A Tephritid Fly Mimics the Territorial Displays of its Jumping Spider Predators
Erick Greene; Larry J. Orsak; Douglas W. Whitman
Science, New Series, Vol. 236, No. 4799. (Apr. 17, 1987), pp. 310-312.
Stable URL:
http://links.jstor.org/sici?sici=0036-8075%2819870417%293%3A236%3A4799%3C310%3AATFMTT%3E2.0.CO%3B2-U
Science is currently published by American Association for the Advancement of Science.
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained
prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in
the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/journals/aaas.html.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic
journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers,
and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take
advantage of advances in technology. For more information regarding JSTOR, please contact support@jstor.org.
http://www.jstor.org
Wed Jan 16 17:10:06 2008
wings with a black felt-ti ped marker. Flies were
first immobiid by ch&g
them at -10°C for
about 30 seconds.
12. To determine that pen markine' on the wing do not
directl cause increased m o d t y , we maintained 20
flies,
of which we marked, within a Plexiglas
cage. Mean age at death for each group did not Mer
Mann-Whimey U test). Further, 0bse~ationSconed that flies did not behave diferently after
ib
6,
marking, nor was their ability to flee spiders hampered.
an NSERC Canada
13. This spdy was sup rted
o ranng grant to f?D.R.%e thank L. Dill, R.
&berg,
and R. Smith for reviewing the manuscript. We also thank L. Dill and B. Lalonde for
advlce and B. Lalonde for photographs.
20 October 1986; accepted 11 F e b v 1987
A Tephritid Fly Mimics the Territorial Displays of Its
Jumping Spider Predators
The tephritid fly Zmmemuta vittigera (Cquillett) has a leg-like pattern on its wings
and a wing-waving display that together mimic the agonistic territorial displays of
jumping spiders (Salticidae). Zmmemuta flies initiate this display when stalked by
jumping spiders, causing the spiders t o display back and retreat. Wing transplant
experiments showed that both the wing pattern and wing-waving displays are
necessary for effective mimicry: Zmsemuta flies with transplanted house fly wings and
house flies with transplanted Zmsemuta wings were attacked by jumping spiders.
Similar experiments showed that this mimicry does not protect Zotwsemuta against
nonsalticid predators. This is a novel form of sign stimulus mimicry that may occur
more generally.
M
OST FORMS OF MIMICRY, SUCH
as cryptic coloration or Batesian
and Miillerian systems, confer
protection against a wide array of predators
(1). We describe a novel form of mimicry in
which an organism mimics its major predator and thereby reduces the risk of being
eaten by it. A tephritid fly, by mimicking the
stereotyped aggressive behavior of one family of spiders, can escape from spiders of this
family but not from other predators.
The fly Zmsemata v i e e r a (Diptera:
Tephritidae) is purported to mimic jumping
spiders (Araneae: Salticidae) (2, 3). Both
sexes have dark wing bands, which resemble
spider legs, and false eyespots on the end of
the abdomen. When disturbed, these flies
hold their wings perpendicular to the body
and wave them up and down (Fig. 1A); this
resembles the agonistic leg-waving behavior
typical of the jumping spiders. However,
there have been no experimental demonstrations that Zmsemata is a spider mimic.
Many flies have dark wing markings and
wing-flicking displays, so Zunosemata might
fortuitously resemble jumping spiders, but
not gain protection from predators by these
features. If Zunosemata is in fact a jumping
spider mimic, it is not clear what types of
predators are deterred. Since salticids are
quick and have a poisonous bite, it has been
suggested that a salticid mimic may be
shunned by many vertebrate and arthropod
predators (3).
Another possibility, which had not been
suggested, is that Zmsemata displays may
specificallymimic salticid territorial displays,
and be effective only against salticid predators (4). Many salticids defend "privacy
spheresyyaround themselves. Wheri two
meet they usually initially perform agonistic
displays (which may turn into courtship
displays depending upon sex and species)
(5). These displays can be performed by
juveniles and adults of both sexes and occur
within and between species. Although the
Flg. 1. ( A ) A female Zmnnata vitt&cra beginning its wing-waving display toward a stalking jumping
spider (Phidippw qacheanw).The jumping spider stopped stalking, waved its legs at the fly, and then
retreated. ( B ) A Zaosnnata a'tt&cra fly with transplanted house fly wings. Such flies can display
normally and fly.
precise details of these stereotyped behavior-
2 displays vary intraspe~ifica"~,salticid agonistic displays generally commence with legwaving (6).
To test the effect of the wing pattern and
the wing-waving display on the behavior of
jumping spiders and other potential predators, we transplanted wings between house
flies (Mwca domestics) and Z m s m a flies
(7). House fly wings are the same general
size and shape as Zonaremata wings, but they
lack pattern. After this operation, the flies
retained complete movement of their wings,
and could display and fly normally (Fig.
1B).
