterfly changes
the angle of light incidence through the entire possible range. To the human eye, a
Morhpo in flight is simply a flickering flash of varying tints of blue. However, to an
other Morhpo, in sunlight, there should be a brilliant shift from blue-green or blue to
ultraviolet, then momentary extinction and back again through the spectral arc; con
ceivably this may be an exceptionally potent stimulus. The well known dipping of
these butterflies to blue papers and other objects I I I. 1:, 1i. ii the wing color
may prove to be a sign stimulus in inter-male or courtship behavior."


In some beetles that live under bark the microsculpture in the cuticle may produce
a type of iridescence. This microsculpture has a characteristic orientation and asym
metrical sculpture, and is thought to be a by-product of the frictional properties of the
cuticle (Crowson 1981). The more common bright iridescence seen in many Coleoptera
is produced from light interference in thin films in the endocuticle. As with iridescence
in feathers, and butterfly scales, these colors vary with the direction of incident light.
The most frequent color is metallic green, but blue, red, gold, and purple are also com
mon (Hinton 1973).
Colors are often a result of an animal's relation to activity and habitat. Green iri
descence typically occurs in diurnal, leaf feeding beetles (Crowson 1981). Beetles that

June, 1997

Behavioral Ecology Symposium '96: Vulinec

are obligate cave dwellers are pale brown, not black, suggesting that this shade is the
natural color when no selection for color occurs. Generally, nocturnal beetles are black
(Crowson 1981).
One group of new world, scarab dung beetles, the phanaeine, is known for irides
cent colors, diurnal habits, and conspicuous behavior (e.g., Edmonds 1994). A conven
tional interpretation for this suite of characters is that the beetles may be bad tasting,
and advertise their unpalatability to bird predators who subsequently avoid them
(Arrow 1951). However, evidence for this hypothesis is minimal, and several other ar
guments for the adaptive benefit of iridescence in these beetles can be invoked with
equal conviction.


Not all examples of iridescence in animals may be adaptive. For example, some fly
larvae infected with a particular virus become iridescent. Unless iridescence attracts
new hosts or agents of dispersal, such coloration is probably an artifact and has no se
lective advantage. But given the striking apparancy of iridescence in diurnal dung
beetles and other insects, it is reasonable to investigate adaptive hypotheses for their


Dung beetles of many species perch on leaves in tropical forests. While much of
this behavior is related to foraging, one beetle species is thought to perch as a way of
regulating body temperature (Young 1984). This beetle, however, is dull black. Bright,
large scarabs probably possess internal mechanism that allow for large increases in
body temperature prior to flight, and sun-basking in these beetles is not necessary
(Young 1984). Further, iridescence reflects light, rather than absorbing it. Possibly, ir
idescence might serve to prevent overheating, allowing diurnal insects to forage in
open habitats. Brilliantly colored species of phanaeines are found in both forests and
more open habitats (Edmonds 1994).

Distracting glare

Hinton (1973) argues that diffraction gratings can produce warning colors, and be
cause some of the light reflected is of the complete spectrum, will also produce intense
glare. This glare might prevent a predator from judging the exact distance of the an


Endler (1990) stressed that the conspicuousness of an animal in its environment
is a function of the receiver's visual system, and the intensity, hue, saturation, and de
gree of contrast between different patches on the animal and its environment. What
may be described as bright when seen out of context by humans may actually be cryp
tic in its environment. Many bright green iridescent leaf beetles could be cryptic to
avian predators; e.g., the iridescence might resemble dew on leaves (Crowson 1981).

Visual signals

Colors can be used to pass information visually from one organism to another. In
the case of insect iridescence, signals are most likely directed at either conspecifics or

Florida Entomologist 80(2)

at diurnal predators. Bats do not use vision in hunting, and generally, defense against
these predators involves interference with their sonar, or evasive action (Dunning &
Roeder 1965). Nocturnal mammals usually hunt by smell, and most diurnal mam-
mals are not thought to have color vision. The main predators that would encounter
visual signals from prey are birds, some reptiles, and other insects (Crowson 1981).
The physiology of bird sight is well known, but behavioral responses to specific col
ors are not. For example, there have been a number of studies examining birds' reac
tion to signals in the ultraviolet spectra. Birds have been known to have receptors
sensitive to UV wavelengths for over 20 years (Bennett et al. 1996), but controlled
studies to determine if they respond behaviorally to these frequencies are rare. Birds
can distinguish between visible light that differs only in the presence or absence of the
UV component (Goldsmith 1980). Studies using filters that screen out particular
wavelengths in choice experiments reveal that female zebra finches respond preferen
tially to males that are displayed behind filters that allow transmittance of both UV
and visible light as opposed to those that only allowed in visible light (Bennett et al.
1996). Because birds make mate choices based on UV reflectance, it might not be sur
prising to find that they perceive and react to UV reflectance in insects (see Parrish
et al. 1984).
A-Aposematism: Bright colored insects, including certain dung beetles, are often
thought to be aposematic. Arrow (1951) recorded an instance where one of the African
ball-rolling beetles Gymnopleurus virens, which is bright green, blue, or crimson, was
shown to induce nausea in a captive baboon. Furthermore, this beetle is usually found
in association with 2 other similarly colored species, which are presumed to be Bate
sian mimics.
On Barro Colorado Island (BCI), two diurnal ball-rolling species, Canthon c. sallei
and C. moniliatus are also brightly colored and conspicuous, flying slowly at 15-30 cm
above the ground (Gill 1991). Canthon c. sallei produces a secretion that repels blow
flies from its food (Belles & Favila 1984), and these beetles captured in flight have an
unpleasant aroma. Canthon angustatus displays with pygidium raised when threat
ened. The secretion of the exposed gland has been shown to repel assassin bugs (Gill
1991). Staphylinids and assassin bugs are known to eat other dung beetles. Small,
metallic colored dung beetles like Ateuchus, and some Canthon, made up 74% of the
captures by a robber fly on BCI (Shelly, cited in Gill 1991). Bats are also known to oc
casionally eat dung beetles (Bellwood, pers. comm.). Burrowing owls are a persistent
predator of north American Phanaiines (Woodruff 1971), and other birds have been
seen to eat them (Sivinski, pers. comm.).
Aposematism in other iridescent beetles has been more convincingly demon
strated. Many cicindelids are iridescent, and a number of these have been shown to be
distasteful (Acorn 1988). One tiger beetle species appears to mimic an iridescent sym
patric blister beetle species. Others are thought to be Mullerian mimics of each other.
There is also a purported Mullerian complex of tiger beetles and mutillid wasps in Af
rica (Acorn 1988). Whether the iridescent colors of dung beetles are aimed at apose
matic deterrence of predation remains to be demonstrated.
B-Social signaling: If the brilliant colors of some dung beetles are used in signal
ing conspecifics, the conspecifics must be able to detect either the colors or some as
pect of them. The eyes of most Scarabaeidae are of the eucone type believed to make
possible the discrimination of colors, and of polarized light (Horridge 1975). Electro
physiological evidence for color sensitivity has been found in some Cetoniinae (Scar
abeidae) (Mazokhin-Porshnyakov 1964).
Most insects appear capable of seeing ultraviolet reflectance. Their visual system
often has one absorbance maximum around 350 nm (Silberglied 1979). Many species
of butterflies use UV for communication (Silberglied 1979). In a number of butterflies,

June, 1997

Behavioral Ecology Symposium '96: Vulinec

ultraviolet patterns and iridescence are not related to the visible wing patterns (Sil
berglied & Taylor 1973), whereas UV reflectance patterns in birds often parallel the
visible patterns (Bleiweiss 1994; Bennett et al. 1996).
To examine iridescent coloration in dung beetles, several species of Phanaeus and
related genera were photographed at various angles and under different lighting con
editions. Because insect perception is vastly different than ours, naturally, the infer
ences made about the photographs must be made with caution (Endler 1990).
Photographs were taken from the front to simulate a beetle's eye view of another, in
teracting, beetle, with 100 ISO Fuji daylight slide film and a Cokin ring flash with
color temperature 5600 K. Iridescent reflectance changed dramatically with the angle
and intensity of the light, as is typical with interference coloration. The iridescence on
the horn and clypeus disappears when not directly illuminated. The same light re
flected onto the subject completely changes the pattern on the prothorax. A front view
of Sulcophanaeus imperatorreveals iridescent spots on either side of the head that re
semble large red eyes. This is unlikely to be the region of the beetle typically encoun
tered by attacking predators. Photographing an iridescent beetle, Phanaeus
mexicanus, with daylight film under flash and UV lights yields even more psychedelic
color patterns; i.e., the insect fluoresces by absorbing UV light and reemitting it in the
visible spectrum. Finally, photographing phanaeine under UV light (Spectroline
model MB 100, peak wavelength 365 nm),with a Kodak UV 18A Wratten filter (passes
only wavelengths between 310 and 400 nm), and a Panasonic AG 150 videocamera
with a TV Zoom lens (6-54 mm; 1:1.4) (Eisner et al. 1988; Bleiweiss 1994; Van der
Kerkovan, pers. comm.), demonstrated UV reflectance from various iridescent areas of
the beetles, notably the front of the pronotum which forms an expansive shield (Fig.
la and b). The UV reflectance could be seen only at specific light angles, and small
changes of light source direction extinguished it. These dramatic and abrupt changes
in light reflectance due to angle (in both the visible and ultraviolet spectra) could be a
potentially efficacious method of communication, either between or within the sexes.
Beetle horns are thought to be used in combats between males for access to fe
males or over resources that attract females. However, male-male encounters are
rarely seen in the phanaeines (Halffter & Lopez 1977; Rassmussen pers, comm.; but
see Otronen 1988). Fighting requires energy and may lead to injury or at least the loss
of a mating opportunity (see Sivinski this symposium). As an alternative to fighting,
I suggest that males assess other males, particularly their size, by the appearance of
the horn, and that this presentation is enhanced by iridescence. While the horn itself
is less reflective, it is highlighted against the backdrop of a bright pronotal shield (Fig
Ib). The relationship between horn size and body size in Phanaeus spp. can be com-
plex and polymodel, however the two characters are generally positively and allomet
rically correlated (J. Sivinski unpublished data; see also Otte & Stayman 1979).
Allometry may be characteristic of structures designed to transmit visual signals con
cerning male body size (e.g., Green 1992; see Sivinski this symposium).
In some phanaeines, such as Diabroctus mimas, male horns are small but the pro
thorasic shield is massive (Edmonds 1972), and may provide a broad signaling sur
face. The various bosses, projections, horns, sculpturing and textures that occur on
phanaeines might be due to adaptations for signaling in different environments or
even result from selection for species isolation through different patterns of reflective
points. There is some intriguing evidence that color patterns in phanaeines may have
simple inheritance patterns similar to that in the butterfly Colias. The blue and green
morphs of Phanaeus difformis were bred in the laboratory with results consistent
with Mendelian ratios in the offspring (Blume & Aga 1976).
Horns and prothorasic shields could also be used in male -female signaling. Fe
males may choose males on the basis of many criteria (Arnold 1983). Hamilton and

Florida Entomologist 80(2)

Fig. la-A photograph of a male Phanaeus vindex taken from a video screen dis
playing the specimen video-taped under light produced by a tungsten bulb.
b-The same specimen video-taped with a camera fronted by an ultraviolet filter
and illuminated only by ultraviolet light. Note the strong UV reflectance of the pro
thorasic shield.

June, 1997

Behavioral Ecology Symposium '96: Vulinec

Zuk (1982) proposed that parasites can influence the evolution of sexually selected
traits. Individuals increase their net fitness by choosing mates with high genetic re
distance to parasites. This model assumes that heritable variation in fitness is main
trained in host-parasite coevolution. Hosts will select mates based on condition
dependent traits that indicate parasite loads. For example, parasites in brightly col
ored birds may cause dulling in the plumage, or a change in courtship displays.
Many birds signal with patches of feathers that reflect UV and/or are iridescent in
the visible spectrum (Radwan 1993). Male hummingbirds have iridescent patches on
head and neck that can be seen only from certain angles (Tyrrell & Tyrrell 1990). In
pigeons, iridescent feathers around the neck region have been hypothesized to inform
potential mates of the health of the bird (Hamilton & Zuk 1982). Birds with little or
no louse infestations are expected to show less iridescence. However, louse damage
doesn't affect the distal end of the feather, which is the part visible on an intact bird.
One study on mate choice in pigeons with artificially enhanced louse loads demon
strated that females are less likely to mate with males which have high louse loads
(Clayton 1990). However, to human observers there is no difference between the birds.
Unfortunately, these birds were not examined under ultraviolet light, and feather re
flectance patterns may be affected in those wavelengths. Clayton hypothesizes that fe
males see louse infestations during close encounters during courtship. Further,
Clayton presents an alternative to the Hamilton-Zuk "good genes" hypothesis that
mates are chosen for their genetic resistance to parasites. He argues that choosiness
may be explained more parsimoniously by a female's aversion to contracting lice her
self or passing them to offspring.
The bright triangular pronotum of Phanaeus vindex and related species can be oc
cupied and partially obscured by phoretic kleptoparasitic flies (Sphaeroceridae), such as
Norbommia frigipennis. The flies ride scarabs down into their subterranean chambers
and deposit eggs in the fecal food-masses and brood balls, where the fly larvae develop
(Sivinski 1983). The dung consumed by rapidly developing fly larvae may decrease the
fitness of the slower developing beetle larvae. Because Phanaeus forms pair bonds and
the pair cooperate in long periods of nest construction (Haftler & Edmonds 1982), it
would be advantageous for a female to determine, prior to mating, that a male carries
flies that might be deleterious to her offspring. Females may choose males that exhibit
traits which clearly demonstrate their freedom from kleptoparasites.
The plausibility of the male advertisement/female mate choice hypothesis is ef-
fected by the absence of obvious behaviors that suggest female comparison of mating
partners in phanaeine dung beetles (Arrow 1951; Otte & Stayman 1979). However, fe
males of most species are turtle-shaped, and difficult for a male to mount. Further
more, the female's genital opening is covered by a plate that would be difficult for a
male to pry open. Such a structure suggests the possibility of covert female choice at
the time of pair formation (Otronen 1988). Arrow (1951) suggests that female beetles
have inadequate vision to assess male horn size. Iridescent and ultraviolet reflective
surfaces, that change radically with small increments in angle of view, may serve as
signal enhancement for the visually impaired.
Since males participate in securing provisions for their offspring they may also
prefer mates without phoretic kleptoparasites. However, females appear to present
fewer opportunities for males to discern an infestation. While females are typically
the same color as males they are generally not horned and often have a less developed
prothorasic shield. Exceptions are the nearly sexually monomorphic Coprophanaeus
lancifer and ensifer (Edmonds 1972). These are extremely large insects that form
brood masses from carrion. Perhaps in keeping with their nocturnal habits they are
among the darkest colored of their tribe. Female horns are used in combats with other
females and males in competitions over cadavers (Otronen 1988).

Florida Entomologist 80(2)

The hypotheses that iridescent surfaces are due to sexual selection through mate
choice and intrasexual competition are not necessarily mutually exclusive. Neither
does the use of iridescent characters in sexual contexts preclude the hypothesis that
iridescent dung beetles may also be aposematic to diurnal predators or even cryptic in
certain habitats. Iridescence in insects may be influenced by numerous selective pres
sures. The least explored is the hypothesis of social signaling, and male -male compe
tition and perhaps intersexual assessment could be important in the evolution of
iridescence in dung beetles. If so, the bright flashing patterns of still other iridescent
insects may more often be territorial or sexual displays rather than aposematic or dis
ruptive predator defenses (Crane 1954).


I thank John Sivinski for his brainchild, and all-around helpfulness and generous
ity. Coleman Kane assisted at great length with the UV photography, and Corey Kane
provided discussions about the Tyndall blue effect. I especially thank Dave Mellow for
untiring support. University of Florida, Institute of Food & Agricultural Sciences,
journal series no.


ACORN, J. H. 1988. Mimetic tiger beetles and the puzzle of cicindelid coloration (Co
leoptera: Cicindelidae). Coleopt. Bull. 42: 28-33.
ARNOLD, S. J. 1987. Sexual selection: the interface of theory and empiricism. In: P.
Bateson, ed. Mate Choice. Cambridge Univ. Press, Cambridge, pp. 67-108.
ARROW, G. J. 1951. Horned beetles: A study of the fantastic in nature. W Junk, Pub
lishers, the Hague. 180 pp.
BELLES, X., AND M. E. FAVILA. 1984. Protection chimique du nid chez Canthon cyanel
lus cyanellus LeConte (Col. Scarabaeidae). Bull. Soc. Entomol. France 88: 602
BENNETT, A. T. D., I. C. CUTHILL, J. C. PARTRIDGE, AND E. H. MALER 1996. Ultraviolet
vision and mate choice in zebra finches. Nature 380: 433-435.
BLEIWEISS, R. 1994. Behavioral and evolutionary implications of ultraviolet reflec
tance by gorgets of sunangel hummingbirds. Anim. Behav. 48: 978-981.
BLUME, R. R., AND A. AGA. 1976. Phanaeus difformis Leconte (Coleoptera: Scara
baeidae): clarification of published descriptions, notes on biology, and distribu
tion in Texas. Coleop. Bull. 30: 199-205.
CLAYTON, D. H. 1990. Mate choice in experimentally parasitized rock doves: lousy
males lose. Amer. Zool. 30: 251-262.
CRANE, J. 1954. Spectral reflectance characteristics of butterflies (Lepidoptera) from
Trinidad, B.W.I. Zoologica (N.Y.) 39: 85115.
CROWSON, R. A. 1981. The biology of the Coleoptera. Academic Press, New York, N. Y.
802 pp.
DUNNING, D. C., AND K. D. ROEDER 1965. Moth sounds and insect-catching behavior
of bats. Science 147: 173-174.
EDMONDS, W. D. 1972. Comparative skeletal morphology and evolution of the Pha
naeine dung beetles (Coleoptera: Scarabaeidae). Univ. Kansas Sci. Bull. 49:
EDMONDS, W. D. 1994. Revision of Phanaeus Macleay, a New World genus of scara
baeine dung beetles (Coleoptera: Scarabaeidae, Scarabaeinae). Contrib. in Sci
ence 443: 1105.
EISNER, T., D. J. ANESHANSLEY, AND M. EISNER. 1988. Ultraviolet viewing with a color
television camera. Bioscience 38: 496-498.
ENDLER, J. A. 1990. On the measurement and classification of colour in studies of an
imal color patterns. Biol. J. Linn. Soc. 41: 315-352.

June, 1997

Behavioral Ecology Symposium '96: Vulinec

GHIRADELLA, H. 1985. The structure and development of iridescent Lepidopteran
scales: the Papilionidae as a showcase family. Ann. Entomol. Soc. Am. 78: 252
GILL, B. D. 1991. Dung beetles in tropical American forests. In: Hanski, I. and Cam
before, Y., eds. Dung Beetle Ecology. Princeton University Press, Princeton,
N.J., pp. 211-229.
GOLDSMITH, T. H. 1980. Hummingbirds see near ultraviolet light. Science 207: 786
GREEN, A. J. 1992. Positive allometry is likely with mate choice, competitive displays
and other functions. Anim. Behav. 43:170 172.
HALFFTER, G., AND W. D. EDMONDS. 1982. The Nesting Behavior of Dung Beetles
(Scarabadeinae): An Ecological and Evolutionary Approach. Institute de Ecolo
gia, Xalapa, Mexico.
HALFFTER, G., AND Y. LOPEZ G. 1977. Development of the ovary and mating behavior
in Phanaeus. Ann. Entomol. Soc. Amer. 70: 203-213.
HALFFTER, G., AND E. G. MATTHEWS. 1966. The natural history of dung beetles of the
subfamily Scarabaeinae (Col.: Scarabaeidae). Folia Entomol. Mexico 12-14: 1
HAMILTON, W. D., AND M. ZUK. 1982. Heritable true fitness and bright birds: a role for
parasites? Science 218: 384-387.
HINTON, H. E. 1973. Natural deception. In Gregory, R. L. and Gombrich, E. H., eds. Il
lusion in nature and art. Charles Scribner's Sons, N. Y., pp. 97-160.
HORRIDGE, G. A. 1975. Arthropod receptor optics. In: Snyder, A. W. and Menzel, R.,
eds. Photoreceptor Optics. Springer-Verlag, Berlin.
HUXLEY, J. 1975. The coloration of Papilio zalmoxis and P antimachus and the dis
cover of Tyndall blue in butterflies. Proc. R. Soc. London Ser. B 193: 441 453.
LALL, A. B., H. H. SELIGER, W. H. BIGGLEY, AND J. E. LLOYD. 1980. Ecology of colors
of firefly bioluminescence. Science 210: 560-562.
MAZOKHIN-PORSHNYAKOV, G. A. 1964. Methods and recent state of knowledge of co
lour vision of insects (in Russian). Ent. Obozr. 43: 503-523.
NIJOUT, H. F. 1991. The development and evolution of butterfly wing patterns. Smith
sonian Inst. Press, Washington, D. C.
OTRONEN, M. 1988. Intra and intersexual interactions at breeding burrows in the
horned beetle, Coprophanaeus ensifer Anim. Behav. 36: 741-748.
OTTE, D., AND K. STAYMAN. 1979. Beetle horns: some patterns in functional morphol
ogy In: Blum, M. S. and Blum, N. A., eds. Sexual selection and reproductive
competition in insects. Academic Press, N. Y pp. 259-292.
PARRISH, J. W., J. A. PTACEK, AND K. A. WILL. 1984. The detection of near-ultraviolet
light by nonmigratory and migratory birds. The Auk 101: 53-58.
POPE, R. D., AND H. E. HINTON. 1977. A preliminary survey of ultraviolet reflectance
in beetles. Biol. J. Linn. Soc. 9: 331-348.
RADWAN, J. 1993. Are dull birds still dull in UV? Acta Ornithol. 27: 125-129.
SILBERGLIED, R. E. 1979. Communication in the ultraviolet. Ann. Rev. Ecol. Syst. 10:
SILBERGLIED, R. E., AND O. R. TAYLOR 1973. Ultraviolet differences between the sul
pher butterflies, Colias eurytheme and C. philodice, and a possible isolating
mechanism. Nature 241: 406-408.
SIVINSKI, J. M. 1983. The natural history of a phoretic Sphaerocerid Diptera fauna.
Ecol. Entomol. 8:419-426.
SIMON, H. 1971. The Splendor of Iridescence: Structural Colors in the Animal World.
Dodd, Mead and Co., N. Y 268 pp.
TYRRELL, E. Q., AND R. A. TYRRELL. 1990. Humminbirds of the Caribbean. Crown
Pub. Inc., N. Y 238 pp.
WOODRUFF, R. E. 1973. The scarab beetles of Florida (Vol. 8, Arthropods of Florida).
Florida Dept. of Agric. and Con. Services.
YOUNG, O. P. 1984. Perching of Neotropical dung beetles on leaf surfaces: an example
of behavioral thermoregulation? Biotropica 16: 324-327.

Florida Entomologist 80(2)


USDA, ARS, Center for Medical, Agricultural and Veterinary Entomology
Gainesville, FL 32604


Occasionally, flies bear sexually dimorphic structures (ornaments) that are used,
or are presumed to be used, in courtships or in aggressive interactions with sexual ri
vals. These are reviewed, beginning with projections from the head, continuing
through elaborations of the legs and finishing with gigantism of the genitalia. Several
functions for ornaments are considered, including advertisement of genetic proper
ties, subversion of female mate choice and "runaway" sexual selection. Neither the
type of ornament nor the degree of elaboration necessarily indicates which of the
above processes is responsible for a particular ornament. Resource distribution and
the resulting possibilities for resource defense and mate choice explain the occurrence
of ornaments in some species. The phyletic distribution of ornaments may reflect for
aging behaviors and the type of substrates upon which courtships occur.

Key Words: sexual selection, territoriality, female mate choice, arms races


Ocasionalmente, las moscas presentan estructuras sexuales dimdrficas (ornamen
tos) que son utilizados o se cree sean utilizadas en el cortejo sexual o en interacciones
agresivas con sus rivals sexuales. Dichas estructuras han sido evaluadas, comen
zando con proyecciones de la cabeza, continuando con las estructuras elaboradas de
las extremidades y terminando con el gigantismo de los genitales. Se han considerado
distintas funciones para dichos ornamentos, incluyendo la promoci6n de sus propie
dades geneticas, subversion de la elecci6n de la hembra por aparearse, y el rehusare
a la selecci6n sexual. Tanto el tipo de ornamento como el grado de elaboraci6n no ne
cesariamente indicaron cual de los process mencionados es el responsible de un or
namento en particular. La distribuci6n de los recursos y la posibilidad resultante de
un recurso de defense y de elecci6n de apareamiento pudieran explicar la aparici6n de
ornamentos en algunas species. La distribuci6n filial evolutiva de los ornamentos
pueden reflejar comportamientos relacionados con la busqueda del alimento y con el
tipo de sustratos sobre los cuales el cortejo sexual se lleva cabo.

In general, the body shapes of flies fall into a few familiar categories, ranging from
the willowy (e.g., Tipulidae) to the robust (e.g., Muscidae). Sporadically added onto
these ordinary forms are extraordinary elaborations apparently fashioned by sexual
selection. These have been called "ornaments," but it is useful to think of them as "or
gans of propaganda," designed to communicate with, and manipulate, potential mates
and/or sexual rivals (c.f, Krebs & Dawkins 1978). In considering the ornaments of
Diptera, first I survey their types and locations, starting with the head and working
back to the genitalia. Then I will address whether the nature of ornaments provides
clues to their "messages" and for whom the messages are intended. Finally, I attempt
to correlate certain forms of decoration with different types of mating systems in var
ious taxa of flies.

June, 1997

Behavioral Ecology Symposium '96: Sivinski


A. Eyes

Sexual dimorphism of the eyes is commonplace in the Diptera, but ornamented
eyes are rare. In order to make this distinction clear, the term "ornament" needs to be
clarified. Males flies, particularly those that swarm, often have larger eyes with por
tions modified to locate the motions of incoming females (e.g., Sivinski & Petersson
1996). However, this sexual difference does not constitute ornamentation. For one
thing, these dimorphic eyes are not suspected of being signaling devices. Colors and
patterns, common in eyes in families such as Tabanidae, Dolichopodidae and Tephriti
dae, and which could act as signals, will not be considered ornaments either. Rather,
ornaments will be defined, perhaps somewhat arbitrarily, as elaborated or novel
structures, sculptures rather than paintings. An example of ornate eyes are those of
the male Brazilian drosophilid Zygotricha dispar Wiedemann (Fig. Ib). They are
much enlarged, and prolonged into sharpened horns that resemble those of a water
buffalo (Bristowe 1925). In certain congeners, the tip of the eye curls like a ram's horn
(Grimaldi 1987; Grimaldi & Fenster 1989).

B. Extensions of the Head Capsule (Stalk-eyes and Antlers)

In eight acalypterate families, male's heads, and occasionally female's heads, are
sometimes stretched laterally until the eyes are supported at the ends of remarkable
"stalks" (Fig. la; Wilkinson & Dodson 1996). There is a considerable literature regard
ing the behavior of stalk-eyed Diopsidae that will be addressed when the significance
of ornaments is discussed (e.g., Burkhardt & de la Motte 1983; de la Motte &
Burkhardt 1983; Shillito 1960, 1976; Wilkinson 1993; Wilkinson & Dodson 1996).
Antlers, projections from the head capsule, occur, to one extent or another, in five
families of flies (Wilkinson & Dodson 1996). Those of the tephritid genus Phytalmia
originate under the eyes and are by far the most elaborate (Fig. Ic; see McAlpine &
Schneider 1978; Schneider 1993). In his classic "The Malay Archipelago", Wallace
(1869) describes his collection of four species from New Guinea: ". .. these horns (of P
cervicornis Gerstaecker) are nearly as long as the body, having two branches, with
small snags near their bifurcation, so as to resemble the horns of a stag. They are
black, with the tips pale ... the eyes (when alive) are violet and green ... The horns
(of P megalotis Gerstaecker (= .. ..i i are about one third the length of the insect,
broad, flat, and of an elongated triangular form. They are of a beautiful pink color,
edged with black, and with a pale central stripe. The front of the head is also pink, and
the eyes violet pink, with a green stripe across them, giving the insect a very elegant
and singular appearance .... The horns (of P alcicornis (Saunders)) are very remark
able, being suddenly dilated into a flat plate, strongly toothed round the outer margin,
and resembling the horns of an elk (moose) . the head (of P brevicornis (Saunders))
is compressed and dilated laterally, with very small, flat horns .."

C. Mouthparts and Face

Mouthparts are occasionally ornamented in the Dolichopodidae. Males of the tiny
Chrysotus pallipes Loew have much enlarged labial palps (see Van Duzee 1924),
which emit silver flashes as males signal from the surface of leaves (Sivinski 1988a).
The expanded gold-silver palpi of the Hawaiian C. pallidipalpus Van Duzee reflect
light as males pursue females (Parmenter 1952). The palpi of males in the closely re

Florida Entomologist 80(2)



Fig. 1. Projections from the heads of acalypterate flies:
a) Stalk-eyes on a male Achias sp. (dorsal view), a large platystomatid fly from
New Guinea. Similar projections in diopsid flies are perceived as signals by both
males and females in the contexts of aggression and mate choice respectively.
b) The head (frontal view) of a male drosophilid, Zygotricha dispar, a tiny, but pug
nacious, fly from Brazil that uses its horn-like eyes in intrasexual combats and per
haps as an advertisement of size directed to potential mates and rivals.
c) The antlered head (lateral view) of a male Phytalmia cervicornis, a large and ag
gressive tephritid fly from the rain forests of New Guinea where males defend ovipo
sition sites from other males and mate with females that come to lay eggs.

lated genus Asyndetus are also sometimes ornate. Those of A. flavipes Van Duzee are
bright yellow and covered with long yellow hairs (Van Duzee 1932). A male of Aph
rosylus raptorWalker, searching for mates on seaweed covered rocks, flashes his large
silver palpi "as he swings his shoulders and head in his stride" (Parmenter 1952). Sil
ver reflections are found on the elongated faces of certain male dolichopodids. In Poly

June, 1997

Behavioral Ecology Symposium '96: Sivinski

medon spp. the face extends to form a "plate or ribbon" that hangs down over the
proboscis (Van Duzee 1927).

D. Setae

Male tephritids often have highly modified setae. Some species of Ceratitis, te
phritids that include the infamous Mediterranean fruit fly, C. capitata (Wiedemann),
bear orbital setae on the face above the antennae. These hairs can be strikingly long;
those of C caetrata Munro reach more than twice the width of the head in length (Mu
nro 1949). The setae, tipped with either black or white expansions (Bezzi 1924), are
waved" about during courtships (e.g., Arita & Kaneshiro 1989).

E. Antennae

Many flies, such as mosquitoes and chironomid midges, bear sexually dimorphic
antennae (see Sivinski & Petersson 1996). In most cases, these differences result from
one sex, usually the male, being adapted to perceive pheromones or acoustic cues.
However, some antennae appear to be modified to emit a signal of their own. Chlo
ropids are rarely dimorphic, but males of the sole species of Gampsocera in Hawaii
have various unique markings and thickened and black aristae (Kanmiya 1989).
Males of Camposella insignata Cole, an acrocerid from Ecuador, have "an astonishing
development" of the third antennal segment that renders it enlarged, flattened and
patterned (Cole 1969). Dolichopodid males sometime have elongated antennae which
are plumed at the tip (e.g., Tachytrechus spp. (Greene 1922)), or in the case of T bin-
odatus Loew, plumed at the tip and in the middle. Tachinids commonly have sexually
dimorphic antennae. Some, such as those of male Lispidae triangularis Aldrich which
contain a much broadened third segment, seem decoratively large (Aldrich 1929). Ex
aggerated and plumed antennae occur in some tephritids (White 1988).


A. Forelegs

Various dolichopodids wave and/or touch potential mates with ornamented fore
legs (Gruhl 1924; Fig. 2a). Males of Neurigonia quadrifasciata Fab. and Poeciloboth
rus nobilitatus (L.) approach a female from the rear and reaching over her, curve their
plumed tarsi over her head (Smith 1959). They then wave their tarsi alternately, one
over each eye. Male Dolichopus omnivoraxVan Duzee wait for foraging females on
floating vegetation (Steyskal 1938). When a potential mate is found, he approaches
with his forelegs extended laterally. The tibiae hang down and forward, displaying a
large black pad on the terminal tarsi. If the female remains still, the male's advance
will bring the pads almost into contact with her eyes. Sometimes the front femora of
dolichopodids are decorated. Those of Tachytrechus olympiae Aldrich are swollen and
marked with a dark spot (Greene 1922). The pinnacle of foreleg ornamentation in the
Dolichopodidae is occupied by Campsicnemus magius (Loew), whose limbs are so
swollen, pendanted, hairy and bizarre that the dipterist Gerstaecker accused his col
league Loew of describing a species from a specimen deformed by fungus (Verrall
1905; Lundbeck 1912; Fig. 2b). Some male asilids in the genera Heteropogon and
Cryptopogon bear decorated front tarsi (Bromley 1933; Wilcox & Martin 1936). Curi
ously, only American species of the latter genus, and not those from Europe, have tar
sal elaborations (Hull 1962). In addition to waving their ornaments, robber fly males

Florida Entomologist 80(2)

June, 1997

Behavioral Ecology Symposium '96: Sivinski

may stroke the female's head and thorax. It is not uncommon for male syrphids to
have dilated front legs, spotted with clumps of setae (e.g.,Verrall 1901). This tendency
achieves the fantastic in the complex decorations of the west African species Tityusia
regulas Hull (Hull 1937). The fore tibia are "enormously thickened, grooved, twisted
and distorted" with an "extremely long, extremely matted" dark pile of fringe. The fore
tarsi are "extravagantly flattened ... the lateral edges of the second, third and fourth
segments prolonged into narrow, down curving lobes." Among acalypterates, the yel
low front legs of the tephritid Ectopomyia baculigera bear a large down-pointing pro
jection on the femur, while the front basitarsis of the male Euphranta maculifemuris
broadened and concave (Hardy 1973).

B. Midlegs

A mosquito, Sabethes cyaneus (F.), bears elongated, iridescent blue and gold scales
that transform the midlegs into "paddles" (Hancock et al. 1990; smaller setae occur on
the other legs as well; S. tarspus Dyar & Knab and some other congeners also bears
leg paddles; Fig. 3; smaller setal expansions occur on the legs of certain Wyeomyia
spp.). Males fly toward resting females with their ornate legs held perpendicular to
their bodies. After landing on twigs, they suspend themselves by their forelegs, then
swing and wave their paddles. Undulating waving motions persist after the initial
coupling, until the genitalia are fully clasped. "Waggling," during which the midlegs
rise and fall, continues throughout the copulation (see Eberhard 1994 for a discussion
of courtship during mating). Remarkable middle tarsi occur in males of the empidid
Rhamphomyia scaurissima Wheeler (Wheeler 1896). The first joint consists of a glob
ular base beset with prominent hairs and a scale-like appendage, the second is large
and symmetrical and has a club-shaped extension clothed in a pencil of long hairs,
and the third is enormously enlarged into a boat-shaped structure. A few tephritids of
the genus Ceratitishave either mid and/or hindlegs expanded and feathered along the
margins (Silvestri 1914). Male dolichopodids sometimes employ ornate midtarsi in
courtship displays (e.g., Qvick 1984). Those of Sympycnus cuprinus are dilated and
fringed with black bristles (Cole 1969; see also Harmston & Knowlton 1943). The mid
legs of certain species of Campsincnemus are much more elaborate (e.g., Curran 1933;
Harmston & Knowlton 1942). Robber flies of the genus Cryptopogon often bear tufts
of black or silver hairs on the tarsi of both the front and middle legs (Wilcox & Martin
1936). In general, ornaments upon the midlegs of flies appear to be rare relative to
forelegs (Wheeler 1896).

C. Hindlegs

Some of the most amazing ornaments in the Diptera adorn males of the platypezid
genus Calotarsa (Fig. 2c). Three species are found in widely separated North Ameri
can locations. Their enlarged hindlegs bear a variety of curious projections and glit
tering aluminum-colored flags (Kessel 1963). Snow (1884) noted how swarming males
.. allow their hindfeet to hang heavily downward and look as if they were carrying
some heavy burden." There is a degree of convergence between the design of the pos
terior tarsi in Calotarsa and the fore tarsi of the syrphid T regulus (Hull 1937; see sec
tion on front legs), but the hover fly has a peculiarity upon its hind tarsi as well, "an
enormous brush of dark, matted hair." Conspicuous hairs decorate the hind tarsi of
certain asilids (Wilcox & Martin 1936). The entire hindleg of males in the genus La
godias is fringed in long flattened setae (Hull 1962). Male anthomyids sometimes
have patterned legs with elongated setae. The hind tibia of Rhynchtrichops aculeipes

Florida Entomologist 80(2)

June, 1997

Behavioral Ecology Symposium '96: Sivinski

Zett. has an odd projection that renders it reminiscent of a wishbone (Seguy 1923).
Males of the dolichopodid genus Scellus are remarkable not only for their caudal rib
bon-like projections (see below), but for the enlarged corkscrew-like spines and long
hairs that project from the hindlegs (e.g., Greene 1924). If these are ornaments, and
not a grooming apparatus for the abdominal projections (or something else), their be
ing on the hindlegs is noteworthy It is my impression that dolichopodid hindlegs bear
fewer peculiar modifications than the midlegs, which in turn are less often orna
mented than the front (e.g., Van Duzee & Curran 1934). Perhaps the presence of a cau
dal appendage creates a posterior focus of attention in females, into which the
hindlegs can be profitably included. Female empidids of the genus Rhamphomyia
have large scale-like setae on their legs. These are held away from the body while in
flight and glitter in the light (Evans 1988).

D. Wings

Like the antennae, wings are commonly sexually dimorphic in size, although this
is often because of adaptations to different flight requirements (e.g., Sivinski & Dod
son 1992). Wings are sometimes dimorphically marked, or have sexually distinct ve
nation (e.g., Alexander 1936; Kanmiya 1989), and serve important roles in courtships
and aggressive interactions (e.g., Land 1993; Lunau 1992), but, for present purposes,
these are not considered to be ornamented. Possible exceptions occur among the oddly
shaped, rounded and patterned wings of certain female empidids who participate in
sex-role reversed swarms (see Cumming 1994) and the combined peculiar wings and
modified tarsi of the dolichopodid Collinellula magistri Aldrich (Aldrich 1932).


A. Enlargement of the Abdomen

Females of the empidid Rhamphomyia longicaudata Loew inflate their abdomens
with air until the pleural membranes are greatly stretched and collapse when punc
tured (Steyskal 1941; Newkirk 1970). Similarly, the membrane of the third abdominal
segment in females of the New Zealand species Hilara flavinceris Miller forms an ex
tensible bladder that stands out to the sides (Miller 1923). Cumming's (1994) examine
tion of the extensive holdings of Empididae in the Canadian National Collection of
Insects and Arachnids (Ottawa) revealed that 29% of the described species ofRhamph
ymyia and 26% of Empis (583 species total) had females with pinnate scales on the legs
or abdomen and pleural sacs. Male abdomens may sometimes be modified as well; that
of the swarming Ugandan stratiomyid Platyna hastata F is expanded and flattened,
and "... brilliantly reflects a white light ... The glistening appearance of the upper sur
face ... is very striking" (Carpenter 1923). Unfortunately, no females were observed, or
at the time had ever been collected, and a sexual dimorphism is only presumed.

B. Modified Glandular Projections

Females of the chironomid Palpomyia brachalis evert long glandular strings from
their abdomens as they participate in sex-role reversed female swarms (Edwards
1920). These have been interpreted as pheromone organs, but their bright orange
color contrasting with the black body suggests a visual role as well. Since similar
tubes in other species of Palpomyia and the related genus Bezia are colorless, their
great size may not be ornamental but a means of increasing surface area for phero
mone dispersal.

Florida Entomologist 80(2)

C. Caudal Ribbons

Males of the dolichopodid genus Scellus have odd, twisted, ribbon-like structures
projecting from the dorsum of the abdomen (Green 1924). Some are as long as the ab
domen itself, fringed and tufted with hairs, or tipped with a spoon-like enlargement.
Often white in color, with black bases and yellow ends, their function is mysterious.
These strange appendages may have evolved solely for communication, or perhaps
they are ornate elaborations of structures that serve an additional purpose (phero
mone dispersion?). In addition to long, twisted, reddish or orange-yellow ribbons,
male S. virago Aldrich have enlarged fore tibia furnished with a large blunt protuber
ance and tufts of curly hairs on the middle tibia. Despite these multiple male orna
ments, the female appears to be more sexually aggressive (Doane 1907); ". . she
seemed suddenly to become very much excited, now squatting low, now rising high
and waving the wings frantically. The cause of this extra excitement was a male fly .
S. He seemed to paying but little attention to her.... (After) facing each other, going
through the curious performance ... The male then turned away and seemed about
to leave, but the female quickly flew in front of him again and began her antics."

D. Modified setae

Males of the large ropalomerid Scatophga gigantea Aldrich have "very striking
long, dense . ." hair on their abdomens (Aldrich 1932). Tephritid fruit flies sometimes
bear modified setae on the abdomen; e.g., males of Trupanea brunnipennis have a
mass of strong yellowish bristles along the posterior margin of the 5th tergite (Hardy
1973). Copiolepis quadrisquamosa Enderlien is perhaps the most dramatically
plumed tephritid (Enderlein 1920). It somewhat resembles the Birds of Paradise with
which it shares habitats in New Britain and New Guinea.

E. Genitalia

It has been argued that the notorious complexity of some male insect genitalia, in
cluding those of certain Diptera, is in fact ornamentation, but ornamentation on a tac
tile level (Eberhard 1985). Giant male genital regions in dolichopodids are employed
in courtships prior to physical contact. A number of species carry enlarged terminalia
(hypopygium) slung under the abdomen. In Dolichopus omnivagus this is raised and
lowered during the male's courtship advance (Steyskal 1938). I observed a more dra
matic effort by an unidentified male on the upper surface of a leaf. It raised itself up
on its long legs, beat its wings and then lowered the hypopygium until it hung perpen
dicular to the body. At this point the genitalia began to slowly twirl. As in some other
structures discussed previously, it is not clear whether the terminal segments are en
large to send a message or if the great size serves a mechanical function and is sec
ondarily used in courtships.


A. Size and Aggression in Horn-eyed, Stalk-eyed and Antlered Flies

The evolution of horn-eyes, stalk-eyes and antlers illustrates how organs of com
munication and manipulation might arise through aggression among members of the
same sex. McAlpine (1979) offers a diabolical hypothesis of how a blunt instrument
(the head) could evolve through deceit into a sophisticated piece of propaganda. Male
flies often fight head to head. The broad head and abundant cheek bristles of the Aus

June, 1997

Behavioral Ecology Symposium '96: Sivinski

tralian platystomadid Pogonortalis doclea (Walker) are used in such combats
(McAlpine 1975). The enlarged and hairy surface area better applies force and pre
vents slippage. Bristles may even become interlocked to grip an opponent, a technique
that may have been further perfected by the clusiid Clusoides gladiator McAlpine,
whose males' facial vibrissae are spiraled (McAlpine 1976), perhaps to twist into those
of a rival's. These elaborations serve as practical weapons, but are they organs of com-
munication; i.e., are they ornaments? Perhaps not, but proceed one step further. Sup
pose, as is often the case, a smaller fly retreats from a confrontation after determining
that his opponent is too large to successfully engage. If the size of the rival is assessed
by the breadth of his head, as gauged by the degree of overlap between the two sets
of eyes, then males can appear large and conquer psychologically by simply widening
the head. As deceitfully widened heads become common, even further exaggeration is
required to sustain a bluff and the resulting "arms race" pulls eyes farther and farther
out until they are held at the ends of extraordinary stalks, each of which may be
longer than the body (e.g., an 8 mm long male of an undescribed diopsid from Borneo
supported eyestalks with a combined span of 20 mm; Burkhardt et al. 1994).
In the end though, there are practical conclusions to arms races. Accumulating ex
penses and increasing vulnerability may dictate the final state of an ornament. Per
haps truly extraordinary ornaments, such as stalked-eyes in certain Achias spp.
(McAlpine 1994), are cases where selection has exploited every opportunity and no
further mechanical demands can be made on the overall "fly design." Wilkinson & Dod
son (1996) found the relationship between antler size and body size withinPhytalmia
spp. reached a plateau. At this point signals are no longer deceptive, they are genuine
burdens that reflect the qualities of their bearers. Wilkinson & Dodson (1996) suggest
that since there is a strong positive allometric correlation between body size and pro
sections from the head, "(ornament) size is an honest indicator of overall size, which it
self is a predictor of fighting success .. (ornament) size could be used by males to
assess an opponents fighting ability, thereby avoiding unnecessary contests." One
might ask why body size should be advertised by an ornament that does not increase
in size at the same rate as the actual body; i.e., why do larger males have proportion
ately longer projections? Positive allometry might allow more accurate judgements of
size; i.e., since a small increase in body size results in a larger and more obvious in
crease in the ornament, "the projection span scale will be finer than the body length
scale." Allometry might also suggest that the cost of stretching the head, in terms of
energy and maintenance, does not increase at the same rate as that of enlarging legs
and guts and the other sophisticated and enervated body parts that make up "size." If
so, larger flies might spend a similar proportion of their resources to advertise their
bulk as smaller individuals but obtain a relatively greater return on their advertising
budget. Still another hypothesis for the existence of allometry is that larger individu
als may be more likely to use force in their interactions with other males. As a conse
quence they might invest more in weapons and propaganda (see Green 1992).
Females in some diopsid species are found in groups associated with individual
males. However, these harems in Cyrtodiopsis white are not the result of males ex
cluding rivals, but of a female preference for males with long stalks (Burkhardt & de
la Motte 1988). Allozyme markers have revealed that males with longer stalks sire
relatively more offspring (Burkhardt et al. 1994). In C. dalmanni, females likewise
prefer longer stalked males (Wilkinson & Reillo 1994). What may have originally been
propaganda to intimidate rival males has come under scrutiny from females and is
now used as a factor in mate choice.
Like eye-stalks, antlers are both weapons and symbols of prowess. Males of
Phytalmia mouldsi clash by rising up on their legs and pushing hard against each
other's remarkable heads, although the antlers themselves do not play a major role in

Florida Entomologist 80(2)

the battle (Moulds 1978). However, those whose horns are experimentally lengthened
or shortened are respectively more and less likely to win fights (Dodson 1989). In ad
edition, males with their horns removed are treated by their rivals like females
(Wilkinson & Dodson 1996). Hence antlers serve, at least in part, as signaling organs.
The massive antlers of P alicornis are more involved with actual pushing.

B. Material Resources and Deception in the Empididae

Horns and stalks have been depicted as evolving through interactions among
males (intrasexual selection), although females might come to prefer a particular
state of ornamentation and influence its form. The ornaments considered from this
point forward are presumed to have originated in a different context, that of interac
tions between the sexes, i.e., intersexual selection. They are employed, or are believed
to be employed, in courtships or in attracting the opposite sex.
A number of male empidids present mates with insects they have killed or stolen
from spider webs (e.g., Chvala 1976). Often these are the only animal meals females
will have as adults. Female mate choice is sometimes based on this nuptial gift and in
certain cases the importance of the gift is so great that a sex-role reversal takes place.
Females swarm and choosey males examine a series of potential mates before feeding
and inseminating a particular individual (Svensson et al. 1989). The addition of a re
source to courtship has consequences for ornamentation. Both sexes have "goods," the
nuptial gift of the male and the eggs of the female, that can be advertised to a poten
tial "customer."
Male Rhamphomyia scaurissima have peculiar growths protruding from the mid
legs (Fig. 4a). I have found no behavioral records for R. scaurissima, but other species
in the genus form swarms. Congeners provide females with a nuptial gift of a small
dead insect which they hold in their legs (Downes 1970; Fig. 4b). Only males with a
gift succeed in mating. Could this mass of swellings and projections deceitfully sug
gest a resource the insect doesn't have or exaggerate the size of one that it does?
On the other side of sexual bartering are females whose apparent fecundity might
influence whether or not they obtain a valuable meal. Females of many Rham
phomyia, Empis and Hilaria species inflate their abdomens while participating in
sex-role reversed swarms (Cumming 1994). It is tempting to think that such swellings
may be exaggerated promises of fecundity directed toward males who provide a nup
tial gift. Larger females are preferred by resource-providing males in other empidids
(e.g., Svensson et al. 1989). Like stalk-eyes, abdominal enlargements may evolve into
"honest advertisements" if only the largest females can fly with the most swollen ab
domens. In Rhamphomyia species females bear glittering setae on their legs. When
extended in flight these ornaments may call attention to the females' abdomens, as
might the coloration of another empidid, an unidentified Alaskan species "garishly
marked with an extensive silvery abdominal 'saddle' which flashes conspicuously as
she crosses beams of sunlight."(Frohne 1959).

C. Good Genes, Manipulation and Runaway Selection

Some ornaments suggest original functions; the air-filled abdomens of female em
pidids may have been false advertisements of fecundity, just as stalk eyes exaggerated
size and dangles from midlegs gave the impression that a male empidid has a nuptial
gift. But putting these instances with perhaps more obvious histories aside, a number
of very puzzling objects remain. Just why does stroking a female's head with tarsal
plumes improve the reproductive success of a male robber fly? If simple species isola

June, 1997

Behavioral Ecology Symposium '96: Sivinski



Fig. 4. A comparison of the appearance of two species of Rhamphomyia:
a) The middle legs of males of the empidid R. scaurissima end in a remarkable
complex of swellings and projections (from Wheeler 1896).
b) These peculiarities are absent from the legs of R. ursinella. However, the orna
ments of R. scaurissima might bear a resemblance to the more mundane species car
trying a nuptial gift, such as the chironomid Smittia sp. (smaller insect figured below;
from Downes 1970). Perhaps originally, ornamented males appeared to be holding a
gift and so were allowed to copulate with females who would otherwise have mated
only when provided with a prey item.