Behavioral trials between jumping spiders
and flies were conducted for 5 minutes in a
glass-topped arena (8). Jumping spiders
were collected on or around Zmsemata
host plant (silver leaf nightshade, Solanum
elaegnifolium). Twenty jumping spiders
representing 11 species (9) were each presented with five treatments: normal Zonosemata, Zmsemata with other Zonusemata
wings (sham operation), Zonusemata with
house fly wings, house flies with Zonosemata
wings, and normal house flies. Each spider
was presented with these treatments in a
random order. All jumping spiders were
hungry when tested: they were given water
but no food for 2 days before the trial.
Individual spiders were never tested more
than twice in one day.
The wing pattern had a profound effect
upon jumping spider behavior (Fig. 2).
Normal Zonosemata and the sham-operated
control flies were attacked or killed less
fiequently than flies in the three remaining
treatments (10). There was no smtisticallv
significant diffeknce in the jumping spider;'
responses to the normal Zonosemata flies and
thi sham-operated control flies (homogeneity test, G = 3.28, P > 0.1), indicating that
the operation itself did not affect spider
responses. Jumping spiders began stalking
these flies within seconds after the trial
began. When the spider approached to within about 5 cm, Zonosma flies usually
began a vigorous wing-waving display.
response, the jumping spiders abruptly
stopped stalking and waved their legs at the
flies. The flies backed away in a zigzag
fashion while waving their wings and flew
off. Most jumping spiders made no further
stalking attempts during the remaining 5
minutes. Jumping spiders were repelled
from both the front and back of the flies. In
E. Gmne, De amnent of Biology, Princeton University, Princeton, b 0 8 5 4 4 .
L. J. Orsak, Institute of Ecology, University of Georgia,
Athens, GA 30602.
and Population ManD. W.Whirman, Insect Biol
agement Research LaboratoV~.s.Dqamnent of
"culture, Agricultural R-ch
Service, P.O. Box 7%;
Tifton, GA 31793.
SCIENCE, VOL. 236
and normal house flics). Of particular intcrest, Zonosernata with house fly wings displayed identically to normal Zonosenzata
flies, but this never elicited the leg-waving
displays from spiders, and all but one were
attacked or killed. House flies with Zonosenzata wings held the wings flat over their
bodies: the pattern was not visible to spiders, which never displayed toward these
flies.
T o determine if these displays also protected Zonosernata flics against other potcntial prcdators, we performed similar expcrimcnts using nonsalticid spiders (Oxyopessalticus), mantids (Mantis rel&iosa), assassin
bugs (Pselliopus zebra), and whiptail lizards
(Cnernidophorous uniparens; Arizona Research Permit 089). All predators wcre
caught on or around the host plant of
Zonosernata. These predators were presented
Fig. 2. Behavioral responses of jumping spidcrs to with three fly treatments (A, C, and E of
fly presentations. The response is the highest level Fig. 2). The same testing protocol was used
of aggression attained during 5 minutes of behav- except that the test chamber sizes were
ioral interaction in a test arena. The fly treatments varied to accommodate the different sizes of
arc: A, normal Zonosemata; R, Zonosemata with
%)nosenzatawings glued on (control for the opera- these predators (11).
The Zonosenzata display is not effective
tion); C, Zonosemata with housc fly wings; D,
house fly with Zonosemata wings; and E, normal against these four types of prcdators (Table
house fly. Sample sizes arc 20 for each fly treat- 1). None wcrc deterred by displaying flies,
ment. The bars connect homogeneous groups (G
tcsts, P's > 0.1). All other combinations are het- as occurred with jumping spiders. For each
predator, there was 110 statistically signifierogeneous (all P's < 0.01).
cant difference in capture times for the three
six trials the jumping spiders stalked from treatment groups (pairwise Mann-Whitney
behind, apparently ~ o 6 s e r v e dby the fly. U tests, all P's > 0.05).
The flies gave spontaneous wing-flicks
In summary, Zonosemata is a specialized
(about one series in 10 seconds). The jump- mimic. Rather than conferring protection
ing spiders stopped stalking, waved thEir against many types of visual prcdators, this
legs, and backed away.
jumping spider mimicry is effective only
However, not all Zonosemata flies with the against jumping spidcrs. The neurological
wing pattern were immune to jumping spi- mechanism of this mimicry is clear: jumping
der attacks: of the 40 Zonosemata in treat- spidcrs possess feature detectors in the retina
ments A and K, 2 wcre attacked, and 8 of their central anterior-median eyes which
others were killed. This is partly a reflection arc excited by waving leg-like patterns (12).
of the conservative bias used in classifying Model presentations to salticids have shown
the behavioral responses: spiders were that leg-like patterns are potent releasers
scored on their highest level of aggression that cause jumping spidcrs to abruptly stop
during
" the 5-minute ex~xriment.Of these what they are doing (usually stalking) and
ten flics that were attacked, six probably display back (13). Thus, the Zonosemata
would have escaped if unconfined since they display mimics a sign stimulus recognized
repeatedly displayed and repelled attacks (a by salticids but not by other predators. This
mean of 2.1 effective displays). Also, even accounts for the extreme specificity of the
though they were eventually attacked, the protection: other examples of sign stimulus
flies displaying with patterned wings were mimicry arc cffcctive against a similarly naraEorded some ~rotection:the latencv to the row range of signal receivers (14).