Florida Entomologist 80(2)

tion is involved in ornamentation, why are such decorations relatively uncommon?
Are ornamented species in some particular danger of engaging in unprofitable hybrid
izations? The opposite is often the case (e.g., West-Eberhard 1984). The spectacular
genus Calotarsa, for example, consists of three widely separated North American spe
cies, one so rare it appears to have never been recollected.
There are a number of other paths that might lead to ornamentation, any one of
which could result in a world with only a single species being inhabited by orna
mented animals.
1) The production and use of expensive and unwieldy growths may provide a po
tential mate (or sexual rival) with an estimate of genetic (or phenotypic) quality; i.e.,
the displayer has foraged well enough or avoided debilitating infections long enough
or is big enough to put on his show (e.g., Sivinski 1988b). Body symmetry is a correlate
of genetic quality and a trait preferred by choosing females in some animals (Moller
1992; Thornhill 1992; Watson & Thornhill 1994). The flags and feathers of some dis
plays could test the genome's ability to produce symmetry.
2) The receiver may be manipulated by an ornament. Nervous systems are imper
fect. A flaw in perception or information processing can be exploited by the behavior
of others (cf. Dawkins 1982). For instance, a resting dragonfly can be "hypnotized" by
tracing a narrowing spiral in the air. Such an event is presumably so rare that selec
tion has not favored a brain resistant to the influence of a moving finger. Perhaps
flaws in female nervous systems allow them to be approached and handled by rhyth
mically waving, plumed, or otherwise ornamented, males.
3) A female preference for extreme examples of a certain characteristic in a mate
begins an episode of "runaway sexual selection." That is, when females prefer the most
ornate male available, genes for both choosing the very elaborate (expressed in daugh
ters, but present in both daughter and sons) and being very elaborate (expressed in
sons, but present in both daughters and sons) can generate a sort of "chain reaction"
self selection for the increasingly extreme. A lucid explanation of this complex proce
dure can be found in Dawkins (1986). This form of selection requires that females sam
ple the range of male decoration and mate with the most ornate. It has been suggested
that such mate comparisons are not typical of insects, who are presumed to have a lim
ited time to acquire courtship experiences and little capacity to remember those that
they had (Alexander et al. 1997). If so, perhaps only rare circumstances, where poten
tial mates are compared simultaneously or where females have unusually good mem
ories, give rise to the occasional "runaway monstrosity" (Sivinski & Petersson 1997).
Could these various kinds of "messages" be recognized by the nature of the orna
ment that carries them? This categorization may prove to be difficult. I can imagine
many ornaments of the "puzzling" variety (those not originally exaggerating size or a
resource) resulting from any of the above. The male robberfly rhythmically stroking
the female's head with leg plumes could be displaying his coordination, seducing her
"hypnotically," or satisfying her taste for an extreme in courtship.
Though similar types of ornaments could be derived from different types of selec
tion, might the different types of selection generate different degrees ofornamenta
tion? To the entomologist's eye not all ornaments are equally elaborate. Some
dolichopodid legs seem to be practical semaphores, others appear contorted and ab
surd (Fig. 2a &b). Would advertisers of genetic quality tend to invest as much in their
displays as participants in a "runaway" situation, or vice versa? Unfortunately, this to
might be a difficult approach to finding meaning. Each type of selection could direct
varying amounts of resources to ornaments, so that complexity and simplicity may
not be indicative of particular sets of selection pressures. For example:
1) There are several explanations for variance in ornaments evolved to advertize
genetic quality." A simple ornament may sometimes be sufficient; i.e., there might be

June, 1997

Behavioral Ecology Symposium '96: Sivinski

types of messages that are just not improved by increased broadcasting. Genetic iden
tity (species identification or lineage identification) is one possibility Under some cir
cumstances, mate choice based on symmetry might select for simplicity If complexity
can overwhelm perception and hide asymmetry, females may come to prefer simpler
ornaments, clearly displayed.
However, there may be few such inherent limitations on how elaborate ornaments
that reflect genetic quality can become. If an ornament is "improved" from the sig
naler's perspective by exaggeration, then potential mates or sexual rivals with new
and higher criteria for what they find attractive or intimidating will be better adapted
than "gullible" individuals with out-of date tastes, and so on and so on (see discussion
of stalk-eyes). An alternative to linked escalation of ornamentation and discrimina
tion is selection for a new ornament that will, at least t. ,i .,1, I, be a more honest
indicator of genetic quality (see also Iwasa & Pomiankowski 1994). Multiple male or
naments are commonly found in the Dolichopodidae (e.g., the genus Scellus; see
It is unlikely that all ornaments are equally burdensome or that all bearers of or
naments would have similar resources to spend on advertisement. Different limits
would lead to variety in ornamentation. On the other hand, some signal systems may
be relatively simple because they have not been in existence long enough for arms
races to bring them to the brink of being maladaptive handicaps to their carriers.
2) Males may exploit weaknesses in female nervous systems, but females might
evolve "immunity", and this could ultimately lead to interspecific differences in the
elaborateness of male ornaments. If the subversion of females' ability to choose a mate
has a sufficiently negative effect on their reproductive success, then flaws in their
brains might be eventually corrected and the degeneration of their sexual control
stopped. Males might then respond with more potent stimuli, escalating yet another
arms race. Assuming different female susceptibilities and different costs to being ma
nipulated, a range of ornamentation could develop in various males.
3) Where runaway sexual selection occurs (if it occurs) the ability of the receiver to
discriminate differences in signals would influence the capacity to choose among
mates, and eventually how far "taste" can dictate male ornamentation. The abilities
of different males to bear the burdens of their "beauty" could also determine how elab
orate any particular display may become. What is extreme in an aerial predator might
appear simple in a fruit fly. Parenthetically, the male empidids who carry objects as di
verse as flower petals (Hamm 1913) and silk balloons, (Kessel 1955; which sometimes,
but not always, contain a prey item), into mating swarms may be using a disposable
ornament" that would not interfere with the other parts of their lives.
Another characteristic of an ornament that might help translate its meaning is the
variance in the display among the individuals of a population. It has been suggested
that when females choose a male trait in lekking species, "modifier genes" to generate
variance in that trait might be selected as well (Pomiankowski & Moller 1995). The
explanation is that the combination of the highest mean value of a character along
with its greatest variance will produce the most extreme manifestations of that trait
in the next generation. In both "runaway selections" and "arms races" extreme indi
viduals can be the most successful (up to a point), perhaps enough so to make up for
extremely unattractive sons that a large variance also produces. But again, an un
usual degree of variance in an ornament could be due to either runaways and many
of the hypothetical arms race causes we have considered. This unenlightening conclu
sion suggests that perhaps the best strategy is to consider the function of each orna
ment individually and not expect that the form of an ornament will immediately
reveal its significance.

Florida Entomologist 80(2)

Ornamentation and Mating Systems

Let us assume that ornate signals are advertisements of male (or less frequently,
female) qualities directed to potential mates and / or sexual rivals. Do these organs of
propaganda occur in any sort of pattern? Are they associated with certain behaviors
and are these behaviors typical of particular mating systems?
There are circumstances where an individual can profitably advertise and situa
tions where it cannot (Burk 1981; Prokopy 1980). One place where there is little profit
in investing in an ornament is where females are predictably located at resources,
(e.g., oviposition sites), and these resources are discrete, scattered and rare. Males can
then wait by the resource and attempt to copulate with an arriving female. Under
these conditions it might be more beneficial for her to immediately mate rather than
spending time and energy choosing a particular male, all the while being distracted
from exploiting the resource. Where there is little opportunity for females to choose,
there is no reason for males to advertise (e.g., Sivinski 1984). If the resource is small
enough for a male to exclude its rivals, then signals directed to competitors can evolve.
Where males cannot predictably locate females by waiting by a resource (e.g., the re
source is common relative to females), then the costs of mate choice are lower, females
may be able to afford to discriminate among males, and males may compete for atten
tion by producing signals.
Can this scheme explain the occurrence of ornaments in flies? Some instances
seem to be textbook examples of the "resource distribution model of sexual selection".
For example, antlered males of Phytalmia spp guard rare, scattered oviposition sites,
"pin holes" in the freshly fallen trunks of particular trees. They dispute with rivals for
control of the resource, through displays of their horns and combat, and females that
attempt to use it must mate with the resident male (Dodson 1987, 1989). The elabo
rate leg decorations of Calotarsa and the facial setae of Ceratitis, which are presum
ably used to communicate with females, adorn males that participate in swarms and
leks, respectively These male aggregations are formed solely for the purpose of mat
ing and in the absence of any of the resources females require (e.g., Sivinski & Peters
son 1996). The sex life of many ornamented flies is unknown, and how well resource
distribution explains ornamentation in general remains to be seen.


While resource distribution seems to be successful in explaining why ornaments
have evolved in certain instances, there are puzzling phyletic patterns (Table 1). Eye
stalks and antlers are concentrated among the acalypterate families. Resource guard
ing is commonly described in acalypterates, but is also found in a number of other
Diptera, including the calypterates which are conspicuous by the scarcity of their or
naments. Also puzzling is the apparent scarcity of elaborate ornaments displayed in
acalypterate courtships (outside of the Tephritidae and related families). Mating be
haviors are often complex and include movements of head and legs, organs orna
mented in other taxa (e.g., section "Conclusion: the locations of ornaments"). Rather
there seems to be a concentration of intersexually selected ornaments in the more
primitive Brachycera.
There is considerable variance in the range of ornamentation within a family Why
are the Dolichopodidae so rich in decorations? Or perhaps even more curious, why
does ornamentation sporadically evolve in otherwise ordinary appearing taxa? The
complicated waving of huge blue leg paddles in Sabethes spp. make up the onlycourt
ships described in the Culicidae! Can resource distributions alone account for either

June, 1997

Behavioral Ecology Symposium '96: Sivinski

the commonness or the rarity of ornaments within various taxa? Are there other fac
tors involved?
Why do dolichopodids seem to bear so many and such various ornaments, on an
tennae, faces, mouthparts, legs and abdomens? As predators, females may not be con
centrated onto a small resource that males can control and this might encourage male
advertisement. But other orthorrhaphous Brachycera, such as the similarly preda
ceous asilids and the closely related empidids, are only occasionally ornamented. One
possible explanation is that dolichopodids, unlike many asilids and empidids, gener
ally feed on small prey that they glean from a surface; i.e., they spend a good deal of
time standing and walking (e.g., Chvala 1976). It may be easier to present a compli
cated display involving the movement of patterned body parts while both parties have
their feet upon the "ground" (or the water's surface in the case of some Campsicne
mus). At least some of the ornamented robber flies both forage for food and display to
mates on substrates, e.g., tree trunks (Wilcox & Martin 1936). Those insects that re
veal their ornaments in flight (e.g., Calotarsa), fly in a slow dignified manner that al
lows their decorations to be seen (Sivinski & Petersson 1996).
Why Sabethes should differ so much from other mosquitoes is a mystery, although
there are two factors that might contribute to their unique ornamentation. First, the
tribe Sabethini is diurnal. Shannon (1931) in Brazil and Haddow & Corbet (1961) in
Africa noted that diurnal mosquitoes were more brightly colored than the drab spe
cies active at twilight or during the night. They presumed that coloration was useless
in the dark. Second, the mating system of Sabathes does not include male swarms or
males waiting by emergence sites, both common behaviors in the Culicidae (see Han
cock et al. 1990). Rather, males patrol areas searching for resting females on twigs, or
occasionally pursue flying females until they land. As in the dolichopodids, there is
more of a stage available for their showmanship than is typical for a mosquito.


Wonders occur everywhere along the bodies of flies. Ornaments that appear to be
used in aggressive interactions with members of the same sex seem to be concentrated
on the head. Since the head is often used in the pushing style of confrontation and
combat typical of Diptera, such elaborations are probably embellishments of weapons
or advertisements of size and the ability to use weapons. They may then take on a pre
sumably secondary function by advertising sexual competitiveness to potential mates
(e.g., stalk-eyes). The rare instances of female ornamentation, swellings and glandu
lar (?) projections are concentrated on the abdomen. The reproductive organs are
likely to be a focus of male interest and where females would center their propaganda.
Male ornaments that appear to be solely directed to females are more widespread, but
still are concentrated in the anterior regions of the body, the head, and fore and mid
The prominence of legs as platforms for signals may be because of their mobility
Movement might enhance perception of the ornament because objects in motion are
more apparent to insect compound eyes. Alternatively, it could be the movement itself
that is embellished by the ornament; i.e., displays of coordination, timing and flexibil
ity made more impressive by the equivalent of a cheerleader's pom poms (or as W. B.
Yeats might say ... "how can we tell the dancer from the dance").
Evidence for it being the motions that are enhanced by the ornaments comes from
the common employment of unornamented legs in communications between flies.
Male forelegs, without decoration, are often used by flies to brush the female's face and
eyes during courtship and copulation. For example, when mating, male Platystoma

Florida Entomologist 80(2)

0 < 0

Q 0


0 =O



rM uH


June, 1997

on c











.: C


1o 0+

o 0

cv. Ic.

< 0 r D 2

< 0 <

< <


, oZ p
Cf 2<

a?- 73



10 0+ 10 10 10 10 10
so soooss

o 0






(^ ;
`E ^a

X I ^
g ,T~

Behavioral Ecology Symposium '96: Sivinski 159

U <
m < n


r Z

W Cc
Z rr, U H a
< m o

0 0-M M
< on

ZZ -

0- IJ-1 F, ~
"03 a M

2 -c8?
Cr <

0 4

0 N~

S 0 2Ko a
Wsg I ;I

<0 n < 0

crZ~ C
Cr -

~Y3 S SI ^3

C rM H < E .
^2 7-^ n:

27 Cr Q2 CO

5^|sw Is

3cr 2 S c


0Co < r-
2 Q~ c y,

? C M Ha;

r^ il I ,vll

Florida Entomologist 80(2)

seminaionis F signal the start of a bout of nuptial feeding with a regurgitant by mov
ing their front legs from the base of the female's wings to the inner margins of her eye
(Michelmore 1928). In a similar vein, copulating males of the micropezid Cardiaceph
ala myrmex alternatively scratch and regurgitate onto their mate's eyes (Wheeler
1924). In Mexico, mounted males of the asilid Efferia cressoni (Hine) rest their fore
tarsi on the females eyes (Dennis et al. 1986). However, in Wyoming they do not. Per
haps the mechanics of copulation remain the same, while selection on signaling does
not. In addition to the actual placing of tarsi on the females' eyes, male flies may wave
relatively unmodified front legs from a distance (e.g., Alcock & Pyle 1979; Spieth
1982). Both forms of signaling, the placing of the foretarsi on (or very near) the female
eye and motions from a distance, might provide more information (or misinformation)
when a more conspicuous front leg is employed. Plain midlegs are also sometimes used
to signal. For example, the particularly complex courtship of the ottiidPhysiphora de
mandate (F) includes sessions where the male raises the middle leg with its light col
ored tarsi on the side away from the female (Alcock & Pyle 1979). Mounted males of
the dolichopodid Scapius platypterus rest their front legs over the female's head while
the midlegs are held to the side near her eyes and waved back and forth (Grootaert &
Mueffels 1988). The unornamented mosquito, Sabethes chloropterus (Humboldt),
quivers its plain midtarsi against its mate's antennae during copulation (Hancock et
al. 1990). Its relative, S. cyaneus, has apparently escalated the display by using spec
tacularly plumed midlegs in a complex visual and tactile sexual performance.
Though wings are mobile, ornamented examples are rare in true flies. Perhaps the
single pair is too critical to survival to bear the additional costs of carrying elaborate
signals. The same combination of mobility and relative expendability characteristic of
fly legs may have concentrated many of the more spectacular displays of birds' onto
their tails.


Martin Aluja, Gary Dodson, D. E. Hardy, James Lloyd, Gary Steck, Bob Sullivan
and Steve Wing all made many valuable comments. Kevina Vulinec displayed her art
istry by drawing the illustrations from specimens provided by the Arthropod Collec
tion at the Florida Division of Plant Industry, Gainesville, Florida. The manuscript
was prepared, with her usual speed and competence, by Valerie Malcolm.


ALCOCK, J., AND D. W. PYLE. 1979. The complex courtship behavior of Physiphora de
mandate (F) (Diptera: Otitidae). Z. Tierpsychol 49: 352-362.
ALDRICH, J. M. 1929. Diptera of Patagonia and South Chile. Part VII. Cyclorrhapha
ALDRICH, J. M. 1932. New Diptera, or two-winged flies, from America, Asia, and Java,
with additional notes. U.S. Nat. Mus. Proc. 81: 128.
ALEXANDER, C. P. 1936. The crane-flies (Tipulidae) of New England fourth supple
mentary list. Occas. Pap. Bost. Soc. 8: 273-292.
ALEXANDER, R. D., D. C. MARSHALL, AND J. R. COOLEY. 1997. Evolutionary perspec
tives on insect mating, in B. J. Crespi and J. C. Choe (eds.) Social Competition
and Cooperation in Insects and Arachnids, Vol. 1. Evolution of Mating Systems.
Cambridge Press, Cambridge (in press).
ARITA, L. H., AND K. Y. KANESHIRO. 1989. Sexual selection and lek behavior in the
Mediterranean fruit fly, Ceartitis capitata (Diptera: Tephritidae). Pac. Sci. 43:

June, 1997

Behavioral Ecology Symposium '96: Sivinski

BEZZI, M. 1924. Further notes on the Ethiopian fruit-flies, with keys to all the known
genera and species. Bull. Entomol. Res. 15: 73118.
BRISTOWE, W. S. 1925. XXI. Notes on the habits of insects and spiders in Brazil.
Trans. Entomol. Soc. London 1924; 475-503.
BROMLEY, S. W. 1933. Courting and mating performances of an asilid fly Psyche. 40: 144.
BURK, T. 1981. Signaling and sex in acalypterate flies. Florida Entomol. 64:30-43.
BURKHARDT, D., AND I. DE LA MOTTE. 1983. How stalk-eyed flies stalk: observations
and measurements of the eyes of Crytodopsis white (Diopsidae, Diptera). J.
Comp. Physiol. 151: 407-421.
BURKHARDT, D., AND I. DE LA MOTTE. 1988. Big 'antlers' are favored: female choice in
stalk-eyed flies (Diptera, Insecta), field collected harems and laboratory exper
iments. J. Comp. Physiol. 162: 649-652.
BURKHARDT, D., I DE LA MOTTE, AND K. LUNAU. 1994. Signaling fitness: larger males
sire more offspring. Studies of the stalk-eyed fly Cyrtoiopis white (Diopsidae,
Diptera). J. Comp. Physiol. A. 1774: 61-64.
CARPENTER, G. D. H. 1923. Notes on Ugandan insects. The Hope Reports. 12: 33.
CHVALA, M. 1976. Swarming, mating and feeding habits in Empididae (Diptera) and
their significance in evolution of the family Acta. Entomol. Bohemoslov. 73:
COLE, F. R. 1969. The Flies of Western North America. Univ. of California Press, Ber-
CUMMING, J. M. 1994. Sexual selection and the evolution of dance fly mating systems
(Diptera: Empididae; Empidinae) Canadian Entomol 126: 907-920.
CURRAN, C. H. 1933. Some North American Diptera. Am. Mus. Nat. Hist. Novit. 682:
DAWKINS, R. 1982. The Extended Phenotype. W. H. Freeman and Co., Oxford.
DAWKINS, R. 1986. The Blind Watchmaker. W. W Norton and Co., New York.
DENNIS, D. S., R. J. LAVIGNE, AND S. W. BULLINGTON. 1986. Ethology of Efferia cres
soni with a review of the comparative ethology of the genus (Diptera: Asilidae).
Proc. Entomol. Soc. Washington 88: 4255.
DOANE, R. W. 1907. Notes on the habits of Scellus viragoAld. Entomol. News 18: 136
DODSON, G. 1987. Tephritid mating systems, with special reference to antlered flies.
Entomol. Soc. Qld. News Bull. 1987: 5962.
DODSON, G. 1989. The horny antics of antlered flies. Aust. Nat. Hist. 22: 604-611.
DOWNES, J. A. 1970. The feeding and mating behavior of the specialized Empidinae
(Diptera); observations on four species of Rhamphomyia in the high arctic and
a general discussion. Canadian Entomol. 102: 769-791.
EBERHARD, W. G. 1985. Sexual Selection and Animal Genitalia. Harvard Univ. Press,
Cambridge, Massachusetts.
EBERHARD, W. G. 1994. Evidence of widespread courtship during copulation in 131
species of insects and spiders, and implications for cryptic female choice. Evo
lution 48: 711 733.
EDWARDS, F. W. 1920. Scent organs in female midges of the Palpomyia group. Annu.
Mag. Nat. Hist. Ser. 9, 5: 365-368.
ENDERLEIN, G. 1920. Zur kenntnis tropischer frucht-bothifliegen. Zool. Jahrb. Abt.
Syst. Geogr. Biol. Tiere. 43: 336-360.
EVANS, H. E. 1988. Observations on swarms of Rhamphomyia sociabilis (Williston)
(Diptera: Empididae). J. New York Entomol. Soc. 96: 316-322.
FROHNE, W. C. 1959. Predation of dance flies (Diptera: Empididae) upon mosquitoes
in Alaska, with special reference to swarming. Mosq. News 19: 711.
GREENE, C. T. 1922. Synopsis of the North American flies of the genus Tachytrechus.
U. S. Nat. Mus. Proc. 60: 121.
GREEN, A. J. 1992. Positive allometry is likely with mate choice, competitive display
and other functions. Anim. Behav. 43: 170-172.
GREENE, C. T. 1924. Synopsis of the North American flies of the genus Scellus. U. S.
Nat Mis. Proc. 26 60-64.

Florida Entomologist 80(2)

GRIMALDI, D. 1987. Phylogenetics and taxonomy of Z .... .. (Diptera: Drosophil
idae). Bull. Am. Mus. Nat. Hist. 186: 103-186.
GRIMALDI, D., AND G. FENSTER 1989. Evolution of extreme sexual dimorphisms:
structural and behavioral convergence among broad-headed Drosophilidae
(Diptera). Am. Mus. Nov. 2939: 125.
GROOTAERT, P., AND H. J. G. MUEFFELS. 1988. Observations sur I'accoupment chez
Sciapus platypterus (Diptera Dolichopodidae). Bull. Annls. Soc. r. Belg. Ento
mol. 124: 275-276.
GRUHL, K. 1924. Paarungs gewohnheit der Dipteren. Zeitschr. f. Wiss. Zool. 122: 205.
HADDOW, A. J., AND P. S. CORBET. 1961. Entomological studies from a high tower in
Mpanga forest, Uganda. V. Swarming activity above the forest. Trans. Roy En
tomol. Soc. London 113: 284-300.
HANCOCK, R. G., W. A. FOSTER, AND W. L. YEE. 1990. Courtship behavior of the mos
quito Sabethes cyaneus (Diptera: Culicidae). J. Insect Behav. 3: 401-415.
HARDY, D. E. 1973. The Fruit Flies (Tephritidae-Diptera) of Thailand and Bordering
Countries. Pacific Insects, Monograph # 31, Bishop Mus. Honolulu, Hawaii.
HARMSTON, F. C., AND G. F. KNOWLTON. 1942. The dipterous genus Campsicnemus in
North America. Bull. Brooklyn Entomol. Soc. 37: 10-17.
HARMSTON, F. C., AND G. F. KNOWLTON. 1943. Five new species of Dolichopodidae.
Bull. Brooklyn Entomol. Soc. 38: 101-107.
HULL, F. M. 1937. A megamorphic and two curiously mimetic flies. Psyche 44: 116-121.
HULL, F. M. 1962. Robber Flies of the World. Smith. Inst. U. S. Nat. Mus. Bull. 224
part 1.
IWASA, Y., AND A. POMIANKOWSKI. 1994. The evolution of mate preferences for multi
ple sexual ornaments. Evolution 48: 853-867.
KANMIYA, K. 1989. Discovery of the genus Gampocera Schiner from Hawaii, with a de
scription of a new species (Diptera: Chloropidae). Proc. Hawaiian Entomol. Soc.
29: 149-153.
KESSEL, E. L. 1955. The mating habits of balloon flies (Diptera: Empididae). Syst.
Zool. 4: 97-104.
KESSEL, E. L. 1963. The genus Calotarsa, with special reference to C. insignis Aldrich
(Diptera: Platypezidae). Wasmann J. Biol. 19: 295-299.
KREBS, J. R., AND R. DAWKINS. 1978. Animal signals: mind-reading and manipulation,
pp. 380-402 in J. R. Krebs and N. B. Davies (eds.) Behavioral Ecology Blackwell
Sci. Pub., Oxford.
LAND, M. F. 1993. The visual control of courtship behaviour in the fly Poecilobothrus
nobilitatus. J. Comp. Physiol. A 173: 595-603.
LUNAU, K. 1992. Mating behavior in the long-legged fly Poecilobothrus nobilitatus L.
(Diptera, Dolichopodidae): courtship behavior, male signaling and mating suc
cess. Zool. Beitr. N. F 34: 465-479.
LUNDBECK, W. 1912. Diptera Dannica 4: Dolichopodidae. Copenhagen.
MCALPINE, D. K. 1973. Observations on sexual behavior in some Australian Platysto
matidae (Diptera, Schizophora). Rec. Australian Mus. 29: 110.
MCALPINE, D. K. 1975. Combat between males of Pogonortalis doclea (Diptera,
Platystomatidae) and its relation to structural modification. Australian Ento
mol. Mag. 2: 104-107.
MCALPINE, D. K. 1976. Spiral vibrissae in some clusid flies (Diptera, Schizophora).
Australian Entomol. Mag. 3: 7578.
MCALPINE, D. K. 1979. Agnostic behavior in Achias australis (Diptera, Platystoma
tidae) and the significance of eye stalks, pp. 221-230 in M. Blum and N. Blum,
Sexual Selection and Reproduction. Academic Press, New York.
MCALPINE, D. K. 1994. Review of the species of Achias (Diptera: Platystomatidae). In
verteb. Taxon. 8: 117-281.
MCALPINE, D. K., AND M. A. SCHNEIDER 1978. A systematic study of Phytalmia
(Diptera, Tephritidae). Syst. Entomol. 3: 159-175.
MICHELMORE, A. P. G. 1928. A mating habit of Platystoma seminationis, F (Diptera,
Ortalidae). The Entomol. 61: 241 243.