first attack was longer for these six flies
How could this mimicry syndrome have
than for Zonosenzata flies with house fly evolved? Wing markings and wing-flicking
wings (median latency was 44 seconds ver- displays arc common among acalyptrate
sus l l seconds; Mann-Whitney U = 61, P flies, and are especially prominent in the
< 0.005). The remaining four flies were courtship displays of tephritid flics (15).
attacked so quickly they had no chance to Salticids can occur at very high densities,
display.
and they can exert strong predation presIn contrast, the spiders responded much sures on many insects, such as tephritid flies,
more aggressively &ward the three other that spend time on exposed vegetation (16).
types of flies (Zonosemata with house fly It is possible that the defensive display of
wings, house flies with Zonosenzata wings, Zonosernata derived from courtship behav-
A
B
C
D
E
Table 1. Capture times for three fly treatments by
potential predators of Zonosemata. The fly treatment symbols arc the same as for Fig. 2. For each
predator there is no statistically significant difference between treatment medians (painvise MannWhitney U tcsts, all P s > 0.05).
Fly
treatment
A
6:
E
ple
(n)
Median
capture
time
(seconds)
Range
(scconds)
Nonsalticid spzder Oxyopes salt~cus
10
80
16-207
10
77
32-149
10
56
32-292
Assassin bufi Pselliopus zebra 8
234
39-632 8
309
52-889 8
162
34-736 A
C
E
Mantis Mantis
12
12
12
A
C
E
rcligiosa
88
64
94
29-435
11-215
41-341
Wh$tail liza~dCnemidophorous uniparens
10
101
47-355
A
C
10
81
24-370
E
10
92
9-1 19
ior, since refinements of displays that deterred jumping spider attacks could confer
large survivorship advantages. A cursory
glance at a museum drawer of flics reveals
many with leg-like wing patterns. Thus,
jumping spider mimicry may be a widespread phenomenon among many species of
flies. Analogous defenses might be expected
in other insects that are also highly apparent
to a specific class of prcdators.
REFERENCES A N D NOTES
1. M. Edmunds, 1)efenc~in Animals: A Sun~cyoj'AntiI'redntorlIefEnces (Longman, Essex, 1974).
2. T. Eisncr, Nat. Hist. 112, 103 (1984). It has bccn
suggested that other tephritids in the Hhn~oletis
pommella s ccies group reseniblc jumping spidcrs
[A. C. Hofson, Ext. Polde~Minn. A ~ t i c Ext.
.
SEW.
122 (1944); L. G. Monteith, Can. Entomol. 104,
257 (1972); G. L. Rush, cited in T. Eisncr (3, p.
1 n4)1
observed that jumping spidcrs sce~nedto avoid t h i
tephritid fly Rha&~tis pomonella, but he did not
propose a mechanisn~.
5. R. R. Jackson, in Spider (:ommunicatwn:Mechanisms
and Ecolu&al S&n$cance, P. N . Witt and J. S.
Rovner, Eds. (Princcton Univ. Press, Princeton, NJ,
1982), PI) 213-247.
6. L. Forster, in Spider Communication: Mechanisms
and Ecolo~ical S&nzficance, P. N. Witt and J. S.
Rovner, Eds. (Princeton Univ. Press, Princcton, NJ,
1982), PI) 161-212.
7. Zonosemata flics wcrc obtained by gently swcepnetting the host plant, silver leaf ni hnhade (Solanum elaegn$dium). Flies were coltckd between
15 June and 5 August 1986, along State Line Road
near Rodeo, NM, and briefly maintained in a 2 by 1
by 1 nl terrarium that contiined potted nightshade
and water, d q powdered milk, and a honey and
watcr solution. Flies wcre cooled in a refrigerator
until inactive, thal held immobile on an operating
board with paper strips and insect pins. Their wings
were cut off distal to a line connecting the nicdial cell
and &
, (Cmnland the intersection of veins KLti
stock terminology). Wings from donor flies wcre
http://www.jstor.org
LINKED CITATIONS
- Page 1 of 1 -
You have printed the following article:
A Tephritid Fly Mimics the Territorial Displays of its Jumping Spider Predators
Erick Greene; Larry J. Orsak; Douglas W. Whitman
Science, New Series, Vol. 236, No. 4799. (Apr. 17, 1987), pp. 310-312.
Stable URL:
http://links.jstor.org/sici?sici=0036-8075%2819870417%293%3A236%3A4799%3C310%3AATFMTT%3E2.0.CO%3B2-U
This article references the following linked citations. If you are trying to access articles from an
off-campus location, you may be required to first logon via your library web site to access JSTOR. Please
visit your library's website or contact a librarian to learn about options for remote access to JSTOR.
References and Notes
14
Aggressive Mimicry in Photuris Fireflies: Signal Repertoires by Femmes Fatales
James E. Lloyd
Science, New Series, Vol. 187, No. 4175. (Feb. 7, 1975), pp. 452-453.
Stable URL:
http://links.jstor.org/sici?sici=0036-8075%2819750207%293%3A187%3A4175%3C452%3AAMIPFS%3E2.0.CO%3B2-T
NOTE: The reference numbering from the original has been maintained in this citation list.