June, 1997

Behavioral Ecology Symposium '96: Sivinski

MILLER, D. 1923. Material for a monograph on the Diptera fauna of New Zealand.
Part 6. Empididae. Trans. New Zealand Inst. 54: 440.
MOLLER, A. P. 1992. Female swallow preference for symmetrical male sexual orna
ments. Nature 357: 38-240.
DE LA MOTTE, I., AND D. BURKHARDT. 1983. Portrait of a stalk-eyed fly. Naturwissen
schaften 70: 451-461.
MOULDs, M. S. 1978. Field observations on the behavior of a north Queensland spe
cies of Phytalmia (Diptera: Tephritidae). J. Australian Entomol. Soc. 16: 347
MUNRO, H. K. 1949. A remarkable new species of trypetid fly of the genus Ceratitis
(sensu strict) from East Africa in the collection of the United States National
Museum. Proc. U. S. Nat. Mus. 99: 499-501.
NEWKIRK, M. R. 1970. Biology of the longtailed dance fly, Rhamphomyia longicaudata
(Diptera: Empididae); a new look at swarming. Ann. Entomol. Soc. Am. 63:
PARMENTER, L. 1952. The courtship of Diptera. Proc. Lond. Entomol. Nat. Hist. Soc.
1952-53: 104-109.
POMIANKOWSKI, A., AND A. P. MOLLER 1995. A resolution of the lek paradox. Proc.
Roy Soc. London B 260: 21-29.
POULTON, E. B. 1913. Empididae and their prey in relation to courtship. Entomol.
Mon. Mag. 49: 177-180.
PROKOPY, R. 1980. Mating behavior of frugivorous Tephritidae in nature. pp. 37-46 in
Proc. Symp. Fruit Fly Problems, XVI Intl. Cong. Entomol., Kyoto.
QVICK, U. 1984. A case of abnormal mating behaviour of Dolichopus popularis Wied.
(Dolichopodidae, Diptera). Notul. Entomol. 64: 93.
SCHNEIDER, M. A. 1993. A new species of Phytalmia (Diptera: Tephritidae) from
Papua New Guinea. Australian Entomol. 20: 38.
SEGUY, E. 1923. Faune de France 6 Dipteres Anthomyides. Les Presses Universitaires
de France. Paris.
SHANNON, R. C. 1931. On the classification of Brazilian Culicidae with special refer
ence to those capable of harboring the yellow fever virus. Proc. Entomol. Soc.
Washington 33: 125-164.
SHILLITO, J. F. 1960. A bibliography of the Diopsidae. J. Soc. Bibl. Nat. Hist. 3: 337
SHILLITO, J. F. 1976. Bibliography of the Diopsidae-II. J. Soc. Bibl. Nat. Hist. 8: 65-73.
SILVESTRI, F. 1914. Report of an expedition to Africa in search of the natural enemies
of fruit flies. Terr. Of Hawaii Brd. Agr. Forest. Bull #3.
SIVINSKI, J. 1984. Sexual conflict and choice in a phoretic fly, Borborillus frigipennis
(Sphaeroceridae). Ann. Entomol. Soc. Am. 77: 232-235.
SIVINSKI, J. 1988a. Unusual female-aggregated mating systems in phorid flies. J. In
sect Behav 1: 123-128.
SIVINSKI, J. 1988b. What do fruit fly songs mean? Florida Entomol. 71: 462-466.
SIVINSKI, J., AND G. DODSON. 1992. Sexual dimorphism in Anastrepha suspense
(Loew) and other tephritid fruit flies: possible roles of developmental rate, fe
cundity and dispersal. J. Insect Behav. 5: 491-506.
SIVINSKI, J., AND E. PETERSSON. 1997. Mate choice and species isolation in swarming
insects, in B. Crespi and J. C. Choe (eds). in Social Competition and Coopera
tion in Insects and Arachnids, Vol. 1. Evolution of Mating Systems. Cambridge
Press, Cambridge (in press).
SMITH, K. G. V. 1959. A note on the courtship and predaceous behavior of Neurigonia
species (Dipt., Dolichopodidae). Entomol. Mon. Mag. 95: 32-33.
SMITH, K. G. V. 1975. Studying the behaviour of Diptera, pp. 241-249 in A. Stubbs and
P. Chandler (eds.) A Dipterist's Handbook. The Amateur Entomologist Soc.,
Richmond, Surrey
SNOW, W. A. 1884. American Platypezidae. Kansas Univ. Quart. 3: 143-152.
SPIETH, H. T. 1982. Behavioral biology and evolution of the Hawaiian picture-winged
species group of Drosophila. Evol. Biol. 14: 351-437.

Florida Entomologist 80(2)

STEYSKAL, G. 1938. The pre-copulatory behavior of the male of Dolichopus omnivagus
Van Duzee (Diptera, Dolichopodidae). Bull. Brook. Entomol. Soc. 33: 193-194.
STEYSKAL, G. 1941. A curious habit of an empidid fly. Bull. Brook. Entomol. Soc. 36:
THORNHILL, R. 1992. Female preference for the pheromone of males with low fluctu
ating asymmetry in the Japanese scorpionfly (Panorpajaponica: Mecoptera).
Behav. Ecol. 3: 277-283.
VAN DUZEE, M. C. 1924. A revision of the North American species of the dipterous ge
nus Chrysotus. Buffalo So. Nat. Sci. Bull. 13: 353.
VAN DUZEE, M. C. 1927. North American species of Polymedon (Diptera, Dolichopo
didae). Ann. Entomol. Soc. Am. 20: 123-126.
VAN DUZEE, M. C. 1932. New North and South American Dolichopodidae, with notes
on previously described species. Am. Mus. Nat. Hist. Novit. 569: 122.
VAN DUZEE, M. C., AND C. H. CURRAN. 1934. Key to the males of nearctic Dolichopus
Latreille (Diptera). Am. Mus. Nat. Hist. Novit. 683: 126.
VERRALL, G. H. 1901. British Flies. vol 8, Syrphidae. E. W Classey, Hampton, En
VERRAL, G. H. 1905. List of British Dolichopodidae with tables and notes. Entomol.
Mon. Mag. 16: 188-196.
WATSON, P. J., AND R. THORNHILL. 1994. Fluctuating asymmetry and sexual selec
tion. Trends Ecol. Evol. 9: 21-25.
WALLACE, A. F. 1869. The Malay Archipelago. Dover Press, New York.
WEST-EBERHARD, M. J. 1984. Sexual selection, competitive communication and spe
cies-specific signals in insects, pp. 284-324 in T Lewis (ed.) Insect Communica
tion. Academic Press, London.
WHEELER, W. M. 1896. A new empid with remarkable middle tarsi. Entomol. News 7:
WHEELER, W. M. 1924. Courtship of the Calobates. J. of Herd. 15: 485-495.
WHITE, I. M. 1988. Tephritid flies (Diptera: Tephritidae). Handbooks for the Identifi
cation of British Insects. 10: 1134.
WILKINSON, G. S. 1993. Artificial selection alters allometry in the stalk-eyed fly Cyr
todiopsis dalmanni (Diptera: Diopsidae). Genet. Res. Camb. 62: 213-222.
WILKINSON, G. S., AND P. R. REILLO. 1994. Female choice response to artificial selec
tion on an exaggerated male trait in a stalk-eyed fly Proc. Roy Soc. London B.
225: 16.
WILKINSON, G. S., AND G. DODSON. 1996. Function and evolution of antlers and eye
stalks in flies. in B. J. Crespi and J. C. Choe (eds.) Social Competition and Co
operation in Insects and Arachnids: Vol 1. Evolution of Mating Systems. Cam
bridge Press, Cambridge (in press).
WILCOX, J., AND C. H. MARTIN. 1936. A review of the genus Cryptopogon Loew in
North America (Diptera-Asilidae). Entomol. Am. 16: 195.

June, 1997

Behavioral Ecology Symposium '96: Gushing


The College ofWooster
Biology Department
931 College Street
Wooster, Ohio 44691


Myrmecomorphs are arthropods that have evolved a morphological resemblance to
ants. Myrmecophiles are arthropods that live in or near ant nests and are considered
true symbionts. The literature and natural history information about spider myrme
comorphs and myrmecophiles are reviewed. Myrmecomorphy in spiders is generally
considered a type of Batesian mimicry in which spiders are gaining protection from
predators through their resemblance to aggressive or unpalatable ants. Selection
pressure from spider predators and eggsac parasites may trigger greater integration
into ant colonies among myrmecophilic spiders.

Key Words: Araneae, symbiont, ant-mimicry, ant-associates


Los mirmecomorfos son artr6podos que han evolucionado desarrollando una seme-
janza morfol6gica a las hormigas. Los Myrmec6filos son artr6podos que viven dentro
o cerca de nidos de hormigas y se consideran verdaderos simbiontes. Ha sido evaluado
la literature e informaci6n de historic natural acerca de las aranas mirmecomorfas y
mirmec6filas El myrmecomorfismo en las aranas es generalmente considerado un
tipo de mimetismo Batesiano en el cual las aranas estan protegi6ndose de sus depre
dadores a trav6s de su semejanza con hormigas agresivas o no apetecibles. La presi6n
de selecci6n de los depredadores de aranas y de parasitos de su saco ovopositor pueden
inducir una mayor integraci6n de las aranas mirmec6filas hacia las colonies de hor

Myrmecomorphs and myrmecophiles are arthropods that have evolved some level
of association with ants. Myrmecomorphs were originally referred to as myrmecoids
by Donisthorpe (1927) and are defined as arthropods that mimic ants morphologically
and/or behaviorally The literature on myrmecomorphs is enormous and has recently
been reviewed by McIver & Stonedahl (1993).
Myrmecophiles were defined by Donisthorpe (1927) as arthropods that live in or
near ant nests. Wasmann (1894) developed a classification system for myrmecophiles
consisting of distinct categories, each suggesting increasing specialization and inte
gration into the host colony However, as pointed out by H6lldobler & Wilson (1990),
such categorization of myrmecophiles can be misleading as some guests take on mul
tiple roles within a colony.
McIver & Stonedahl (1993) stated that myrmecomorphy and myrmecophily both fall
under the general category of ant mimicry, since even myrmecophiles which lack mor
phological resemblance to ants may mimic chemical or textural characters of their hosts.
However, myrmecophiles may not mimic their hosts in any way and may simply be tol

Florida Entomologist 80(2)

erated by their otherwise aggressive hosts because they are either neutral in odor or are
below some critical size to be recognized by the hosts as intruders (Cushing 1995a). Be
cause of this and because the selective pressures involved in the evolution of myrmeco
morphy and myrmecophily are quite different (discussed below), it is more useful to
view these as separate phenomena and not as subcategories under ant mimicry.
The literature on myrmecomorphs and myrmecophiles in general has been sum
marized by McIver & Stonedahl (1993) and by H11dobler & Wilson (1990). The pur
pose of the present paper is to expand coverage of myrmecomorphs and
myrmecophiles in the Order Araneae.


Table 1 presents information about known spider myrmecomorphs. The putative
ant models are those to which the mimics bear a generic or specific resemblance and
which are sympatric with the mimics. In fact, the majority of the models presented are
found in the same microhabitat as the mimics and are often collected with them. De
tails about the natural history of the mimics or about the form of their mimicry are
also presented. As far as possible, the taxonomy of the spider myrmecomorphs follows
that presented by Brignoli (1983) or Platnick (1993). The taxonomy of the models fol
lows that presented by Bolton (1995).

Morphological and Behavioral Adaptations

The morphological adaptations involved in achieving a resemblance to ants among
spider myrmecomorphs were first discussed by Banks (1892). Reiskind (1972, 1977)
lists and illustrates these morphological adaptations and they are described in
McIver & Stonedahl (1993). They include a variety of color and body-form modifica
tions that give the spider the appearance of having three body segments instead of
two and of having long, narrow legs instead of shorter, more robust legs. Mandibles,
compound eyes and even stings are sometimes mimicked by the spiders through mod
ifications in the chelicerae, pigmentation in the cuticle, or special positioning of the
spinnerets. In many cases, the extent to which the mimics resemble a particular
model is extraordinary (see Fig. 1). Reiskind (1977) compares specific features of the
mimic with similar features in the model which enhance the many cases of species
specific mimicry found among spider myrmecomorphs.
The overall body of spider myrmecomorphs is much narrower than non-mimics,
and this appears to reduce their fecundity. Female myrmecomorphs lay fewer eggs per
eggsac than non-mimetic spiders of similar size (Bristowe 1939, 1941, Collart 1941,
Edmunds 1978, Wanless 1978, Bradoo 1980, Boev6 1992). However, myrmecomorphs
may compensate for this limitation by laying more eggsacs so that their life-time fe
cundity may be about equal to that of non-mimetic spiders.
McIver & Stonedahl (1993) list myrmecomorphs which show morphological, be
havioral, or pattern mimicry. All spider myrmecomorphs are morphological mimics,
and the majority are also behavioral mimics. Spider myrmecomorphs move in a much
more erratic, more ant-like fashion than non-mimics. This behavior is described
throughout the literature for most of the species of myrmecomorphs (Pocock 1908, Do
nisthorpe 1927, Bristowe 1941, Marson 1946, 1947, Reiskind 1972, 1977, Wanless
1978, Wing 1983, Brignoli 1984, Fowler 1984, Oliveira 1988, Lighton & Gillespie
1989, Boev6 1992). Behavioral mimicry also involves raising either the first or second
pair of legs and using them to mimic the movements of antennae (Reiskind 1977,
Jackson 1986). This functionally reduces the number of legs in the mimic from four

June, 1997

Behavioral Ecology Symposium '96: Gushing

a m^


Fig. 1 (drawn from photographs in Reiskind 1977). a) A female Zuniga magna
Peckham (Salticidae). Note how the front pair of legs is used as pseudo-antennae and
how the abdominal constriction mimics the third body segment of the model. b) A fe
male Mymarachneparallela (Fabricius) (Salticidae). Note the constriction of the ceph
alothorax and the lighter band of setae around the mid-section of the abdomen-both
of which add to the illusion of additional body segments. The legs of this mimic have
been effectively "shortened" through the lighter pigmentation of the terminal seg
ments. The darkening of the metatarsal segments of the first pair of legs adds to the
antennal mimicry as it gives the illusion that the pseudo-antennae are being held off
the ground when, in fact, the legs are in contact with the substrate. c) A male Syne
mosyna americana (Peckham) (Salticidae). Note the constrictions of the cephalotho
rax and the abdomen. The color pattern of the spider also closely mimics the
coloration of the model.

Florida Entomologist 80(2)

pairs to three. Sometimes the terminal segments of these mimetic antennae are
darker giving the impression that the mimic has clubbed antennae (Reiskind 1977).
Pocock (1908) suggests that behavioral mimicry may have evolved before morpholog
ical mimicry among spider myrmecomorphs. Bristowe (1941) agrees with this view.

Transformational and Polymorphic Mimicry

Spiders undergo gradual metamorphosis. During the earlier developmental stages
(instars) only smaller ant species found in the vicinity would serve as appropriate po
tential models for young myrmecomorphic spiders. Because of this, it might be pre
dicted that the suite of models would change as the spiders passed through each
successive instar. A mimetic complex in which the identity of the model species
changes as the mimic develops is called transformational mimicry (Mathew 1935) and
has been documented for several species of myrmecomorphic spiders (see Table 1). In
fact, McIver (1989) predicts that transformational mimicry "probably occurs in most
systems where the ant-mimic develops through gradual metamorphosis." Wanless
(1978) believes that transformational mimicry may occur in the majority of Myrma
rachne myrmecomorphic species (Salticidae). In a study of transformational mimicry
complexes among Myrmarachne spp., Edmunds (1978) demonstrated that the model
species involved in each example of transformational mimicry were either positively
associated with one another or tolerated each other's presence in the area. In other
words, the set of models mimicked by each instar of the spider were always present in
the same habitat.
In several species of myrmecomorphic spiders, the adults are polymorphic. It is
thought that each morph mimics a different model. Such polymorphic mimicry ap
pears to be fairly common among myrmecomorphic species (see Table 1). In some
cases, there is sexual dimorphism among the adult spiders and the sexes each mimic
a different model (Reiskind 1970, Cutler 1980, Wanless 1978, Oliveira 1988).
In all these cases of polymorphic mimicry each morph either corresponds to one
model ant species that is also polymorphic or to two or more different model species.
For example, light yellow or brown morphs of Synemosyna aurantiaca mimic
Pseudomyrmex flavidulus (F Smith) and P oculatus (F. Smith) while black morphs
mimic P gracilis (Fabricius) and P sericeus (Mayr) (Table 1 and Oliveira 1986). In
these polymorphic mimicry systems, it is not known to what extent the different color
forms of the mimic are sympatric nor to what extent the polymorphism is a result of
differential predation. Predators could be eliminating the "wrong" color morph from
an area where its model is absent creating an apparent geographic separation of the
different morphs or the different color morphs could be genetically distinct.

Adaptive Significance of Myrmecomorphy

McIver & Stonedahl (1993) discuss the adaptive significance of myrmecomorphy in
depth. Four different hypotheses have been proposed to explain myrmecomorphy: 1)
Wasmannian mimicry, 2) Mtillerian mimicry, 3) Aggressive, or Peckhammian mim
icry, and 4) Batesian mimicry. In Miillerian mimicry, both the model and the mimic
are unpalatable. As Mclver and Stonedahl point out, the hypothesis that myrmeco
morphs are Miillerian mimics is not well supported, especially for spider myrmeco
morphs. Although the ant models may be unpalatable to most predators, there is no
evidence that the spider mimics are unpalatable. Therefore, this hypothesis will not
be discussed.
Wasmannian mimicry involves the evolution of resemblances between a model
and its mimic that facilitates a mimic living with its host (Rettenmeyer 1970). Retten

June, 1997

Behavioral Ecology Symposium '96: Gushing






NZ o



C;6 00~
ciC r- i

C x 00 NM
o o

-D aa
2 ^ 2
o3 Qt 0 C t-

0 c^ ^ 3 V3 C

C -c

c ~c3 "g

Cz C.E m
co -o

C C 9


2j -o~v
S a o

4 s

R4 R R



b ~ E
o n
H^ Eon ^ ^ OR
0 z- ^ 0 0 C

t o I I
a -f i o
yi *y g) r

i/l i/ 11 q
,^ ,^^ ^
o c+~J. *+J^ +J ~




- t-

C- V




S-c "

aaC 9

$ co2 on
3 "
;O (
3 ^.^
(3 - -
Ci Ci Ci CO'

Florida Entomologist 80(2)





4 U"




00 00

00 i

to to

C o CM
-: 2 non
- U C
C" C -








C- V
U C-
0 Cv



C -



fc; -


U -
c- E

E t
A" C




- 3 t
1 Li tZ' t
Y 3 so ? 8 vW4
S ^ ^ ^ V)

N C c~ c
- ? N C
a;E -s -^ e ^a
o c CQ

II I I L | ^i 5
'il ~ vl ^i i c
sS c
3 B ~

r- r-2 -

; s
* *-i *i *i *i *
Vi; V Vl Vl C S
-' -'3 -'3 -3 -3 -3 "
C'o) C CC C C

June, 1997

M '9

^ 5 E
caS I 4 w
C a ^ e-

Li (D S if

m~ 0,
:9 IA -
e C -^ g

Behavioral Ecology Symposium '96: Gushing




2; L


C .
CD 0
on .t

t>- LO Oi 0
CD 00

N- (DCC -

t: : U













.g F
4; S)

c II




I~ C

s? -S -

^ ho a; S
;Q -^ O^ti

F11 R

.3 .3 8 l
^ q ^ ^'^ ^ q

2 z

a h

s' o 1

















Florida Entomologist 80(2)








-E F

a- a




c'J c~
at o

o a





E s

June, 1997







N- N-

0- C-
o a

0 0

C Mt


'S -o

s K

bS :


Kr, -0

- ^
O to
~s e^





Behavioral Ecology Symposium '96: Gushing 173

7o C
N- E
i- yi rlv
CQ 0 tO L0 Q^

C N N- v N-
a t at a2 I
Cc (- -'!)
c. -f -f -a c

at c c aN--
3 iS cn 0) en m n
I K i 2! ^ i ^ s
i I C. I I. it 1 N
o a, C -
0 a '2 E -a -C A
2 em mO 3 a ca ~ o-
d o o o ^; o o cn

- lo -'
C C Z C it S
0^ 0 0 0

" 3E ima s ^
Qt C-~ C l

|C I i I ri

< M c a o; ^ ^
w" E 3 E C-E E

O! ^N Y 1 I II t:o

^ S g^.a' g.i ln
i L L, Er S S' q l .p r '
BM1 [Sa, ^^^ s a, Oa-
C-2 C" CE

7 A (A
"x e 0PE E

,a /-^ ? -

I I II 11 1

oE E .
sr 3
C- 8C
2 2 0
C, _~ :M~
c.? v

x cr 2i o W

o E E
$ ~

7; cE~-~

o 5>
C- Q ( -

Florida Entomologist 80(2)


















m- a- t, vi
/- c-R R S

I E 1 1

^ ~ R -S8
i3 q; q; q q; (
s-~~ ~ y)^ ^ J J


Ci 0


a t




-I c x,

- OE

E ,
<"0 0
_ U t/3

' .a ^
SE S .S_

Co ^


June, 1997






~ .j
C O 3
--- c
K i ^
co ,u


0,0,0 0,

Behavioral Ecology Symposium '96: Gushing


m- E


o R

^ -
*^ 3
r l!
31 1 o

ho '"; ?S S
3 3 a cS c3 L
Si! s S S
p.~ ^ -S
a ^~ f-^ ^

S i-
C O `

*^ S 'e
Q_ i -~ '

-~ CT3 G
6 "^ 6, C
Ir II !r r

'S ,

i it

R^- ^ S
^ '3 R^ E

Florida Entomologist 80(2)

2 00

0 E
-i 'c
m E
00 0 o 3 O f
q oo 0 t- c mo t- o LO 2

-1 e: : : oorrr r
C- 0 fc C- 2 0n ^ 0 C DQ c-
000 -O c cn c- ^ c

--GL^ r ^- r^ CI -CO
< ..g vl aO m 3m. a oc c
y ~ ?2M O;OOOUQ
< 8

o 0

I lvl l II

C- C.S ^ I
Kc v

~< -
2^ i -ia*C,<

S 2 c. t ^

la i1 a

gKj B S *2 -B^ E
5H .a o E c 3 3 ^ 0^

0i0 S P ~ -
sf Iw o s s^ r 3.

-^ ^ 0 ar -
^ ~ ~ ~ ~ c :0 :- cOO ^ *^' ''
,, m r;^ ^D( &
o-3 ~
J -e : -
S Eo ja

C- m bi


June, 1997

Behavioral Ecology Symposium '96: Gushing

c o

N>- Lfl
C "

- C

C iC







-i E

ca- Nac ca
bt 3 E
%~t --,' C-

;~ E -


-~ 2
c .
U C~iji







ti5 5

a, ra < C *

" ^-^ '3 -^ 3re
o og

v rl '" ^ ^3 ^
S E2 SKj l
IIg~~ l~i I ^^1 ^
K ^ ^ r
lllS Il |-
Tv, 9 ^
V K 5

a'a, a o a B 1 a a P a



EE 3
-s ^
re 4;r

1| I

Florida Entomologist 80(2)

June, 1997








ca ca

~- B-

C ~ C Y
am c -Y, .9

e 0 ff -


vi v Ci rr l
re -, -, -, -S, ,

CC ^3re 73 tO


Cc a,

2E E
C, .
C A~





ir ^ ^

S g-g

sqo,,2^ &






Behavioral Ecology Symposium '96: Gushing


C~ -










^ ^

-a c ar

re "-c2> 3

cvl ^s o-oi^
MS ~~~ C "

a2 -CESS L-a

3l &, ,


6 4 -0




Florida Entomologist 80(2)

meyer considers the relationship between the model and mimic to be either exploit
ative on the part of the mimic or beneficial to both the model and the mimic. As
Wasmannian mimics are, by definition, myrmecophiles, they will be discussed in the
section on spider myrmecophiles.
At least some spider myrmecomorphs are clearly aggressive, or Peckhamian mim
ics (Table 1 and McIver & Stonedahl 1993). Aggressive mimicry complexes involve a
predator mimicking its prey (Wickler 1968). In such a system, the prey species acts as
both model and operator (in the terminology of Vane-Wright 1980), or as both model
and selective agent. The aggressive mimics often use both morphological resemblance
as well as behavioral tactics to attract and prey on the models. For example, the th
omisid, Amyciaea forticeps O.P.-Cambridge, assumes the alarm attitude of its model
(abdomen and "antennae" raised). This apparently attracts workers of the model,
Oecophylla sp. (which have good eyesight). When an ant approaches, the spider at
tacks it (Table 1, Hingston 1927, Bristowe 1941). The aphantochilid, Bucranium sp.
carries dead ants of the genus Cephalotes aloft, perhaps as a mimetic device (chemical
mimicry?) to attract other ants (Table 1, Bristowe 1941). This same strategy is used
by Aphantochilus rogersi O.P.Cambridge (Table 1, Oliveira & Sazima 1984). Oliveira
& Sazima (1984) suggest that "close similarity of integument texture (granular) and
pilosity of body and legs (sparse hairs) apparently facilitates the obligatory intimate
contact A. rogersi must make with cephalotines in order to capture an ant among
other ants." The models may, therefore, exert selective pressure for more perfect mim
icry in their own predators.
However, not all myrmecomorphs that prey on their models are aggressive mimics.
In order for the spider to be considered an aggressive mimic, the model must be the
operator, or selective agent. This is unlikely for models which have poor eyesight (the
majority of ants) or which do not approach or investigate the spider. Table 1 lists as
aggressive mimics only those spiders that lure their prey to them using a behavioral
strategy and/or a behavioral strategy combined with morphological similarity
Most myrmecomorphic spiders are probably Batesian mimics (Pocock 1908, Bris
towe 1941, Marson 1947, Reiskind 1977, Edmunds 1978, Wanless 1978, Parker 1984,
Oliveira & Sazima 1984, Oliveira 1986, Parker & Cloudsley-Thompson 1986, Cutler
1991, McIver & Stonedahl 1993). Ants are generally considered to be distasteful, nox
ious, or unpalatable to vertebrate and invertebrate predators. Many species are par
ticularly aggressive and will mob predators that attack individual ants (H11dobler &
Wilson 1990). Others have particularly potent bites or stings or a hard cuticle with
spines making them less appealing prey for most vertebrate and invertebrate preda
tors. Myrmecomorphic spiders would, therefore, gain protection against generalist ar
thropod predators.
However, it has been suggested that myrmecomorphy in spiders is not an example
of Batesian mimicry since there are so many predators that do specialize on ants (Bri
gnoli 1984). The myrmecomorph would be trading one set of predators for another. In
stead, Brignoli (1986) proposed that myrmecomorphy allows the spider "to live in
many different habitats from which most other species, which ants perceive as differ
ent from themselves, are excluded." Certainly, specialized ant predators exist. Certain
species of Crabronid wasps stock their nests with ants (Pocock 1908, Bristowe 1941).
Species of wasps in the genus Tracheliodes are also ant specialists (Krombein 1967).
Some spiders are specialist ant predators (H11dobler 1971, MacKay 1982, Porter &
Eastmond 1982). McIver & Stonedahl (1993) cite additional examples of vertebrate
and invertebrate ant predators.
Nevertheless, Edmunds (1978) points out that myrmecomorphy in spiders proba
bly provides protection, despite the existence of specialized ant predators, since spi
ders respond much differently to disturbance (including attack by an ant predator)

June, 1997

Behavioral Ecology Symposium '96: Gushing

than the models. Ants, when disturbed, tend to respond aggressively to the threat,
whereas spiders tend to dodge the threat, hiding beneath a leaf or in a crevice, or
dropping on a drag line. It has been noted that spider myrmecomorphs, which are also
behavioral mimics, abandon their ant-like gait when disturbed (Emerton 1911, Mar
son 1947, Fowler 1984, Brignoli 1984). This sudden, unexpected change in the behav
ior of the spider would most likely facilitate its escape from an ant predator. Marson
(1947) points out that living in close proximity to their models (often in the midst of
foraging ants), as do many spider myrmecomorphs, also reduces the risk of predation,
even by ant predators, simply because the likelihood of an ant predator preying on a
less common mimic than one of the more common models is slim.
Important agents selecting for myrmecomorphy in spiders are probably spider
predators such as sphecid or pompilid wasps (Pocock 1908, Bristowe 1941, Edmunds
1978, Wanless 1978, Parker & Cloudsley-Thompson 1986). These predators might not
recognize myrmecomorphic spiders as potential prey. However, it has been reported
that some wasps, such as Trypoxylon placidum Cameron, Pison sp., and an unidenti
fled sphecid wasp, had myrmecomorphic spiders of the genus Myrmarachne in their
nest cells (Richards 1947, Edmunds 1978). However, these may be isolated instances
of individual wasps that have learned to differentiate Myrmarachne mimics from
their models. It is generally uncommon to find myrmecomorphic spiders in the nest
cells of spider hunting wasps (Bristowe 1941).
Indirect support for the hypothesis that myrmecomorphs are Batesian mimics lies
in the fact that, in general, myrmecomorphic spiders mimic either the dominant ants
in a habitat or aggressive, well protected ants (Edmunds 1978). Edmunds (1978) fur
their points out that transformational and polymorphic mimicry provide indirect sup
port for the hypothesis that myrmecomorphy in spiders evolved as an anti-predator
strategy. "Evidence for the strength of predator selection in perfecting the resem
balance between mimic and model is the infrequency of finding a Myrmarachne with
the 'wrong' species of ant, and the occurrence of different color morphs of mimic where
ever the model has a different colour" (Edmunds 1978).
Direct experimental studies in which arthropod predators have been presented
with choices between myrmecomorphic and non-mimetic prey also support the hy
pothesis that myrmecomorphs are Batesian mimics (Oliveira 1985, Mclver 1987,
Mclver 1989, and Cutler 1991). The results of these experimental studies are summa
rized in Mclver & Stonedahl (1993). In general, the predators avoid the myrmecomor
phs and the models while preying readily on the non-mimetic species, and they treat
the mimic as if it were an ant.


Table 2 presents information about known spider myrmecophiles. Included in this
table are those spiders that have either occasionally or exclusively been found in or
just outside ant nests. H11dobler & Wilson (1990), in their review of myrmecophiles,
included as myrmecophiles spiders that were specialized ant predators such as Ste
atoda fulva I'.. ., I .,-:) (Theridiidae) (H11dobler 1971), Euryopis cokiLevi (Theridi
idae) (Porter & Eastmond 1982), and Latrodectus hesperus Chamberlin & Ivie
(MacKay 1982). Although these spiders have evolved specialized hunting strategies
for capturing ants, they probably do not feed exclusively on ants and are only occa
sionally or never found inside ant nests. Therefore, they are omitted from Table 2. A
few other spider genera listed as myrmecophiles in H11dobler & Wilson (1990) are
more accurately described as myrmecomorphs and are included, instead, in Table 1.
The ants with which the myrmecophilic spiders are associated are also listed in
Table 2, as is information about the natural history of the spiders. Very little informa

Florida Entomologist 80(2)

tion is known about spider myrmecophiles. Only a very few studies have investigated
aspects of the spider-ant associations in any depth (Shepard & Gibson 1972, Noonan
1982, Porter 1985, Cushing 1995a, 1995b). Much more work must be done to deter
mine how the spiders become integrated into the host colonies, how the ants react to
these guests, what adaptations enable the spiders to live inside the nests, and to what
extent the spider affects the life of the host colony.

General Information about Myrmecophily

Many arthropods have evolved symbiotic relationships with ants. Some are found
at the periphery of the nest, either near the entrances or on refuse piles; others are
found within the chambers of the nest, either in the peripheral chambers or deeper in
the nest in the brood and storage chambers (H11dobler 1977, H11dobler & Wilson
1990). They range from tiny collembolans to beetles and caterpillars many times the
size of their hosts. These myrmecophiles have evolved various adaptations enabling
them to exist in this hostile environment. Many of the myrmecophiles acquire cutic
ular hydrocarbons similar or identical to those of their hosts (Vander Meer & Wojcik
1982, Vander Meer et al. 1989). This allows them to become integrated with hosts that
are otherwise hostile to intruders with foreign, non-colony odors. Others, such as
some staphylinid beetles and lycaenid caterpillars, have evolved specialized glands
that produce appeasement substances (reviewed in H11dobler & Wilson 1990).
In many myrmecophiles, the evolution of a symbiotic association can be intimated
through an examination of extant species that show varying degrees of behavioral in
tegration (H11dobler & Wilson 1990). For example, Akre & Rettenmeyer (1966) de
scribed species of staphylinid beetles that show varying degrees of association with
army ants. Some species live only around the edges of the bivouacs or in the refuse
piles but are not otherwise integrated into the colonies, others are found running
along the edges and sometimes within the emigration columns of ants, and yet others
are found directly in the midst of ants in the center of the emigration colonies. Some
species even hitch rides on the booty or the brood carried by ants. Certain staphylinid
species can only live within a narrow range of conditions found within colonies and die
shortly after removal from the colonies.
If each stage in this process of gradual integration into colonies is correlated with
the evolutionary history of the lineages, then the various adaptations of the myrme
cophiles leading to greater integration could be viewed as characters on the phyloge
netic tree (Brooks & McLennan 1991). Kistner (1979) takes this idea a step further by
superimposing the phylogenies of termites in the family Rhinotermitidae with their
associated termitophiles in the family Staphylinidae to illustrate the evolution of host

Adaptations of Myrmecophilic Spiders

Myrmecophilic spiders are unique because their close relatives apparently have no
preadaptations to a symbiotic lifestyle. Most spiders are solitary predators and sym
biosis with other arthropod groups should be rare; yet myremcophilic spiders are
found in at least 12 different families (Table 2). Some of these species may be only oc
casional visitors into ant colonies, using the entrance and upper chambers as tempo
rary refuges (see Table 2). However, some appear to be commensals that have become
more dependent on the conditions present within the nest and spend their entire lives
within this complex ecosystem.
Masoncus pogonophilus Cushing (Linyphiidae) is the best known example of the
latter group of spider myrmecophiles (Porter 1985, Cushing 1995a, 1995b). This spi

June, 1997

Behavioral Ecology Symposium '96: Gushing 183


~~o ooo oo st ^ .s
-C'Jt/ t
C- at
cn C c'J
at at ata at

2 cc ccg2^ S3 |S .
z s -5 -5 z l
-a a C a C a aQ

a3 23 ^- ^ CC

o a c^ C C
> 2 S

S ~1 c. vJ-
C tC U^
C ThC ~E C- E

l ~~v ,il |
ci 1 1C M ll-S

I s3a
2 CEa a c
-^ C C-C 3 h
E c of l o 3 c i^ Q Q

^ ~ o~ ~ -o 3 v m S ^u ^o c^ &
c. *C o oflCy -S
So -c QQ

o ;C-uc

=- e~LC
z o

7 -Y
C (O
2 El -/
cl o ao
I 0

C- ~ H ~- -
-c~~ S-'
H ~~~ C/l OC Om

Florida Entomologist 80(2)



at C
0 c

-2 C
C- C

r 3 0 "
Mr o ol


C .


-C- Cl 3-cC C
'?~8 a '

! *a 3" 'o-

A -e E i2 5
^C- Ofl C-
3 C -i V






3 O
C -





CCoa$ E

-co C-Q
-- Ce U
Qt CM~xC

C- 7,IC
a u,-E -
C-c- *,


- .:








June, 1997





C oa



Behavioral Ecology Symposium '96: Gushing 185


2 C
C- tf

Cc 0

-1 C o- CN

ct c o
:: :-m -
z 3
~ ~ C-c-

a, c-,


C- C- ,: Ymo ~ '

C C aM C- a,

cE nrl ~ oama

Li 11 C~~~n~'= M il Li~ UC-C- a
~C m C-C-
-1 3 a, E

0 03 0m

2 g,,Y~, m v m
Z m ~ ~ II1 C v ~ MC Q a-
-C- -= ~ b rr

w~ ~ ~E <2z m E
C-- ~o o~io
-c~s~ ,~~oJ
I ~ ~~~~~ m -C-CUoY I

Florida Entomologist 80(2)






3a a
i^ 3









I i^ i


-E T

2.ft ^
a a

I 1
; a;?
*S s
I P?-
z^ R^c

'e E

SU m

-c x -c-c

June, 1997




C -



- E

yi o

-,X C-I
.^ ^

Yo '*- -

c^ ^. m8
m1 EE m

/1 m

*: -c .ES!i i
iiiiit ^q
Q O iiQ^SL
mna, y) (j QiC )C





c u C

a ^
O iL

Behavioral Ecology Symposium '96: Gushing 187


3 Ct
U -
a, M t-
1 i
o N N
r" ^1 00 CT1 c'
C- C'Jo

.... So= 49
z 0- 2 &
7 -9
at C
-c Qt
SI I v, I |: a
on - a

< 6 .
on 0

0 ^ 0 3 6 3
oX a
ooC -o
G ^c : CCO %.QH^ ^ 0r
C"o cC ~ ~ Cc "; o3C
-C 3 cZ-c c-
3 2^ ^s 1^1 ] | |

o I- < C I c
C.C x z- a o a
s s s ^ ti a, >, -a,- na 'yt g a
- Cc. c-Cs so

H 2 0-.-- o- g ~ 5 'SS 3 o e Sic
7 - CC--

g3 (^ 3 E 'Z
M ~~ co o Oi -' i
A Ec o ^ CC
C- B -, a, C2: 0:.i 1

o g~- i m a, m'

i I, I, ^i | i i
Cr C i C I Pi t
aS ^l2 v i Is ~ e v l c
Cr CCCV^[3 0,
|S U

IS1 I 1

I o .2i m ,
a "

2 S>- II
EU ; ,

3~~--,- -,51 I ^
: *
-z C

7" K
00 53 r- 0

Ho on cn- =

Florida Entomologist 80(2)

der lives within the colony chambers of the Florida harvester ant, Pogonomyrmex ba
dius (Latreille). All life stages of M. pogonophilus are found inside the nests
throughout the year. The spiders feed on collembolans springtailss) found in the nest
chambers. When the host ants emigrate to a new nest site, the spiders (and collembo
lans) move with the ants along the emigration trails (Cushing 1995a, 1995b). There
is also evidence that spiders disperse between ant nests (Cushing 1995a, and in
prep.). The mechanism by which spiders locate new host colonies or become integrated
into new colonies is not yet known.

Adaptive Significance of Myrmecophily

Some myrmecophilic spiders may be considered Wasmannian mimics since they
are also myrmecomorphs (see Table 2). However, in Wasmannian mimicry, the model
itself (in this case, the host ant) is the selective agent (Rettenmeyer 1970). In other
words, the resemblance of the spider to the host ant must have been selected for by the
host itself and must facilitate the integration of the spider into the host colony. How
ever, very little is known about any of the myrmecomorphic myrmecophiles. Most of
them apparently spend at least some of their time outside the ant nests (see Table 2)
where they would be subject to predation by visually hunting predators in which case
their morphological resemblance to the host ants may simply be another example of
Batesian mimicry. The host ants may have little, if anything to do with their myrme
As H11dobler & Wilson (1990) propose, an ant colony can be considered an isolated
ecosystem. Arthropods that have evolved mechanisms for integrating themselves into
this specialized community are greeted with a stable microclimate, abundant food,
and protection from predators and parasites. Predation pressures, in particular, may
trigger greater integration into the ant societies in these myrmecophilic spiders since
association with the aggressive hosts may afford a high degree of protection to the
guests. Several of the myrmecophiles, such as Mastigusa arietina (Thorell) (Dic
tynidae), Eilica puno Platnick and Shadab (Gnaphosidae), Masoncus pogonophilus
(Linyphiidae), and Thyreosthenius biovatus O.P.-Cambridge (Linyphiidae) lay their
eggsacs inside the chambers of the host's nest (Donisthorpe 1908, 1927, Noonan 1982,
Porter 1985, Cushing 1995a, 1995b). Spiders are particularly vulnerable to eggsac
parasitism (Bristowe 1941). Eggsac parasitism or predation may also be a particu
larly important factor selecting for greater integration into ant colonies.


Detailed studies of myrmecomorphic spiders and their associated models can pro
vide insight into the ecological and evolutionary implications of mimicry. The hypoth
esis that myrmecomorphic spiders are Batesian mimics must be further tested
experimentally (Cutler 1991). It is important to use not only generalist predators in
such experiments, but also, if possible, spider predators as these may also be impor
tant selective agents for the evolution of more exact mimicry
The distribution of mimics and models, especially in transformational or polymor
phic mimetic complexes must be documented as Edmunds (1978) has done for Myr
marachne spp. in order to determine what effect community structure among the
model species has on the distribution and survival of color morphs in the mimic. It is
not known to what extent the geographic distribution of intraspecific polymorphic
mimics is dictated by genetic patterns or by differential predation of morphs in areas
with and without the appropriate model.

June, 1997

Behavioral Ecology Symposium '96: Gushing

A great deal more research must be done to uncover (literally) the basic natural
history of myrmecophilic spiders. For most spider myrmecophiles, it is unknown to
what extent they are obligate versus occasional guests in ant nests. For the obligate
guests, it must be determined how the spiders become integrated into the nests, how
they maintain the association, what part they play in the life of the colony, and what
part the colony plays in the survival of the guest. Noonan (1982) indicates that, for the
myrmecophile Eilica puno Platnick and Shadab (Gnaphosidae), the host ants protect
and tend the spider's eggsacs (see Table 2).
Certain families of spiders, such as the Linyphiidae and Liocrannidae, seem to
have more myrmecophilic representatives than others. It would be interesting to de
termine the phylogenetic relationship between myrmecophilic taxa and their free-liv
ing relatives. Are there certain preadaptations that make myrmecophily more likely
for certain lineages and less likely for others? It is important to document those
myrmecophilic spiders that may be encountered in the field. Studies of myrmecophilic
spiders can provide insight into the evolution of interspecific associations between so
cial hymenopterans and their guests.


Thanks to Lloyd R. Davis, Jr. for his help with the nomenclature of the ants and
for his helpful comments on an earlier draft of this manuscript. Thanks also to San
ford Porter and Robert Neumann for their comments on an earlier draft of this manu


AKRE, R. D., AND C. W. RETTENMEYER 1966. Behavior of Staphylinidae associated
with army ants (Formicidae: Ecitonini). J. Kansas Entomol. Soc. 39: 745-782.
BANKS, N. 1892. Mimicry in spiders. Proc. Entomol. Soc. 2: 174-176.
BANKS, N. 1895. A list of the spiders of Long Island, N.Y., with descriptions of new spe
cies. J. New York Entomol. Soc. 3: 7693.
BANKS, N. 1930. Ant-like spiders of the genus Myrmarachne from the Philippines.
Psyche 37: 207-218.
BERLAND, L. 1927. Sur une Araignee Myrm6comorphe de Nouvelle-Caledonie. Bull.
Soc. entomol. France 95: 52-55.
BHATTACHARYA, G. C. 1939. On the moulting and metamorphosis of Myrmarachne
plataleoides. Camb. Trans. Bose Res. Inst. 12: 103-114.
BOEVE, JL. 1992. Association of some spiders with ants. Revue suisse Zool. 99: 8185.
BOLTON, B. 1995. A New General Catalogue of the Ants of the World. Harvard Uni
versity Press, Cambridge, MA.
BRADOO, B. L. 1980. A new ant-like spider of the genus Myrmarachne (Salticidae)
from India. Current Science 49: 387-388.
BRIGNOLI, P. M. 1983. A Catalogue of the Araneae described between 1940 and 1981.
Manchester Univ. Press. 755 pp.
BRIGNOLI, P. M. 1984. Some remarks on myrmecomorph spiders. Newsl. British
Arachnol. Soc. 41: 34.
BRIGNOLI, P. M. 1986. Myrmecomorphism again. Newsl. British Arachnol. Soc. 47: 5
BRISTOWE, W. S. 1939. The Comity of Spiders, Vol. 1. Ray Society, London.
BRISTOWE, W. S. 1941. The Comity of Spiders, Vol. II. London: Ray Society 560 pp.
(pp. 414-460).
BROOKS, D. R., AND D. A. MCLENNAN. 1991. Phylogeny, Ecology, and Behavior. Uni
versity of Chicago Press, Chicago and London, 434 pp.
CHAMBERLIN, R. V 1925. A new clubionid spider of the genus Phruronellus from Cal
ifornia. Proc. Biol. Soc. Washington 38: 78.

Florida Entomologist 80(2)

COLLART, A. 1929a. Quelques observations sur une araignee mimetique. Revue Zool.
Bot. Afr. 18(2): 147-161.
COLLART, A. 1929b. Quelques notes sur les Myrmarachne araignees oecophylliformes.
Bull. Cerc. Zool. Congol. 5: 117-118.
COLLART, A. 1941. Notes Compl6mentaires sur Myrmarachne foenisex Simon
araignee myrm6comorphe du Congo Belge. Bull. Mus. R. Hist. Nat. Belg. 17: 1
CUSHING, P. E. 1995a. Natural History of the Myrmecophilic Spider, Masoncus
pogonophilus Cushing, and its Host Ant, Pogonomyrmex badius (Latreille),
Ph.D. dissertation, University of Florida, Gainesville.
CUSHING, P. E. 1995b. Description of the spider Masoncus pogonophilus (Araneae,
Linyphiidae), a harvester ant myrmecophile. J. Arachnol. 23: 5559.
CUTLER, B. 1970. A Taxonomic Revision of the Genus Consingis, Simon 1900. Ph.D.
dissertation, University of Minnesota, Minneapolis.
CUTLER, B. 1980. Synemosyna bicolor is the female of Synemosyna americana (Ara
neae, Salticidae). J. Arachnol. 9: 342-343.
CUTLER, B. 1981. A revision of the spider genus Paradamoetas (Araneae, Salticidae).
Bull. American Mus. Nat. Hist. 170: 207-215.
CUTLER, B. 1985. Taxonomic notes on neotropical species in the genus Synemosyna
(Araneae: Salticidae). Studies on Neotropical Fauna and Environment 20: 83
CUTLER, B. 1991. Reduced predation on the antlike jumping spider Synageles occiden
talis (Araneae: Salticidae). J. Insect Behav 4: 401-407.
DONDALE, C. D., AND J. H. REDNER 1972. A synonym proposed in Perimones, a syn
onym rejected in Walckenaera, and a new species described in Cochlembolus
(Araneida: Erigonidae). Canadian Entomol. 104: 1643-1647.
DONISTHORPE, H. 1908. A few notes on myrmecophilous spiders. The Zoologist 12:
DONISTHORPE, H. 1927. The Guests of British Ants, Their Habits and Life Histories.
London: Routledge and Sons. 244 pp. (pp. 184-201)
EDMUNDS, M. 1978. On the association between Myrmarachne spp. (Salticidae) and
ants. Bull. British Arachnol. Soc. 4: 149-160.
EIDMANN, H. 1937. Die Gaste und Gastverhaltnisse der Blattschneiderameise Atta
sexdens L. Z. f. Morph. u. Okol. d. Tiere 32: 391-462.
EMERTON, J. H. 1911. New spiders from New England. Trans. Connecticut Acad. Arts
and Sciences 16: 383-407.
ENGELHARDT, W. 1970. Gestalt und Lebensweise der "Ameisenspinne" Synageles ve
nator (Lucas) Zugleich ein Beitrag zur Ameisenmimikryforschung. Zoolog. Anz.
185: 317-334.
FAGE, L. 1938. Quelques Arachnides provenant de fourmilieres ou de termitieres du
Costa Rica. Bull. Mus. National d'Hist. Natur. 4: 369-376.
FOWLER, H. G. 1981. Behavior of two myrmecophiles of Paraguayan leaf cutting ants.
Rev. Chilena Ent. 11: 69-72.
FOWLER, H. G. 1984. Note on a clubionid spider associated with attine ants. J. Arach
nol. 12: 117-118.
GALIANO, M. E. 1965. Salticide (Araneae) formiciformes. IV. Revision del genero
Sarinda Peckham, 1892. Rev. Mus. Argentino Ciencias Naturales "Bernardino
Rivadavia" 1: 1312.
GALIANO, M. E. 1967. Salticidae (Araneae) formiciformes. VIII. Nuevas descripciones.
Physis 27: 2739.
GALIANO, M. E. 1969a. Salticidae (Araneae) formiciformes. VII. El genero Myrma
rachne MacLeay, 1939, en America. Rev. Mus. Argentino Ciencias Naturales
"Bernardino Rivadavia" 3: 107-148.
GALIANO, M. E. 1969b. Salticidae (Araneae) formiciformes. IX. Adicion a las revi
siones de los generous Martella y Sarinda. Physis 28: 247-255.
GALIANO, M. E. 1975. Salticidae (Araneae) formiciformes. XV Descripcion de Corcov
etella aemulatrix, genero y especie nuevos. Physis 34: 33-39.

June, 1997

Behavioral Ecology Symposium '96: Gushing

GRAY, B. 1971. Notes on the biology of the ant species Myrmecia dispar (Clark) (Hy
menoptera: Formicidae). Insect. Soc. 18: 71 80.
GREENSLADE, P. J. M., AND R. B. HALLIDAY. 1983. Colony dispersion and relationships
of meat ants Iridomyrmexpurpureus and allies in an arid locality in South Aus
tralia. Insect. Soc. 30: 82-99.
HARKNESS, R. D. 1977. Further observations on the relation between an ant, Catag
lyphis bicolor (F.) (Hym., Formicidae) and a spider, Zodarium frenatum (Simon)
(Araneae, Zodariidae), Entomol. Mon. Mag. 112: 111-121.
HINGSTON, R. W. G. 1927. Field observation on spider mimics. Proc. Zool. Soc. London
HOLLDOBLER, B. 1971. Steatoda fulva (Theridiidae), a spider that feeds on harvester
ants. Psyche 77(2): 202-208.
HOLLDOBLER, B. 1977. Communication in social Hymenoptera. In: How Animals
Communicate (ed. T A. Sebeok), pp. 418-471. Indiana University Press, Bloom
ington, IN.
HOLLDOBLER, B., AND E. O. WILSON. 1990. The Ants. Harvard University Press, Cam
bridge, MA, 732 pp.
JACKSON, J. F., AND B. A. DRUMMOND, III. 1974. A Batesian ant-mimicry complex
from the mountain pine ridge of British Honduras, with an example of trans
formational mimicry. American Midl. Nat. 91: 248-251.
JACKSON, R. R. 1986. The biology of ant-like jumping spiders (Araneae, Salticidae):
prey and predatory behaviour of Myrmarachne with particular attention to M.
lupata from Queensland. Zool. J. Linn. Soc. 88: 179-190.
KISTNER, D. H. 1979. Social and evolutionary significance of social insect symbionts.
In: Social Insects, vol. 1 (ed. H. R. Hermann), pp. 339-413. Academic Press, NY.
KOMATSU, T. 1961. Notes on spiders and ants. Acta Arachnol. 17(2): 25-27.
KROMBEIN, K. V. 1967. Trap-Nesting Wasps and Bees: Life Histories, Nests, and As
sociates. Smithsonian Press, Washington, D.C., 570 pp.
LEVI, H. W. 1986. Ant-mimicking orb weavers of the genus Ildibaha. In Proceedings
of the Ninth International Congress of Arachnology, Panama 1983, ed. W G.
Eberhard, Y. D. Lubin, B. C. Robinson, pp. 159-162. Washington, D.C.: Smith
sonian Institution Press. 333 pp.
LIGHTON, J. R. B., AND R. G. GILLESPIE. 1989. The energetic of mimicry: the cost of
pedestrian transport in a formicine ant and its mimic, a clubionid spider. Phys
iol. Entomol. 14: 173-177.
MACKAY, W. P. 1982. The effect of predation of western widow spiders (Araneae: The
ridiidae) on harvester ants (Hymenoptera: Formicidae). Oecologia 53(3): 406-411.
MARSON, J. E. 1946. Observations on the ant-mimicking spider Myrmarachneplata
leoides Cambr. in India. Entomol. Mon. Mag. 82: 52-53.
MARSON, J. E. 1947. The ant mimic Myrmarachne plataleoides. J. East Africa Nat.
Hist. Soc. Natl. Mus. 19: 62-63.
MATHEW, A. P. 1934. The life-history of the spider Myrmarachneplataleoides. J. Bom
bay Nat. Hist. Soc. 37: 369-374.
MATHEW, A. P. 1935. Transformational deceptive resemblance as seen in the life his
tory of the plant bug Riptortus pedestris, and of a mantis Evantissa pulchra. J.
Bombay Nat. Hist. Soc. 37: 803-813.
MATTHEW, A. P. 1954. Observations on the habits of the two spider mimics of the red
ant, Oecophylla smaragdina (Fabr). J. Bombay Nat. Hist. Soc. 52: 249-263.
MCIVER, J. D. 1987. On the myrmecomorph Coquillettia insignis Uhler (Hemiptera:
Miridae): arthropod predators as operators in an ant-mimetic system. Zool. J.
Linn. Soc. 90: 133-144.
MCIVER, J. D. 1989. Protective resemblance in a community of lupine arthropods. Na
tional Geogr. Res. 5: 191-204.
MCIVER, J. D., AND G. STONEDAHL. 1993. Myrmecomorphy: morphological and behav
ioral mimicry of ants. Ann. Rev. Entomol. 38: 351-379.
MELLOLEITAO, C. F. 1925. Dois interessantes arachnideos myrmecophilos. Physis 8:
228 237

Florida Entomologist 80(2)

MELLOLEITAO, C. F. 1926. Algumas aranhas do Brasil Meridional. Boletim do Museu
Nacional 2: 1-18.
MUKERJEE, D. 1930. Description of a new ant-mimicking spider, Synemosyna trans
versa sp. nov. J. Bombay Nat. Hist. Soc. 34: 200-201.
MYERS, J. G., AND G. SALT. 1926. The phenomenon of myrmecoidy, with new examples
from Cuba. Trans. Royal Entomol. Soc. London 74: 427-437.
NOONAN, G. R. 1982. Notes on interactions between the spider Eilica puno (Gna
phosidae) and the ant Camponotus inca in the Peruvian Andes. Biotropica 14:
OLIVEIRA, P. S. 1985. On the mimetic association between nymphs of Hyalymenus
spp. (Hemiptera: Alydidae) and ants. Zool. J. Linn. Soc. 83: 371-384.
OLIVEIRA, P. S. 1986. Ant-mimicry in some spiders from Brazil. Bull. Soc. Zool. France
111: 297-311.
OLIVEIRA, P. S. 1988. Ant-mimicry in some Brazilian salticid and clubionid spiders
(Araneae: Salticidae, Clubionidae). Biol. J. Linn. Soc. 33: 1-15.
OLIVEIRA, P. S., AND I. SAZIMA. 1984. The adaptive bases of ant-mimicry in a neotro
pical aphantochilid spider (Araneae: Aphantochilidae). Biol. J. Linn. Soc. 22:
OLIVEIRA, P. S., AND I. SAZIMA. 1985. Ant-hunting behaviour in spiders with emphasis
on Strophius nigricans (Thomisidae). Bull. British Arachnol. Soc. 6: 309-312.
PARKER, J. R. 1984. Spiders associated with ants. Newsl. British Arachnol. Soc. 41: 4.
PARKER, J. R., AND J. L. CLOUDSLEY-THOMPSON. 1986. Myrmecophilous spiders.
Newsl. British Arachnol. Soc. 45: 24.
PIZA, S. DE T. 1937. Novas esp6cies de aranhas myrmecomorphas do Brasil e consid
erac6 sobre o seu mimetismo. Revista do Museu Paulista 23: 307-319.
PLATNICK, N. I. 1993. Advances in Spider Taxonomy 1988 1991, with Synonymies and
Transfers 1940-1980. New York Entomological Society in association with The
American Museum of Natural History, New York, 846 pp.
POCOCK, R. I. 1908. Mimicry in spiders. J. Linn. Soc. (Zool.) 30: 256-270.
PORTER, S. D. 1985. Masoncus spider: a miniature predator of Collembola in har
vester ant colonies. Psyche 92: 145-150.
PORTER, S. D., AND D. A. EASTMOND. 1982. Euryopis coki (Theridiidae), a spider that
preys on Pogonomyrmex ants. J. Arachnol. 10(3): 275-277.
REISKIND, J. 1965. The taxonomic problem of sexual dimorphism in spiders and a syn
onymy in Myrmecotypus (Araneae, Clubionidae). Psyche 72: 279-281.
REISKIND, J. 1969. The spider subfamily Castianeirinae of North and Central America
(Araneae, Clubionidae). Bull. Mus. Comp. Zool. 138(5): 163-325.
REISKIND, J. 1970. Multiple mimetic forms in an ant-mimicking clubionid spider. Sci
ence 169: 587-588.
REISKIND, J. 1972. Morphological adaptation for ant-mimicry in spiders. Fifth Inter
national Congress of Arachnology (Brno, 1971): 221 226.
REISKIND, J. 1977. Ant-mimicry in Panamanian clubionid and salticid spiders (Ara
neae: Clubionidae, Salticidae). Biotropica 9: 18.
REISKIND, J., AND H. W. LEVI. 1967. Anatea, an ant-mimicking theridiid spider from
New Caledonia (Araneae: Theridiidae). Psyche 74: 2023.
RETTENMEYER, C. W. 1970. Insect mimicry. Ann. Rev. Entomol. 10: 43-74.
RICHARDS, O. W. 1947. Observations on Trypoxylon placidum Cam. (Hym., Sphe
coidea). Entomol. Mon. Mag. 83: 53.
ROEWER, C. F. 1935. Zwei Myrmecophile Spinnen-Arten Brasiliens. Ver6ffentlichun
gen Deut. Kolon. Ubersee Mus. 1: 193-197.
ROEWER, C. F. 1942. Opiliones, Pedipalpi und Araneae von Fernando Poo. Ver6ff.
Deutsch. Kolon. Uebersee-Mus. Bremen 3(3): 244-258.
SHELFORD, R. 1902. Observations on some mimetic insects and spiders from Borneo
and Singapore. Proc. Zool. Soc. London. 2: 230-284.
SHEPARD, M., AND F. GIBSON. 1972. Spider-ant symbiosis: Continusa spp. (Araneida:
Salticidae) and Tapinoma melanocephalum (Hymenoptera: Formicidae). Cana
dian Entomol. 104: 1951 1954.

June, 1997

Behavioral Ecology Symposium '96: Gushing

VANDER MEER, R. K., AND D. P. WOJCIK. 1982. Chemical mimicry in the myrmecophil
ous beetle, Myrmecaphodius excavaticollis. Science 218: 806-808.
VANDER MEER, R. K., D. P. JOUVENAZ, AND D. P. WOJCIK. 1989. Chemical mimicry in
a parasitoid (Hymenoptera: Eucharitidae) of fire ants (Hymenoptera: Formi
cidae). J. Chem. Ecol. 15: 2247-2261.
VANE-WRIGHT, R. I. 1980. On the definition of mimicry Biol. J. Linn. Soc. 13:1-6.
WALLER, D. A., AND J. C. MOSER 1990. Invertebrate enemies and nest associates of
the leaf cutting ant Atta texana (Buckley) (Formicidae, Attini). In: Applied
Myrmecology: a World perspective (ed. R.K. Vander Meer, K. Jaffe, and A.
Cedeno), pp. 255-273. Westview Press, Boulder, CO.
WALSH, J. H. 1892. On certain spiders which mimic ants. J. Asiat. Soc. Bengal 60: 104.
WALTER, E. V., L. SEATON, AND A. A. MATHEWSON. 1938. The Texas leaf cutting ant
and its control. USDA Circular No. 494: 1-18.
WANLESS, F. R. 1978. A revision of the spider genera Belippo and Myrmarachne (Ara
neae: Salticidae) in the Ethiopian region. Bull. British Mus. (Nat. Hist.) 33(1):
WASMANN, E. 1894. Kritisches Verzeichniss der myrmecophilin und termitophilen Ar
thropoden. Felix Dames, Berlin xi + 231 pp.
WICKLER, W. 1968. Mimicry in Plants and Animals. World University Library,
McGraw-Hill Book Co., New York.
WING, K. 1983. Tutelina similis (Araneae: Salticidae): an ant mimic that feeds on
ants. J. Kansas Entomol. Soc. 56: 5558.


Behavioral Ecology Symposium '96: Van Hook


Department of Entomology and Nematology
Box 110620
University of Florida
Gainesville, FL 32611


Large, conspicuously colored insect taxa, due to associated logistical and anthro
pocentric biases in knowledge, public support and legislative consideration, are fa
vored as targets of species protection, environmental monitors and education tools.
They are also vulnerable to collection and perhaps, due to ecological specializations
associated with apparency, to extinction. I discuss the implications for conservation.

Key Words: insect conservation, insect coloration, insect apparency, insect conserve
tion policy


La taxa de insects grandes y visiblemente coloreados, por estar asociados a pre
juicios en el conocimiento antropoc6ntrico y logistico, apoyo pfiblico y consideraci6n le
gal, son favorecidos como blanco por grupos dedicados a la protecci6n de species, a
monitorear el medio ambiente y a utilizarlos como herramienta educativa. Estos in
sectos tambi6n son vulnerable a la colecci6n y quizas, debido a especializaciones eco

Florida Entomologist 80(2)

16gicas asociadas con la apariencia, tambien son vulnerable a la extinci6n. Se
discuten las implicaciones en t6rminos de conservaci6n.

Insects dominate terrestrial ecosystems in terms of species, biomass, number of in
dividuals, and importance of ecological roles (Ricklefs et al. 1984, Wilson 1987, 1988).
Approximately 80% of all described metazoan species are insects (Samways 1992). In
sect global distribution is highly biased, with over 50% living on less than 7% of the
earth's surface in tropical rain forests (Samways 1994). It has been estimated that
only about 5% of insect species have been described, and significant information is
thought to exist on less than 1% of these (Raven 1990). Almost all of that information
has been garnered from either pest or charismatic species in temperate areas; neither
are representative samples of insect biological diversity.
Insects are susceptible to the same anthropogenic threats as vertebrates. Wilson
(1988) estimates that species extinctions are occurring at a rate of at least 1000 times
faster than before human-induced extinction pressures. However, most insect popular
tion declines and extinctions go unnoticed or unappreciated. This is largely a result of
apparency-related obstacles. Small size and inconspicuous habits, together with tre
mendous diversity and a mostly tropical distribution, make insects largely invisible to
human attention and concern.
Humans most readily learn about, care about, and make sacrifices for animals
that are apparent, familiar, aesthetically appealing, and demonstrate positive bene
fits to mankind. Such glamorous species often enjoy special privileges in species-ori
ented conservation efforts, due to research, funding, and political and public support.
On the other hand, they may be particularly susceptible to anthropogenic impacts, di
rectly through their status as commodities and indirectly through special ecological
needs associated with their apparency. Familiar and appealing insects also offer spe
cial advantages for larger-than-species-scale conservation efforts. Existing historical
and ecological knowledge, sampling methodologies, expert interest, and financial sup
port enhance the use of glamorous species as indicators of biological diversity and as
monitors of environmental change.
Animal apparency and aesthetic appeal are related to their coloration, morphol
ogy, behavior, and size. Strikingly colored, shaped, or behaviorally interesting crea
tures benefit from heightened conservation attention. However, with regard to
insects, size is a limiting factor. Glamor status occurs only when a species is large
enough 1) to be well-studied and sampled, and 2) to overcome anthropocentric biases
that favor larger animals. Such threshold sizes are atypical of insects; by far, most in
sects are minute or small. In North America, described insect species vary from less
than 1 mm to approximately 15 cm, with more than half being less than 6 mm long
(Borror & White 1970). Of approximately 430 species of adult beetles collected by fog
going in four tropical forests in Brazil, 97% were less than 8 mm in length (Erwin
1983). Although speciose groups, including Coleoptera, Lepidoptera and Diptera, con
tain some of the largest, most well-studied and charismatic species, flamboyance is
not typical even within these taxa.
I will briefly discuss the roles of the environment, life history strategies, sexual se
election and predator defense in shaping adaptations associated with apparency and
beauty in insects. I will then consider how these adaptations influence 1) the way we
value species, 2) anthropogenic threats to their populations, and 3) how we treat spe
cies in conservation policy and management.

June, 1997

Behavioral Ecology Symposium '96: Van Hook 195


Sources of variation in apparency and aesthetic appeal in insects include adapta
tions associated with the environment, particular life history strategies, predator de
fense, sexual selection, and the interplay of these. Population and species differences
in size, shape, and coloration can reflect variation in temperature, humidity, and day
length. Differing ecotypes contain species with characteristic arrays of adaptations
that influence diversity at all levels of biological organization. For example, tundra ec
system thermal and daylight constraints and structural simplicity limit within and
among species variation. Few species are naturally rare. Such environments and their
relatively uneventful insect species are seldom the focus of conservation efforts. Only
under larger-scale efforts that focus on preservation of representative ecotypes are
such landscapes and species given significant conservation attention. On the other
hand, tropical areas are generally characterized by seasonally stable, structurally
complex environments, and strong biotic selective pressures. Intra and interspecific
diversity is extreme in size, coloration, and associated behaviors, habitat specialize
tion, and distribution patterns. Such environments receive high conservation priority.
Ironically, these areas represent our greatest conservation challenges specifically be
cause such extreme variation eludes current conservation strategies.
Phenotypic variation can occur through the season or through a species' range and
can be genetic and/or environmentally controlled. The nature of this control is impor
tant in conservation theory and practice. Color polymorphisms can be sources of bio
logical diversity (Samways 1994), sources of confusion in monitoring populations
(Crother 1992), and focuses of interest for researchers and collectors. Differing color
forms can be associated with differing macro or micro-habitat preferences affecting
within-species vulnerabilities to anthropogenic changes to the environment. For ex
ample, the peppered moth (Biston betularia) demonstrates how genetic polymor
phisms in color, form and associated resting surface preferences can lead to color
form-specific responses to human-induced changes to the environment.
Where biological variation over space is relatively gradual, conservation efforts
are often directed at range extremes. These areas are often marked by apparency-re
lated characters. Variation in insect color, morphology, behavior and size are also used
to map areas of abrupt environmental change, increased variability within and
among species, and speciation hot spots.
Complex life cycles are common in insects, especially speciose taxa with large,
readily apparent species such as Coleoptera, Lepidoptera, and Diptera. Variation in
appearance and ecological specialization associated with different life stages can be
extreme and influence both our concern and our ability to conserve species. Extreme
differences in life stage forms can lead to conflicts of interest, taxonomic confusion and
a need for increased research efforts.
Insect size and coloration are integral parts of life history strategies that are asso
ciated with factors such as mobility, longevity, degree of habitat specialization, activ
ity periods, flight patterns, predator avoidance, and feeding strategies. For example,
reproductive rate is related to insect size. Some of the largest and most dramatically
colored insects, such as the birdwings (Ornithoptera, Papilionidae), are considered K
selected species. Because such long-lived species produce relatively few eggs at a low
rate, they can be especially vulnerable to extinction. Furthermore, large, conspicuous
species often live in closed or sedentary populations that are thought to be especially
threatened by habitat fragmentation (Thomas 1984, Thomas & Mallorie 1985).
Large, conspicuous species, due to their associated ecological specializations, may
exist as groups of local sub-populations. Such metapopulations are potentially buff

Florida Entomologist 80(2)

ered from extinction within an area by re-colonization from surrounding sub-popula
tions. Such populations may by particularly vulnerable to habitat destruction and
require regional approaches to their conservation (Murphy et al. 1990). Metapopula
tions are thought to be common in insects in general and differ from those used to de
velop vertebrate models used in population viability assessments and recovery plans
(Murphy et al. 1990).
Insect apparency is increased when aggregations are formed in association with
roosting, feeding, mating, predator defense, moderation of the environment, or over
wintering. Such aggregations increase the apparency of the group to potential preda
tors and to collectors. Insect behavior and aggregation site characteristics may
further increase exposure. For example, swarming or hilltopping mating aggregations
often occur in exposed locales and individuals remain in flight almost constantly. Al
though long-term aggregations associated with warning coloration may function to
deter predation (Vulinec 1990), they may be especially vulnerable to collection due to
the increased accessibility and the economic value associated with aposematism.
When aggregations involve a large portion of the population for an extended time,
their members become especially vulnerable to very specific habitat threats.
At least in tropical butterflies, aggregations are associated with other life-history
strategies, such as restricted home ranges, low reproductive rates, and increased lon
gevity (Turner 1975), that magnify their vulnerability to human-induced threats to
their habitat. Congregated organisms also bias our perception and measurement of
rarity, thereby favoring aggregated species and populations in conservation priority
Predation pressures not only impact conservation efforts through the evolution of
aggregation behavior, but as selection agents in the evolution of warning coloration,
concealment, crypsis, and mimicry. Depending on the adaptation and the particular
threat faced, these adaptations may benefit or harm the conservation of a species. For
example, insects that rely on concealment or blend into their background are often
small and drab. While such adaptations may lessen potential anthropogenic threats
such as insect collection, inconspicuous species are not likely to be well-studied and
their conservation is unlikely to gain public support. Because their inconspicuousness
is dependent on specific associations with a component of the insect's environment,
these insects can be especially vulnerable to changes in their habitat. Cryptic species
that mimic a particular object such as a leaf, due to their often exposed resting and
feeding habits and their associated low population densities, may be particularly vul
nerable to climate change, pollution, and pesticides. Furthermore, once discovered by
collectors, cryptic species may become economically valued due to their aesthetic in
trigue. For the same reason, they may also gain research favor and public support.
Chemically protected aposematic insects are thought to have evolved bright, bold
coloration, often involving black, yellow, orange or red, as an advertisement of their un
palatability to visual predators such as birds and lizards. Aposematic insects are
among the most-valued aesthetically, the most-studied, and the most-exploited in trade
largely because human visual systems are also attuned to such colors and patterns.
Such species often attain their chemical protection by sequestering secondary sub
stances produced by plants as adaptations against phytophagous insects. Plant and
insect counter-adaptations lead to very specific and dependent relationships. Such co
adaptations increase insect vulnerability due to the indirect impacts of disruptions to
their associated plant ecology. Insect-plant associations can restrict geographical and
ecological tolerance increasing susceptibility to climate change (Samways 1994). Ba
tesian and Mullerian mimicry complexes are commonly associated with warningly
colored insects. These complexes can be confusing taxonomically Because of scientific
interest in using mimicry complexes as tools to understand ecology and evolution, the

June, 1997

Behavioral Ecology Symposium '96: Van Hook

associated species are often some of the best understood ecologically and genetically.
Their apparency also enhances the practicality of their study
Sexual selection pressures account for many adaptations that we associate with
beauty in insects, and why we often prize males as commodities. Secondary sexual
characters that function in intrasexual interactions and/or mate choice include size,
bright coloration, ornamentation, weaponry and behaviors. In terms of intraspecific
visual signaling coloration, size, and shape can be viewed as a compromise between
the needs to attain mates and avoid being eaten. For example, the bright and bold up
per wings of many male butterflies are exposed in flight and used in intraspecific sig
naling while the under wings, exposed when resting and feeding, are cryptically
colored to avoid predation. Such visual compromises are common in some of our most
glamorous butterfly species. They are sources of intraspecific genetic diversity and are
of great interest to researchers and collectors.
Sexual dimorphisms in coloration in butterflies are a common result from the in
terplay of sexual selection and predation pressures. Females are relatively inconspic
uous while the brightly colored males risk increased predation for increased
reproductive success. Because males generally show the most dramatic coloration,
and the taking of males is generally thought to have no effect on future population
numbers, potential threats by collectors or predators can be lessened or negated by
such adaptations. However, in female-limited mimicry complexes in which females,
but not males, mimic other brightly colored, chemically protected species, it may be
that females stand out and are particularly sought after.


Economic, ethical, ecological, educational, and practical values attributed to par
ticular species are used to prioritize conservation efforts (for reviews see IUCN 1983,
Morris et al. 1991, New 1995, Pyle et al. 1981, Samways 1994). Insect apparency and
beauty impact how we value insects in all of these areas.
Insects are generally little valued economically (IUCN 1983). Small size and the
associated lack of information contribute to this situation (Samways 1994). While
some species are used as sources of products, medicines, and biological control, only a
very few taxa are commercially valuable because of their apparency These are typi
cally traded as dead stock. Worldwide, dead stock trade of insects is valued at tens of
millions of dollars annually (Morris et al. 1991). Single specimens of birdwings have
been advertised for up to US $7000 (Morris et al. 1991). An unexplored way that
aposematic insects might be commercially valuable is through their co-evolved chem
ical systems with plants. Because insect conspicuousness can serve as a flag for novel
plant chemistry, such insects might be used as probes to survey for medicines or other
useful bio-chemicals (Kremen et al. 1993).
Ethical arguments for insect conservation that are based on intrinsic values of in
dividuals are philosophical extrapolations of human-based morality (Lockwood 1987).
Such an ethical framework cannot favor individuals of undescribed species nor indi
viduals of endangered versus common species (Samways 1994). Individual-based
moral consideration is effectively apparency-biased because only insects that are ob
vious to humans gain support on such grounds (New 1995). Furthermore, aestheti
cally-valued taxa are more likely to gain support, especially when economic, cultural,
ecological or practical values conflict with moral consideration (Samways 1990). Eth
ical considerations are practically without application in undeveloped countries and
are often incongruent with community or landscape level conservation strategies.
However, by focusing on the interdependence of the genome and the individual, con

Florida Entomologist 80(2)

flicts between ethical and biological justifications of insect conservation are often rec
onciled (Samways 1994). The above arguments do not negate the importance of moral
consideration of individuals of all species. Such consideration is commendably the ba
sis of insect collection ethical codes.
Insects are by far most valued in conservation for their ecological roles. They are
key components in the composition, structure and function of ecosystems (Hafernik
1992, Ricklefs et al. 1984, Wilson 1987). Insects are abundant herbivores and detriti
vores influencing directly and indirectly elemental cycling and net primary productive
ity (Seastedt & Crossley 1984). Ecological importance and beauty only rarely coincide.
The human bias in favor of the apparent and beautiful may be particularly short
sighted in this regard.
Charismatic species can be successfully used in the communication of issues,
needs, knowledge, and the benefits of insect conservation (Salwasser 1991). Environ
mental educational objectives in which glamorous insect species are particularly use
ful include the study of diversity, abundance and biomass, complexity, species
radiation, history, biological and economic importance, and interaction with plants
(Robinson 1991, in New 1995). Conservation studies that demonstrate population de
lines of glamorous species, especially butterflies, have increased the general public's
awareness of the need to protect insects and their habitats (Hafernik 1992, Samways
1989, Thomas 1984).
Amateur involvement in conservation efforts, such as the Fourth of July Butterfly
Count (Swengel 1990) and the Entomological Society of Victoria Butterfly Mapping
Scheme (New 1990(92)) are generally limited to well-appreciated, large, easily as
sessed taxa. Zoo and museum displays use almost exclusively large, apparent, and at
tractive species. Butterfly gardening has become a popular hobby and is promoted as
a means to effectively demonstrate important ecological principles using mostly large,
attractive species. It is common for naive butterfly gardeners to want to discourage
unattractive larval stages that devastate their store-bought plants. However, the as
sociation of the less conspicuous, and usually less attractive, larvae with the appreci
ated, sought after adults teaches the need for the less-than-beautiful and the need to
provide habitat. Experienced gardeners usually come to appreciate larval forms and
Large, conspicuous insects offer unparalleled opportunities for conservation-re
lated research. Unlike their vertebrate counterparts, they are accessible, easily
reared, short lived, diverse, and inexpensive to study Theoretical studies of apparent
species provide important models for developing conservation methodology and set
ting conservation priorities that are unique to invertebrates (e.g. Hanski & Thomas
1994, Ehrlich & Murphy 1987, Murphy et al. 1990, Murphy & Weiss 1988).


Public support for conservation continues to rest on emotional rather than intel
lectual motives, and has been garnered primarily by the cute and cuddly vertebrates.
Most adults dislike or are afraid of arthropods. This reflects our biased awareness of
almost exclusively injurious insects (Byrne et al. 1984, Kellert 1993). Modern agricul
ture, and the usual resulting information bias toward small, unattractive, harmful
pests, is largely responsible for such negative public perceptions (Barnes 1985). In
nate fears may also contribute to human biases against insect conservation, espe
cially when species are inconspicuous, unattractive, and economically unimportant
(Kellert 1993). Such fears may be especially well-ingrained by certain aposematic in
sects. On the other hand, it is also the bright, big and bold insects, especially beneficial

June, 1997

Behavioral Ecology Symposium '96: Van Hook

ones, that can be used most effectively to overcome ignorance, prejudice, innate fears,
and anthropocentric biases against the small and often ugly world of insects (Morris
1987, Samways 1992).
Insects face the same anthropogenic threats as vertebrates, including changes to
their habitat, impacts by exotics, pollution, climate change, pesticides, and, poten
tially, their collection for profit. The most important threat is habitat loss, fragmenta
tion, and/or degradation. Unlike that seen in vertebrates, there is no general positive
relation between insect size and their vulnerability to extinction (Samways 1994).
This suggests that, although large, conspicuous species may sometimes face increased
vulnerabilities due to their associated ecological needs, the conservation focus on
large, conspicuous species is not biologically sound in general.
Climate change potentially affects insects both directly and indirectly through
plant associations (Dennis 1993, New 1995). Apparency-related aspects of butterfly
biology have led to their use as models for understanding the direct impacts of atmo
spheric pollutants and for predicting the indirect effects of climate change. For the
same reasons, butterflies are promoted as monitors of climate change (Dennis 1993).
Pesticides have been blamed for insect species extinctions, but there have been no
documented cases of such extinctions (IUCN 1983, Pyle et al. 1981, Thomas 1984).
This is not to say that pesticides have not or can not lead to insect extinctions under
certain circumstances. Furthermore, pesticides may influence insect community
structure by changing the distribution and relative abundance of species (Samways
After habitat destruction, the negative impacts of non-indigenous species is con
sidered the greatest threat to insect conservation. Including all known animal taxa,
by far, most documented non-indigenous species in the US are accidentally introduced
insects. Their impact is assumed to come primarily through interspecific competition
and increased predation pressures (US Congress, Office of Technology Assessment
1993). However, the impacts of these non-indigenous species are rarely documented,
except for economically important or charismatic species, because insects are gener
ally unapparent, unappreciated, and, therefore, neglected in conservation.
Size also influences our knowledge of the environmental risks posed by biological
control organisms. For example, microorganisms are thought to offer the greatest po
tential in biological control. However, due to their great diversity, minute size, and in
S li.I we know almost nothing about their biology and ecology (Pimentel
1980). Classical biological control agents, such as nematodes, fungi, protozoa, bacte
ria, and viruses may have host ranges beyond their targeted species (Pimentel 1980,
Samways 1988). However, their potential impact is rarely studied. When impacts are
assessed, they are judged by aesthetically pleasing or economically valued species.
For example, Bacillus thuringiensis has been shown to negatively affect more than
135 non-target species (Laird 1978, in Pimentel 1980), but it has generally received
positive reviews because it has not been documented to be harmful to the natural en
emies of economically important pest species. Bacillus thuringiensis israelensis, used
in mosquito control, has received conservation attention because it has been shown to
cause mortality in mayfly and dragonfly larvae (Zgomba, Petrovic & Srdic 1986, in
Samways 1994). The general lack of conflict of interest between insect conservation
and classical biological control lies partly in the fact that biological control is most of
ten aimed at small, inconspicuous, unpopular, exotic species, while conservation ef
forts are aimed at large, conspicuous, popular, rare, and, often, specialized species
(Samways 1988). The general absence of public demand for more strict pre and post
release assessments of imported exotic biological control agents is related to the fact
that obvious, charismatic species have rarely been noticeably impacted. Bacillus thu
ringiensis, released for gypsy moth control, may raise public concern if butterflies,

Florida Entomologist 80(2)

even non-indigenous species, are found to be negatively impacted as suspected (Har
brecht 1991).
Insect collection for trade, commodity production and research is biased toward
large, apparent species due to their aesthetic value and practical advantages. Taxa
that are valued by collectors may benefit through the associated increased knowledge
that is necessary for most species-oriented conservation efforts. Live trade of insects
is highly biased toward large, aesthetically pleasing species but these are often bred
from very few wild-caught animals. Butterfly farming and ranching are considered vi
able sustainable use strategies in which very high demand species are reared or en
courage to breed by providing them with host plants in their natural environment.
A subset of these are then collected and used for economic gain, while the remaining
are left to maintain or even boost natural populations.
The potential impact of collectors on insect populations remains a hotly debated
topic, especially among lepidopterists. For recent controversial opinions, see The News
of the Lepidopterists' Society (38 (1-2) 1996). The consensus appears to be that collec
tors rarely, if ever, are the primary cause of insect population or species extinctions
(IUCN 1983, Morris 1987, New 1995, Orsak 1978(81), Pyle et al. 1981, Samways
1994, Thomas 1984). However, the scientific study of the impact of collection on vul
nerable species is lacking. Insect collection is considered an ethical issue, but only spe
cialist trade of wild-caught specimens, where value is heightened by rarity, is
considered potentially threatening to populations. These rare species are often K-se
elected. Low reproductive rates, limited ranges and very specific host plant associa
tions can increase vulnerability to collection and the habitat destruction that can be
associated with economic gain. Parnassius apollo and New Guinea birdwings are ex
amples of K-selected insects that are apparently threatened by collecting (Pyle


Insect conservation policy primarily addresses the protection of rare species, with
provisions for those species' habitats, and/or general restrictions on insect collection.
Such policies are extensions of vertebrate-based conservation philosophies and are
generally not objective nor consistent (New 1995). This is partly due to logistical con
straints related to the small size and inconspicuousness of most insect taxa. It is eas
ier to assess population status, develop management plans, and monitor large,
conspicuous species. Their conservation need is more likely to be demonstrated by
pre-existing data necessary to document population decline. Their study is more
likely to secure funding and public support.
Just as with vertebrates, charismatic insect species are sometimes intentionally
given conservation priority for political reasons. In Britain, the Swallowtail, Papilio
machaon, was included in the Wildlife and Countryside Act (1981) as a political ploy.
Its inclusion was based on glamour status and historical focus and was contrary to sci
entific data that indicated a low priority in conservation need (Morris 1987). Further
more, charismatic species that are not considered a high conservation priority may be
listed because their preservation is expected to serve as an umbrella for other species.
Such an umbrella can be quite effective. Habitat protection for the El Segundo blue,
Euphilotes bernardino allyni, has helped to protect 15 other less-glamorous inverte
brate species that co-inhabit the preserved California sand dune ecosystem (Mattoni
The US Endangered Species Act (ESA) of 1973 is considered the most powerful
conservation policy in the world. Although the ESA theoretically gives equal status to
all species, in practice charismatic species are strongly favored. Fewer than 10% of

June, 1997

Behavioral Ecology Symposium '96: Van Hook

listed species received more than 90% of the funding in 1990, and none of these is an
insect (New 1994)! ALL insects receive little attention relative to their representative
ness in species diversity or their ecological importance. Small size, lack of aesthetic
appeal, and associated lack of knowledge, support, and funding further bias listing ef-
forts within the Insecta (Boecklen 1987, Hafernik 1992, Murphy 1991, Van Hook
1994). As of 1989, 95% of the 427 insect species assessed for listing were not listed due
to insufficient information (Opler 1991). There have been 28 insects put on the list of
endangered and threatened wildlife, 19 of which are butterflies. Recovery plans exist
for only four species, all of which are butterflies (Opler 1995).
The ESA is an example of the wrong approach at the wrong scale (sensu Murphy
1989). The policy is criticized for lack of scientific bases, ineffectiveness and inconsis
tent use (Mann & Plummer 1992, Murphy 1991, Noss 1991, Rohlf 1990, Salwasser
1991, Scott et al. 1987, Tangley 1984, Wilcove 1992). All of these err in disfavor of in
sect conservation, and especially the less charismatic taxa. For example, vague terms
like endangered and threatened have no consistent biological meaning. What consti
tutes a species is debatable, especially in plants and invertebrates. The use of such
vague terminology creates both intended and unintended biases in conservation ef-
forts and apparent species are often favored (Rohlf 1991). Below-species-level knowl
edge and conservation consideration are very rare in insects and restricted to
glamorous taxa (Wilcove et al. 1992).
ESA biases in species listing that are related to apparency and appeal include 1)
information is related to charisma, 2) species that are less charismatic are slower to
be listed, even when data is available, 3) once listed, more attention and funding are
directed toward charismatic species, and 4) small, inconspicuous species are difficult
to survey (Tangley 1984). Once filtered through these biases, listed species are those
thought to be especially threatened by anthropogenic impacts. Ecological specialize
tions associated with, but not limited to, large, apparent species can increase these
vulnerabilities (Murphy 1991).
Apparency-related biases in listing are also characteristic of state agency insect
conservation policies. For example, the Technical Advisory Committee on Endangered
Species for the Florida Committee on Rare and Endangered Plants and Animals is
constrained in its efforts to identify rare species and develop recovery plans by appar
ency-related problems. These include small size, diversity of types, seasonality of
form, lack of information and taxonomic problems (Weems 1977). Listed species are
not necessarily the most worthy. They reflect the interests of taxonomic specialists
and amateurs who provide the historical knowledge base needed to demonstrate pop
ulation declines.
International conservation policies also generally favor charismatic species. For
example, the criteria for nomination for the listing on the Berne convention (the Con
servation of European Wildlife and Natural Habitats, 1979) includes a provision that
the species must be easy to identify. Minute, inconspicuous insects, have undeveloped
taxonomies, even in relatively well-studied areas like Europe, preventing listing of
the major chunk of insect biological diversity. The International Union for the Conser
vation of Nature and Natural Resources (IUCN) Red Data Book is intentionally bi
ased toward some glamorous groups, like butterflies and dragonflies, due to their high
profile related to size, coloration, ease of identifying, and taxonomist specialization
(Samways 1994). The aim is for these taxa to serve are umbrella species for the lesser
endowed, less conspicuous species (Pyle 1978(81)). This bias is exemplified in the Swe
den Red Data Preliminary List (1987) which includes 786 species, of which over 300
are Coleoptera and over 250 are Lepidoptera. This predominance reflects the high di
versity of these groups, but also reflects their relatively greater number of large, con
spicuous species compared to other groups. The Convention of International Trade in

Florida Entomologist 80(2)

Endangered Species of Wild Fauna and Flora (CITES) of 1973, an international agree
ment aimed at protecting rare species from economic abuses through trade, lists 10
insect species. All of these are lepidopterans (New 1995). The Bonn Convention on the
conservation of migratory species of wild animals (1979) lists only the charismatic
monarch butterfly.
As noted above, overcollection is rarely considered to threaten insect populations.
Furthermore, there is no evidence that restrictions on collection benefit insect popu
lation numbers (Hama et al. 1989, in Sibatani 1990(92)). Inconsistent with this
knowledge, most insect conservation policy consists only of restrictions on collectors
or insect trade (Pyle et al. 1981). When broader-based policies exist, collection and
trade restrictions are usually retained (e.g. ESA). This is a carry over of vertebrate
based evidence that population declines result from overexploitation. We need scien
tific consensus on if, when, where, and how collection impacts insect populations if we
are to develop more appropriate insect conservation policy.
Biases in our perception and appreciation of insects contribute to the problems of
policy restrictions on insect collection. These include 1) broad restrictions are often
without biological rationale and may unduly restrict amateur interest, 2) restrictions
are often biased in enforcement, 3) policy often does not reflect species need, but aes
thetic appeal and the associated higher levels of knowledge, 4) bureaucratic, and en
forcement costs may compete with habitat protection, 5) insect surveys necessary to
document population declines are severely restricted, and 6) restrictions can increase
exploitation when the perceived rarity is related to value (New 1995).
All insect collection is prohibited without a permit in most protected areas in
many, especially developed, countries. These restrictions are meant to serve as um
brella protection measures, but such policies lack scientific bases and unduly inhibit
the gathering of information and the development of amateur interest (Morris 1987,
New 1990(92), Samways 1994, Sibatani 1990(92), and Thomas 1984). For example, in
an effort to protect one species, the satyrid (Erebia christi), in some areas of Switzer
land it is illegal to carry a butterfly net (New 1995). At the other extreme, The Indian
Wildlife Protection Act lists approximately 450 butterfly taxa as protected and prohib
its specifically the collection of these species (New 1995). The identification problems
associated with small size and inter and intraspecific phenotypic variation make
such policies practically self defeating.
All-inclusive collection restrictions are rarely enforced due to the necessary costs
and bureaucracy. However, recently, the US Fish and Wildlife Service has brought
several collectors to court over the taking of insect specimens without a permit on pro
tected lands. Sporadic, inconsistent enforcement is biased in time and in space, is re
stricted primarily to charismatic taxa such as butterflies, and has estranged amateur
collectors (for recent accounts of this controversy see News of the Lepidopterists' Soci
ety38 (1-2) 1996).
When collection restrictions are less than all-inclusive, they are focused toward
charismatic taxa. Such restrictions assume collection can negatively impact insect
populations in general and then use collector interest to direct restrictions. This is a
conservative approach that results from a lack of information. Under the British
Wildlife countryside act of 1981, it is illegal to kill, take, or sell 14 insects (Drewett
1988), all of which are relatively conspicuous species. The Federal Republic of Ger
many prohibits collection of large lepidopterans (Morris 1987). In some European
countries all butterflies are protected from collection. Interestingly, the pestiferous
white pierids are exempted from such restrictions (Collins 1987, in New 1995).
In Germany, all Odonata are protected from collection, while the impact of acid
rain in their conservation is largely ignored (Samways 1994). In Japan, protection leg
isolation is limited almost exclusively to collection prohibitions for butterflies, thought

June, 1997

Behavioral Ecology Symposium '96: Van Hook

to be of little or no benefit, while preservation of insect habitats is ignored (Sibatani
Many conservation, amateur, and scientific organizations have published volun
tary insect collection codes. These often cover all species, but are aimed at glamorous,
not necessarily rare, species. These restrictions are based on ethical rather than bio
logical grounds. They rightly discourage collection of very rare species, over-collection
of any species, and wasteful collection methods.


Practical problems with both species and larger-scale approaches to conservation
that are related to apparency and appreciation of insects include 1) the paucity and
complexity of taxonomic and ecological knowledge, 2) monitoring problems, and 3) bi
ases in research, funding, amateur interest, and public support (New 1995).
There is approximately one taxonomist for every 425 described insect species
(Samways 1994). This ratio creates a taxonomic impediment that becomes even more
daunting when we consider that fewer than 5% of existing insect species are thought
to be described (Raven 1990). The taxonomic limitations arising from the practical dif
ficulties of observing and studying very small organisms is so great that microorgan
isms must be classified functionally rather than morphologically (Chapin et al. 1992).
Funding and expertise interest are biased toward aesthetically appealing and eco
nomically important species, and both are most lacking in undeveloped, tropical areas
where insect species diversity is highest.
Apparency differences associated with life stage, microhabitat, sex, and season are
not appreciated by traditional taxonomic methods but are critical to ecologically-ori
ented conservation efforts (Samways 1994). Phenotypic variation, such as color poly
morphisms, cryptic species, and sibling species, also confuse species-status
determination. It is difficult to accurately assess the population status, develop man
agement plans, or monitor such ambiguous groups. The use of dead specimens further
compounds the problem of taxonomic designation. For example, the satyrid butterfly,
Oeneis bore, has two color forms that behave as separate species in the field but is
treated as one species using phenotypic techniques that rely on dead specimens (Fer
ris 1986). Naturalists studying live animals in their natural habitats and molecular
systematic methods are necessary to overcome some of the shortcomings of tradi
tional taxonomic methods. Both practical and theoretical problems with species sta
tus designation have not been adequately confronted in species-level conservation
approaches. Community and landscape-level conservation strategies overcome some
of these taxonomic-related problems but these approaches also rely heavily on species
The extreme diversity, small size, inconspicuous habits, and the taxonomic and
ecological ignorance associated with these aspects of insect biology prevent species
by-species inventorying. New (1995) suggests three strategies aimed at getting
around this problem: 1) the use of indicator groups, 2) taxonomic reduction, and 3) the
use of ecologically functional groups. Taxonomic reduction includes grouping by
higher than species level taxa and grouping by morphological characters or recognize
able taxonomic units. Both taxonomic reduction and the use of functional groups rely
on apparency-related adaptations to alleviate other apparency-related obstacles in in
sect conservation. For example, size, coloration, and morphological structures related
to feeding strategies are used to group species with similar ecological function.
The assessment of potential impacts of climate change, pesticides, non-indigenous
species, and collection on insect populations is primarily restricted to aesthetically
pleasing and economically important species. This reflects interest, knowledge and

Florida Entomologist 80(2)

monitoring methodologies that are beauty or necessity-biased. The potential impacts
of pesticides are little-studied for any non-target species. To date, monitoring the im
pacts of non-indigenous species, including biological control agents, is almost exclu
sively restricted to economically important or glamorous species (Ehler 1991). It is
expected that further studies will confirm the environmental safety of most classical
biological control agents (Samways 1988). However, the potential environmental dan
gers of releasing irretrievable, mobile, evolving organisms and our paucity of data on
the impacts of these non-indigenous species on their new environment are forming a
barricade to the development of this important pest-control strategy (Samways 1994).
Most attempts to document potential negative impacts of collecting on insect popular
tions come from studies on attractive taxa, almost exclusively butterflies. This is ap
propriate since they are particularly sought after, but the documented impacts or lack
thereof may not be representative of such a diverse group as the Insecta.

Problems Associated with Single Species Approaches to Insect Conservation

Species listing and the development of recovery plans are very demanding in
terms of both historical and ecological information and financial and public support.
We can afford these costs only for relatively glamorous species and only in relatively
wealthy nations. For example, in the IUCN Invertebrate Red Data Book (1983), all ex
amples of anthropogenic impacts on insects resulting from changes in land, water,
pollution, loss of associated species, and importation of exotics were documented for
large, apparent species in developed countries. Only water-pollution impacts were
noted for inconspicuous species, probably reflecting a long history of using inverte
brates as environmental indicators of water quality (Kremen et al. 1993).
Under the ESA, conservation priorities are based on biological uniqueness, degree
of threat, and opportunity for success (Mann & Plummer 1992). Each of these is
highly biased in favor of apparent and appreciated species for practical and emotional
reasons. Most conservation efforts have been aimed at butterflies because they are ob
vious, enjoy high amateur interest, are easy to see and study, and are both harmless
to humans and beneficial as pollinators. These aspects of butterfly biology make de
termination of uniqueness and threat easier to identify and also incite public support
necessary to monitor and manage insect populations. In contrast, inconspicuous and
unattractive parasites are generally ignored in conservation, even though they are
considered extremely diverse and of conservation concern due to their generally ex
treme, obligatory specializations (Windsor 1995). Parasites are often only discovered
when their hosts become extremely rare or extinct, and then they are often dismissed
or even attacked in an effort to boost their host's survival (Windsor 1995). The demise
of the Passenger Pigeon stands as one of the most exemplary, best-appreciated species
losses. The simultaneous loss of its lice parasite has gone unnoticed and without con
cern (Stork & Lyal 1993, in Windsor 1995).
Species-oriented management plans are restricted almost exclusively to butter
flies. The European Large Copper butterfly, Lycaena dispar batava, has been aug
mented since the 1930s, and this effort is expected to remain necessary for its
continued survival in the wild (New 1995). Such costly efforts are not feasible for even
the most well-studied and well-appreciated species in developed countries. They are
likely to be counter-productive in understudied, speciose areas such as the tropics.
As with vertebrates, intensive management efforts, such as captive breeding,
translocation and reintroduction programs, are initiated when species are at the
brink of extinction with little chance of recovery These risky and unpredictable tactics
are costly in terms of funding, time, expertise, and research. They are restricted to
charismatic or economically important species in developed countries. In Europe, 323

June, 1997

Behavioral Ecology Symposium '96: Van Hook

insect reintroductions or reinforcements have been attempted, with less than 60% es
tablished. All of these efforts were directed at butterflies (Oates & Warren 1990, in
Samways 1994).
Recent releases of captive-bred Schaus Swallowtails (Papilio aristodemus pon-
ceanus), an endangered subspecies under the ESA, demonstrate the sort of practical
considerations that insect apparency forces on intensive management efforts. Re
searchers had to change from releasing cryptic pupae to adults because 65-99% of the
pupae were lost to predation when placed in their natural habitat. Even in this rela
tively well-studied species, it is unknown how these rates compare to natural levels of
predation, but it is thought that unnatural densities, positioning, and artificial pupa
tion bases used in the releases may have voided the larvae's crypsis (Jaret Daniels,
pers. comm.).
Variation in insect form and function related to seasonality, polymorphic types,
sexual differences, and life stage specializations is an obstacle in species-oriented con
servation strategies (Samways 1994). Such variation adds both confusion and time to
the listing process and complexity to management plans, with apparent forms being
favored. For example, critical habitat protection under the ESA includes areas outside
the geographic area typically occupied, including hilltopping, hibernation, and aesti
vation areas. These are more likely to be known, and their preservation supported, for
charismatic species.

Insect Apparency Biases and Implications for Large-Scale Conservation Strategies

It is not feasible or biologically rational to appraise insects species-by-species for
conservation needs, due to their extreme diversity in species and ecological roles, and
habitat requirements. More and more, single-species approaches are combined with
ecosystem approaches to conservation. Larger-scale (than species) approaches rely on
reducing the volume and complexity of information necessary to preserve and manage
species and natural areas through innovative methods of assessment, management,
and monitoring (Hunter 1991). These approaches rest on empirical knowledge, ecolog
ical theories, and model development that are in their infancy (Salwasser 1991).
Large, brightly colored insects are most likely to contribute to each of these. They are
also more likely to enjoy funding priority and expert attention.
Large-scale approaches in conservation relieve the need to prioritize conservation
efforts by values associated with charisma. However, biases toward large, conspicuous
species are retained for monitoring and assessing conservation sites. Five types of
species are of paramount importance in ecosystem approaches to conservation. These
include 1) species used as indicators of diversity or monitors of environmental change,
2) keystone species: those that play a critical role in the structure and function of an
ecosystem, 3) umbrella species: those whose conservation serves to protect other spe
cies, 4) flagship species used as a focus for funding and generating support, and 5) spe
cies that are particularly vulnerable to extinction due to their biology and/or ecology
(Noss 1991). The discovery and use of each of these types of species are apparency-bi
ased due to disproportional levels of information and the prevailing anthropocentric
conservation perspective.
Insects are increasingly used as indicators of biogeographic zones, areas of ende
mism, community richness, diversity, naturalness, typicalness, and centers of evolu
tionary radiation in conservation planning (see Kremen 1992, Kremen et al. 1993,
and references therein). Favored groups are readily observed and collected, are well
known taxonomically and ecologically, and are valued aesthetically and/or economic
cally (Kremen et al. 1993). These biases are intended and often necessary. They may
or may not be biologically legitimate. For example, dipteran and hymenopteran para

Florida Entomologist 80(2)

sitoids are potentially good indicator species, due to their association with diverse eco
logical niches and microhabitats, widespread occurrence and correlated trends with
other groups. However, Disney (1986b, in New 1995) showed that mapping the distri
butions of Diptera is limited due to practical shortcomings associated with their ap
parency We must use the distribution and abundance of obvious, easily sampled
To better assess representativeness, uniqueness, and typicalness of areas in order
to set conservation priorities, we need to develop more efficient sampling methods
(New 1995). The use of parataxonomists or amateurs to help with the tremendous
amount of sorting and identification necessary for conservation-related work is be
coming increasing popular (but see Rosenberg et al. 1986). Such innovative ap
preaches are necessary and effective, but accuracy is generally sacrificed for efficiency
Furthermore, the loss of accuracy is not consistent across taxa, but is apparency bi
ased. For example, in a study of the performance of non-specialists in assessing sam
ples of aquatic insects, Cranston and Hillman (1992) showed that increased
variability was correlated with small body, increased number of closely related taxa,
and morphological variability within species.
In management, insects are used to monitor human disturbance and ecological
change, including changes in habitat, ecological disruption, climate change, and pol
lution. Insects are sometimes favored as monitors over vertebrates because they are
particularly sensitive, respond rapidly, and offer a smaller-scale probe (e. g. Kremen
et al. 1993, New 1995, Sparrow et al. 1994, Thomas & Mallorie 1985). They are also
increasingly used to supplement vertebrate monitoring because of their unique habi
tat needs and responses to anthropogenic threats. Useful groups must be ecologically
specialized and, due to the need for reproducible sampling methods and historical in
formation, they are generally large, and apparent. Butterflies are preferred as indica
tor species and monitors of environmental change specifically because of their
apparency and charisma (Samways 1994). They are specialized, well known taxonom
ically and ecologically, have established monitoring methods, and strong amateur in
terest and public support (Kremen 1992, Kremen et al. 1993, Thomas 1991).
Umbrella species are notable taxa that are characteristic of a particular habitat
that, when preserved, benefit many unstudied, unappreciated species in the commu
nity The value of a particular species to serve as a protective umbrella is based on eco
logical requirements, such as the need for large diverse habitats. However, the need
for historical and ecological information, as well as public support, favors the desig
nation of apparent and appealing species for this role (New 1994). The monarch but
terfly is an example of an umbrella species. The focus of research, conservation
attention, and public support for monarch conservation enhances the potential to pre
serve the remaining flora and fauna of the highly fragmented, isolated fir forest relics
that constitute their threatened overwintering grounds in the highlands of Mexico.


Large, conspicuously colored insect taxa are given special attention in species-ori
ented conservation. This focus is both legitimate and intended. It is based on special
threats and ecological needs associated with a species' apparency and on conservation
values, public support and policy aims. However, the apparency bias is also a some
times unintended, and sometimes unnoticed, bias, resulting from practical aspects of
insect ecology and conservation methodology
Empirical and theoretical contributions by entomologists are needed to improve
existing species-focused conservation efforts, to better develop larger scale ap
preaches, and to help build conservation policies that better reflect the unique conser

June, 1997

Behavioral Ecology Symposium '96: Van Hook

vation needs of insects. Particularly important is the need to develop generally
applicable population models using representative insects, to develop better sampling
methods, and to better integrate conservation and agriculture programs.
In species-oriented conservation efforts, ALL insects are relatively small and in
conspicuous, and are highly disfavored in conservation efforts relative to their verte
brate co-inhabitants. This reflects the lack of ecological and taxonomic knowledge,
research, funding, public and policy support, and sampling problems. These impedi
ments point to the value of increasing large-scale conservation research, education,
and policy directives. The use of insects as tools for assessing, managing and monitor
ing landscapes promotes ecosystem and regional approaches that are critical to all fu
ture conservation efforts. Large-scale conservation strategies also rely on both
intended and unintended biases toward large, conspicuous insects. Entomologists can
help to identify and lessen detrimental biases and document strengths through theo
retical and empirical contributions.
The relatively recent focus on insects as targets and tools in conservation points to
the need to broaden the discipline of entomology and to better bridge our work with
amateurs, ecologists and conservation biologists. The study of pest and glamour spe
cies has much to offer conservationists. However, to achieve the broader goals of sus
tainability in agriculture and conservation, entomologists need to discard our own
biases. We need to better address the 99% of species not generally considered in pest
oriented research (Wilson 1987). We cannot afford to cut our funding or attention to
the development of innovative, ecologically sensible solutions to pest problems. How
ever, we can no longer ignore the fact that sustainable agriculture rests on functioning
natural ecosystems both near and far from the agricultural fields. These natural sys
tems are insect-dominated, but by neither characteristically beautiful or pestiferous
species. Their study is critical both to the future sustainability of agriculture and to
agriculture's contribution to the conservation of biological diversity.
Effective and efficient conservation strategies cannot depend solely on public sup
port for charismatic species that are emotionally valued. Although we can, and must,
learn affinities for species that are unfamiliar to and different from ourselves, we
most readily learn about, care about, and make sacrifices for species that are appar
ent, aesthetically appealing and demonstrate positive human benefits. Exposure to
glamorous insects will help bridge the gap between our natural human affinity for the
cute and cuddly and the needed appreciation of often non-intuitive ecological princi
ples. Until we cross that bridge, we will continue to make irresponsible personal and
social decisions. Increasing the ecological awareness of policy makers and the general
public is the most important and timely conservation challenge. Entomologists study
the most diverse, ubiquitous and, arguably, most important taxa in conservation. We
are the best equipped to rid negative biases against insects and to instill an appreci
ation of the importance of insects to our sustainable future.


I would like to thank John Sivinski for organizing this year's Behavioral Ecology
Symposium and for giving me the opportunity to learn more about insect conservation.
John Sivinski, Thomas C. Emmel and Jaret Daniels provided constructive criticisms on
this manuscript. James Castner, Skip Choate, Jaret Daniels, Peter Kovarik, James
Lloyd, Paul Skelley, and Mike Thomas kindly offered their assistance, advice and the
use of their slides which were crucial to both my talk and this paper. And finally, thanks
to David Kirschke and Sweetpea Van Hook for their patience and reminders that,
seemingly contrary to my above thesis, the world is more than a bunch of bugs. Univer
sity of Florida, Institute of Food & Agricultural Sciences, journal series no. R-05632.

Florida Entomologist 80(2)


BARNES, J. K. 1985. Insects in the New Nation: A cultural context for the emergence
of American entomology. Bull. Entomol. Soc. America 31(1): 21 30.
BOECKLEN, W. J. 1987. Conservation of insects: mass extinction, scientific interest,
and statutory protection, p. 40-75 in S. B. Vinson and R. L. Metcalk [eds.], En
tomology Serving Society: Emerging Technology and Challenges. Entomologi
cal Society of America, Lanham, MD.
BORROR, D. J., AND R. E. WHITE. 1970. A Field Guide to the Insects of American North
of Mexico. The Peterson Field Guide Series. Houghton Mifflin Co., Boston.
BYRNE, D. N., E. H. CARPENTER, E. M. THOMS, AND S. T. COTTY. 1984. Public attitudes
toward urban arthropods. Bull. Entomol. Soc. America 30(2): 40-44.
CHAPIN, F. S., III, E-D. SCHULZE, AND H. A. MOONEY. 1992. Biodiversity and ecosys
tem processes. Trends Ecol. & Evol. 7(4): 107-108.
CRANSTON, P. S., AND T. HILLMAN. 1992. Rapid assessment of biodiversity using 'Bi
logical Diversity Technicians'. Australian Biol. 5: 144-154.
CROTHER, B. I. 1992. Genetic characters, species concepts, and conservation biology.
Conserv Biol. 6(3): 314.
DENNIS, R. L. H. 1993. Butterflies and Climate Change. Manchester University Press,
DREWETT, J. 1988. Never mind the whale, save the insects. New Scientist 120: 32-35.
EHLER, L. E. 1991. Planned introductions in biological control p. 21-39 in L. R. Gin
zburg [eds.], Assessing Ecological Risks of Biotechnology. Butterworth-Heine
mann, Boston.
EHRLICH, P. R., AND D. MURPHY. 1987. Conservation lessons from long-term studies
of checkerspot butterflies. Conserv. Biol. 1(2): 122-131.
ERWIN, T. L. 1983. Tropical forest canopies: the last biotic frontier. Bull. Entomol. Soc.
America 29: 14-19.
FERRIS, C. D. 1986. Presidential address, 1986: Unexplored horizons -the role of the
amateur lepidopterist. J. Lepidop. Soc. 40(4): 247-254.
HAFERNIK, J. E., JR 1992. Threats to invertebrate biodiversity: Implications for con
servation strategies, p. 171-195 in P. L. Fiedler and S. K. Jain [eds.], Conserva
tion Biology: The Theory and Practice of Nature Conservation, Preservation
and Management. Chapman and Hall, NY.
HANSKI, I., AND C. D. THOMAS. 1994. Metapopulation dynamics and conservation: a
spatially explicit model applied to butterflies. Biol. Conserv. 68: 167-180.
HARBRECHT, D. 1991. Rescuing rare beauties. National Wildlife, 29(5): 4-9.
HUNTER, M. L., JR. 1991. Coping with ignorance: The course-filter strategy for main
training biodiversity, p. 266-281 in K. A. Kolm [ed.], Balancing on the Brink of
Extinction: The Endangered Species Act and lessons for the future. Island
Press, Washington, DC.
SOURCES). 1983. The IUCN Invertebrate Red Data Book. IUCN, Gland, Swit
KELLERT, S. R. 1993. Values and perceptions of invertebrates. Conserv. Biol. 7(4): 845
KREMEN, C. 1992. Assessing the indicator properties of species assemblages for natu
ral areas monitoring. Ecol. Appl. 2(2): 203-217.
JAYAN. 1993. Terrestrial arthropod assemblages: Their use in conservation
planning. Conserv Biol. 7(4): 796-808.
LEDERHOUSE, R. C. 1990. Avoiding the hunt: Primary defenses of Lepidoptera cater
pillars, p. 175-189 in D. L. Evans and J. O. Schmidt [eds.], Insect Defenses:
Adaptive Mechanisms and Strategies of Prey and Predators. State University
of New York, Albany.
LOCKWOOD, J. A. 1987. The moral standing of insects and the ethics of extinction.
Florida Entomol. 70: 70-89

June, 1997

Behavioral Ecology Symposium '96: Van Hook

MANN, C. C., AND M. L. PLUMMER 1992. The butterfly problem. The Atlantic Monthly
269(1): 47-70.
MATTONI, R. H. T. 1992. The endangered El Segundo blue butterfly J. Res. Lepid. 29:
MORRIS, M. G. 1987. Changing attitudes to nature conservation: the entomological
perspective. Biol. J. Linn. Soc. 32: 213-23.
MORRIS, M. G., N. M. COLLINS, R. I. VANE-WRIGHT, AND J. WAAGE. 1991. The utiliza
tion and value of non-domesticated insects, p. 319-347 in N. M. Collins and J.
A. Thomas [eds.], The Conservation of Insects and their Habitats. Academic
Press, London.
MURPHY, D. D. 1989. Conservation and confusion: Wrong species, wrong scale, wrong
conclusions. Cons. Biol. 3(1): 82-84.
MURPHY, D. D. 1991. Invertebrate conservation, p. 181-198 in K. A. Kohm [ed.], Bal
ancing on the Brink of Extinction: The Endangered Species Act and Lessons for
the Future. Island Press, Washington DC.
MURPHY, D. D., K. E. FREAS, AND S. B. WEISS. 1990. An environment-metapopulation
approach to population viability analysis for a threatened invertebrate. Con
serv Biol. 4(1): 41-51.
MURPHY, D. D., AND S. B. WEISS. 1988. Ecological studies and the conservation of the
Bay Checkerspot butterfly, Euphydryas editha bayensis. Biol. Conserv. 46: 183
NEW, T. R. 1990(92). Conservation of butterflies in Australia. J. Res. Lepid. 29(4): 237
NEW, T. R. 1995. Introduction to Invertebrate Conservation Biology. Oxford Univer
sity Press, Oxford.
NEW, T. R., AND N. M. COLLINS. 1991. Swallowtail Butterflies. An action plan for their
conservation. IUCN, Gland.
Noss, R. F. 1991. From endangered species to biodiversity, p. 227-245 in E. K. A.
Kohm [ed.], Balancing on the Brink of Extinction: The Endangered Species Act
and Lessons for the Future. Island Press, Washington DC.
OPLER, P. A. 1991. North American problems and perspectives in insect conservation,
p. 9-32 in N. M. Collins and J. A. Thomas [eds.], The Conservation of Insects and
their Habitats. Academic Press, London.
OPLER, P. A. 1995. Conservation and management of butterfly diversity in North
America, p. 316-324 in A. S. Pullin, in association with the British Butterfly
Conservation Society [eds.], Ecology and Conservation of Butterflies. Chapman
and Hall, London.
ORSAK, L. J. 1978(81). Endangered insects of the world proceedings: Introduction to
the proceedings and an update on terrestrial arthropod conservation. Atala
6(1-2): 1-18.
PIMENTEL, D. 1980. Environmental risks associated with biological controls. Ecol.
Bull. 31: 11-24.
PYLE, R. M. 1978(81). International problems in insect conservation. Atala 6(1-2): 56-58.
PYLE, R., M. BENTZIEN, AND P. OPLER 1981. Insect conservation. Ann. Rev. Entomol.
26: 233-259.
RAVEN, P. H. 1990. The politics of preserving biodiversity. BioScience 40(10): 769-774.
RICKLEFS, R. E., Z. NAVEH, AND R. E. TURNER 1984. Conservation of ecological pro
cesses. Commission on Ecology Papers Number 8. Environmentalist, 4(Supple
ment No. 8): 3-16.
ROHLF, D. J. 1990. Six biological reasons why the endangered species act doesn't work
-and what to do about it. Conserv. Biol. 5(3): 273-282.
ROJAS, M. 1992. The species problem and conservation: What are we protecting? Con
serv Biol. 6(2): 170-178.
ROSENBERG, D. M., H. V. DANKS, AND D. M. LEHMKUHL. 1986. Importance of insects
in environmental impact assessment. Environ. Manage. 10(6): 773-783.
SALWASSER, H. 1991. In search of an ecosystem approach to endangered species con
servation, p. 247-265 in K. A. Kohm [ed.], Balancing on the Brink of Extinction:

Florida Entomologist 80(2)

The Endangered Species Act and Lessons for the Future. Island Press, Wash
ington, DC.
SAMWAYS, M. J. 1988. Classical biological control and insect conservation: Are they
compatible? Environ. Conserv. 15: 349-354.
SAMWAYS, M. J. 1990. Insect conservation ethics. Environ. Conserv 17(1): 7-8.
SAMWAYS, M. J. 1992. Some comparative insect conservation issues of north temper
ate, tropical, and south temperate landscapes. Agric., Ecosyst. & Environ. 40:
SAMWAYS, M. J. 1994. Insect Conservation Biology. Chapman and Hall, NY.
SCOTT, M. J., B. CSUTI, K. SMITH, J. E. ESTES, AND S. CAICCO. 1987. Beyond Endan
gered Species: An integrated conservation strategy for the preservation of bio
logical diversity. Endangered Species Update 5(10): 43-48.
SEASTEDT, T. R., AND D. A. CROSSLEY, JR. 1984. The influence of arthropods on eco
systems. BioScience 34(3): 157-160.
SIBATANI, A. 1990(92). Decline and conservation of butterflies in Japan. J. Res. Lepid.
29(4): 305-315.
SPARROW, H. R., T. D. SISK, P. R. EHRLICH, AND D. D. MURPHY. 1994. Techniques and
guidelines for monitoring neotropical butterflies. Conserv. Biol. 8(3): 800-809.
SWENGEL, A. B. 1990. Monitoring butterfly populations using the Fourth of July but
terfly count. American Midl. Nat. 124: 395-406.
TANGLEY, L. 1984. Protecting the "Insignificant". BioScience 34(7): 406-409.
THOMAS, C. D., AND H. C. MALLORIE. 1985. Rarity, species richness and conservation:
Butterflies of the Atlas Mountains in Morocco. Biol. Conserv. 33: 95-117.
THOMAS, J. A. 1984. The conservation of butterflies in temperate countries: past ef
forts and lessons for the future, p. 333-353 in R. I. Vane-Wright and P. R. Ackery
[eds.], The Biology of Butterflies. Princeton University Press, Princeton, NJ.
THOMAS, J. A. 1991. Rare species conservation: case studies of European butterflies,
p. 149 197 in I. F. Spellerberg, F B. Goldsmith, and M. G. Morris [eds.], The Sci
entific Management of Temperate Communities for Conservation. Blackwells,
TURNER, J. R. G. 1975. Communal roosting in relation to warning color in two helico
niine butterflies (Nymphalidae). J. Lepidop. Soc. 29: 221-226.
Species in the United States. No. OTA-F-565, U.S. Government Printing Office,
Washington DC.
VAN HOOK, T. 1994. The conservation challenge in agriculture and the role of ento
mologists. Florida Entomol. 77(1): 42-73.
VULINEC, K. 1990. Collective security: Aggregation by insects as a defense. p. 251-288
in D. L. Evans and J. O. Schmidt [eds.], Insect Defenses: Adaptive Mechanisms
and Strategies of Prey and Predators. State University of New York Press, Al
WEEMS, H. V., JR. 1977. The Florida endangered insect program. Florida Entomol.
60(1): 57-63.
WILCOVE, D. S., M. MCMILLAN, AND K. C. WINSTON. 1993. What exactly is an endan
gered species? An analysis of the U.S. Endangered Species List: 1985-1991.
Conserv Biol. 7(1): 87-93.
WILSON, E. O. 1987. The little things that run the world: The importance and conser
vation of invertebrates. Conserv. Biol. 1(4): 344-346.
WILSON, E. O. 1988. The current state of biological diversity, p. 3-18 in E. O. Wilson
[ed.], Biodiversity National Academy Press, Washington DC.
WINDSOR, D. A. 1995. Equal rights for parasites. Conserv Biol. 9(1): 1-2.

June, 1997

Castineiras et al.: Distribution ofNeoseiulus cucumeris 211


University of Florida, IFAS, Tropical Research and Education Center
18905 SW 280th Street
Homestead, FL 33031


The distribution of the predacious mite Neoseiulus cucumeris (Oudemans) and its
prey, Thrips palmi Karny, was studied in eggplant plots in Homestead, Florida. Neo
seiulus cucumeris was more abundant on fruits (X = 3.39 0.20) than on leaves (X =
0.95 0.16) and it was not found in the flowers. Thrips palmi was more abundant on
the leaves (X = 17.97 5.07) than on the fruits (X = 3.22 0.70) and flowers (X = 0.93
0.03). Predacious mite populations on the fruits and leaves increased with T palmi
populations increase. Both predator and prey populations were low on the youngest
leaf (Xp,,t = 0.00 + 0.00; X y = 1.75 0.28) and high on the oldest leaf (Xp,_ = 1.92
+ 0.79; X, = 50.83 + 11.64). Neoseiulus cucumeris and T.palmiwere more abundant
on the adaxial surface of the leaf (X ,__ =- 1.58 0.56; Xp,,_ = 42.77 8.29). Pred
ators aggregated mostly on the adaxial base of the midrib vein. The fourth leaf is rec
ommended for population sampling studies because the predators aggregate at the
base of the adaxial midrib and T palmi population levels are not extreme on that leaf.

Key Words: thrips, predacious mites, distribution, biological control


Fue estudiada la distribuci6n del acaro depredador Neoseiulus cucumeris (Oude
mans) y de su presa, Thrips palmi Karny, en Homestead, Florida. Neoseiulus cucume
ris fue mas abundante en los frutos (X = 3.39 0.20) que en las hojas (X = 0.95 0.16)
y no fue encontrado en las flores. Thrips palmi fue mas abundante en las hojas (X =
17.97 + 5.07) que en los frutos (X = 3.22 0.70) y flores (X = 0.93 0.03). En los frutos
y las hojas, la poblaci6n del acaro depredador aument6 con la poblaci6n de T palmi.
Ambas poblaciones fueron bajas en la hoja mas joven (Xdedor 0.00 + 0.00; XP
1.75 0.28) y altas en la hoja mas vieja (Xde,.doa 1.92 0.79; Xs~ 50.83 11.64).
Neoseiulus cucumeris y T palmi fueron mas abundantes en el env6s de la hoja (X, _
mms = 1.58 0.56; X,,, = 42.77 8.29). Los depredadores se agregaron mayoritaria
mente en la base de la vena central, en el env6s de la hoja. Se recomienda la cuarta
hoja para studios de muestreo porque los depredadores se concentran en la base del
env6s de la vena central y porque los niveles polacionales de T palmi no son extrema
damente altos o bajos en esa hoja.

The melon thrips, Thrips palmi Karny, is an important vegetable pest in South
Florida, attacking beans, cucurbits, eggplants, peppers, and potatoes (Seal & Bara
nowski 1992). Thrips palmi was described from Sumatra in 1925. It was considered
an insect without economic importance for more than 50 years, but since 1978 it be
came a major threat to vegetable growers in Asia (Sakimura et al. 1986). In 1985 T

Florida Entomologist 80(2)

palmi was detected in the Caribbean (Denoyes et al. 1986), and in 1991 it was found
in Homestead, Florida (South 1991). Losses of more than 10 million dollars caused by
T palmi were reported on peppers in Palm Beach County, Florida, in 1993 (Nuessly
& Nagata 1995).
The predacious mite Neoseiulus cucumeris (Oudemans) has been tested in the field
as a potential biological control agent for suppression of T palmi on eggplants (Casti
neiras et al. 1997). Neosiulus cucumeris is mass reared on fungus mites in wheat bran
and sold for release in commercial greenhouses (Hoy & Glenister 1991). In eggplants,
N. cucumeris is released by sprinkling the bran on top of the leaves (Castineiras et al.
To evaluate the efficacy of a biological control agent, both the predator and the
prey must be monitored from the moment of release through harvest; thus, knowledge
of their distribution within the plant is essential.
There is no information on the distribution of N cucumeris in eggplant. Thrips
palmi is known to be more abundant on eggplant leaves than on flowers and fruits
(Kawai 1988). We examine here the distribution pattern of N cucumeris and T palmi
within eggplants where controlled releases of the predator were made.


The study was conducted from Oct. 1995 through Apr. 1996 at the University of
Florida Tropical Research and Education Center in Homestead. Three 11 x 12.5 m
plots spaced 2.5 m apart were set in beds 0.2 m high and 0.9 wide, covered with black
polyethylene mulch to retard weed growth. Five-week old eggplant (Solanum melon
gena var. Classic) seedlings were transplanted 0.6 m apart in double rows on 12 Oc
tober 1995. A mix of maneb [1.38 kg (AI)/ha] and copper hydroxide [2.88 kg (AI)/ha]
was sprayed weekly to prevent diseases. Weeds in the interbed spaces were controlled
with a mixture of paraquat [0.87 kg (AI)/ha] and diquat [0.83 kg (AI)/ha].
Neoseiulus cucumeris (IPM Laboratories, Inc., Locke, NY) was released in wheat
bran on the top of the leaves at a ratio of one predator per prey which is the recom-
mended ratio for biological control of T palmi by N. cucumeris (Castineiras et al.
1997). Number of T palmi per plant was estimated before predator releases by aver
aging the number of larval and adult thrips on the second, fourth, and sixth leaves of
10 shoots on 10 randomly selected plants per plot and multiplying the mean by the av
erage number of leaves per plant. The first leaf longer than 2.5 cm from the base to the
apex on a shoot was considered the terminal leaf. One hundred predators per plant
were released on week 7 after transplanting, when thrips population averaged 99.0
per plant, and 200 predators per plant were released on week 10 after transplanting,
when thrips population averaged 198.5 per plant.
A sample of ten flowers, 30 fruits, and 30 leaves per plot was taken at random on
the first and second week after each release. The fruit sample consisted of 10 small (2
4 cm long), 10 medium (5-10 cm long) and 10 large (15-20 cm long) fruit taken at ran
dom within each plot. The leaf sample consisted of the first, fourth, and seventh leaves
of a shoot taken at random on each of 10 plants per plot. All samples were collected
separately and taken to the laboratory in plastic bags.
The number of T palmi larvae and adults and all stages ofN cucumeris inside the
flowers, under the fruit calyx, and on the leaves was counted under the microscope.
The leaf surface was divided in two halves, from the center to the tip and from the cen
ter to the base. Each half was also divided into 4 areas: Abaxial and adaxial leaf sur
faces and abaxial and adaxial midribs.
The data from the four samplings were averaged for each replicate. Data were
square root transformed and analyzed using general linear models (SAS Institute,

June, 1997

Castineiras et al.: Distribution ofNeoseiulus cucumeris 213

Inc. Cary, NC). A one-way ANOVA was used for fruit data, and three-way ANOVAS
were used for leaf data. Leaf position (first, fourth, and seventh), leaf side abaxiall and
adaxial) and leaf area (tip, base, midrib tip and midrib base) were considered the main
effects in the three-way ANOVAS. Curves for N cucumeris against T palmi popular
tions on leaves and fruits were fit by nonlinear regression analysis using TableCurve
2-D (Jandel Scientific, Inc., San Rafael, CA).


Neoseiulus cucumeris was observed on the fruits (X =3.39 0.20) and leaves (X =
0.95 0.16) but not inside the flowers. The number of T palmiwas lower in the flowers
(X = 0.93 0.03) than on the fruits (X = 3.22 0.70) and leaves (X = 17.97 5.07), as
previously documented (Kawai 1988).
Predators tend to aggregate where prey densities are high (Varley et al. 1974). Re
gressions of N cucumeris density on T palmi density yielded significant relationships
for the leaves [No. N cucumeris = 1.18 + 0.01(No. T palmi)12; r2 0.99, F 867.17]
and fruits [No. N cucumeris = 24.72 + 4.04(No. T palmi0 ; r2 0.95, F = 63.63]. The
regression equations show that increases in prey population were followed by in
creases in predator population on both leaves and fruits. Neoseiulus cucumeris also
congregates on cucumber and cabbage leaves with high thrips population densities af
ter release (Gillespie 1989, Hoy & Glenister 1991).
Neoseiulus cucumeris and T palmi populations increased with fruit size (Table 1).
Thrips palmi was on the fruit from the developing ovary phase through fruit maturity.
After petal abscission, when the fruits were 2-4 cm long and the calyx began to open,
T palmi and N cucumeris aggregated under the sepals.
Neoseiulus cucumeris preferred the ridges of the underside of the fruit sepals over
the leaves for oviposition. F 'li 1, i... percent of N cucumeris eggs were found under
the sepals and 11% on the leaves. On the leaves, predator eggs were always found at
the base of the adaxial midrib, hidden under the trichomes.
In the shoots, numbers of N cucumeris and T palmi increased from the first
through the seventh leaf (Table 2). Both predator and prey were more abundant on
the adaxial surface of the leaves (Table 3). Neoseiulus cucumeris populations concern
treated mostly on the midrib base. However, T palmi populations distributed all over
the leaf surface and seemed to avoid the midrib tip (Table 4).
Analyses of variance showed significant interactions of leaf position, leaf surface,
and leaf area for both the predator and the prey (Tables 5 and 6). Neoseiulus cucum
eris population density was highest on the adaxial midrib base of the seventh leaf. On


N cucumeris T palmi
Fruit length (Mean SE)' (Mean SE)2

2-4cm 1.23+ 0.12 1.76+ 0.08
5-10 cm 3.56 0.12 3.40+ 0.23
15-20 cm 5.36 0.14 4.50+ 0.15

'ANOVA on square root transformed data, untransformed means are presented. F = 257.69, p > 0.0001, df =
2, 6.
ANOVA on square root transformed data, untransformed means are presented. F = 67.20, p > 0.0001, df =
2. 6.

Florida Entomologist 80(2)

June, 1997


N cucumeris T palmi
Leaf position (Mean SE)' (Mean SE)2

First 0.00 + 0.00 1.75 +0.28
Fourth 0.54 +0.30 20.04 + 3.92
Seventh 1.92 0.79 50.83+ 11.64

'ANOVA on square root transformed data, untransformed data means are presented. F = 282.04, p > 0.0001,
df 2, 48.
2ANOVA on square root transformed data, untransformed data means are presented. F = 2702.70, p > 0.0001,
df 2, 48.


N cucumeris T palmi
Leaf side (Mean SE)' (Mean SE)2

Abaxial 0.05 + 0.03 5.63 + 0.99
Adaxial 1.58 +0.56 42.77 +8.29

'ANOVA on square root transformed data, untransformed means are presented. F = 501.89, p > 0.0001, df
2ANOVA on square root transformed data, untransformed means are presented. F = 3883.74, p > 0.0001, df

the fourth leaf, predators were found only at the base of the adaxial midrib (Table 5).
Highest T palm density was found on the adaxial tip of the seventh leaf (Table 6).
Considering that predator and prey only coincided on leaves and fruits, both leaves
and fruits can be used for sampling proposes. It is more convenient to sample the
leaves because they are easier to handle than the fruits. The fourth leaf is best for
monitoring N cucumeris and T palmi because the predators aggregate at the base of


N cucumeris T palmi
Leaf areas (Mean SE)' (Mean SE)2

tip 0.00+ 0.00 31.88 + 11.52
base 0.38 +0.21 31.00 + 10.61
midrib tip 0.08 0.04 5.22+ 1.10
midrib base 2.80+ 1.05 28.72+ 1.05

'ANOVA on square root transformed data, untransformed means are presented. F = 344.98, p > 0.0001, df =
'ANOVA on square root transformed data, untransformed means are presented. F = 443.89, p > 0.0001, df =
3 48.

Castineiras et al.: Distribution ofNeoseiulus cucumeris 215

C. C.

SE E 2n E E


- +1



t 3

- +I




June, 1997

Florida Entomologist 80(2)

SE E 2n E
7 B 7 B B

7 7 7

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs