Featured

Artifice Among Asilidae: An investigation of insect mimicry, focusing on robber flies

While creating this page, I remembered that I wrote a paper in college on mimicry in robber flies. Spoiler alert: it’s almost all conjecture.

Mallophora orcina eating what it’s thought to mimic, a bumblebee

INTRODUCTION

Mimicry occurs when an organism has evolved to resemble another organism, most often to the direct benefit of the mimic.  Mimicry has been recorded in many animal orders, such as birds and reptiles, but it is most common in arthropods.  A mimic is successful when it goes unnoticed, or when it tricks another organism, often referred to as the dupe, into thinking the mimic is harmful or harmless, as the situation requires.  Mimicry is often visual, as vertebrate predators such as birds rely heavily on their sense of sight, and may include structural, pattern, or behavioral adaptations of the mimic.  Mimicry may also be auditory (Barber & Conner, 2007; Marshall & Hill, 2009) or olfactory (Ruxton, et al., 2004), the latter case being especially important when the dupe is another arthropod.

Several categories of mimicry have been defined, and are often named after the scientist who originally described it.  As with many fields of biology, these categories do not describe all situations of mimicry, and the boundaries between these categories are often loose.  Mimic systems can be organized into defensive mimicry, aggressive mimicry, and reproductive mimicry.

DEFENSIVE MIMICRY

Adaptations that permit an organism to go unnoticed, to frighten or to deter a potential predator are defensive.  These adaptations, resulting from very long evolutions (Jolivet, 1998), include camouflage, Batesian mimicry, Müllerian mimicry, automimicry, and Wasmannian mimicry.

Camouflage

Camouflage is effective when the mimic is not perceived as separate from the background environment such as vegetation or substrate, a characteristic sometimes known as homochromy.  Stick insects in the genus Phyllium represent a well-known example of camouflage.  These insects imitate a leaf including the associated fungi, lichens, insect damage, and leaf veins (Jolivet, 1998).  Some insects, such as Reduviid nymphs and lacewing nymphs, undertake an active role in their camouflage, adorning themselves with debris and cast skins (Jolivet, 1998).

Batesian Mimicry

Batesian mimicry, coined by Henry Walter Bates in 1862, describes organisms which closely resemble other harmful, distasteful, or unprofitable organisms, but are harmless or otherwise suitable prey themselves.  A Batesian mimic is successful when a predator cannot distinguish between the model and mimic, so there is perpetual evolutionary pressure for the mimic to achieve perfect mimicry with respect to the perception of the predator (Gilbert, 2005).  However, some mimics are readily distinguished from their model by human observers.  These mimics may simply share a common characteristic with a model, and potential predators generalize that the character is associated with a harmful or unprofitable organism.  Possessing conspicuous warning colors or sounds combined with chemical or mechanical defenses, a trait known as aposematism, is common among models, and mimics have evolved to take advantage of those conspicuous features.  Imperfect and distinguishable mimics persist when the model is especially noxious, so that predators will not risk an encounter with a mimic (Gilbert, 2005).

The relationship between Batesian mimics and their models often depends on relative densities; greater densities of Batesian mimics will cause predators to attack the model more frequently (Rowland, et al., 2007).  This suggests that Batesian mimics are most successful at low relative population densities.  For this reason, mimics often resemble models that are generally abundant (Turner, 1984).  However, if the model is particularly distasteful, predators will avoid both the mimic and the model, even when the mimic greatly outnumbers the distasteful model (Gilbert, 2005). 

Examples of Batesian mimicry are numerous, including arthropods from many orders.  The Bornean grasshopper, Condylodera tricondyloides, mimics better defended tiger beetle species in appearance and behavior (Ruxton, et al., 2004).  Some swallowtail caterpillars resemble bird feces, and thus are able to dissuade potential predators.  Several neotropical Sphingidae caterpillars resemble snakes and are able to startle predators.  Vertebrates are also known to mimic arthropods, such as juvenile Eremias lugubris lizards in southern Africa, which mimic noxious Anthia beetles in color, gait, and size (Ruxton, et al., 2004).  Some adaptations, such as the eye spots exhibited by many Lepidoptera, who expose them to frighten away potential predators (Triplehorn & Johnson, 2004), could be considered Batesian mimicry, but there is no specific model.

Müllerian Mimicry

Müllerian mimicry, named after Johannes Friedrich Müller in 1878, describes situations where two or more organisms, each of which is defended, have co-evolved to appear similar to each other.  These organisms often use common warning colors, lights and sounds, such that predators must learn fewer avoidance signals (Triplehorn & Johnson, 2005).  Müllerian mimics are both mimics and models of each other, so each benefits from a reduction in predation.  Müllerian mimicry can be considered as an “interspecific form of aposematism,” where the convergence of warning signals reduces the cost of predator education (Ruxton, et al., 2004).  Some organisms in Müllerian complexes may be less defended than other members of the same complex, though the overall result is usually reduced predation on both the less defended and better defended organisms (Rowland, et al., 2007). 

While it was long thought that edible Viceroy butterflies mimicked the unpalatable Monarch butterfly, it is now known that Viceroys can be just as unpalatable as the Monarch (Ritland & Brower, 1991), so these two species can be considered Müllerian mimics.  Neotropical Heliconius butterflies are perhaps the most widely discussed Müllerian mimics, where 54 species of this distasteful genus have converged to form four mimicry complexes, or mimicry rings (Ruxton, et al., 2004).  Gilbert (2005) suggests that many harmless Batesian mimics resemble complexes of harmful Müllerian mimics, and that the traditional view of a “one-to-one correspondence between model and mimic” is often incorrect.  Indeed, since many harmful organisms have converged in appearance from the benefits of Müllerian mimicry, such as bumblebees in eastern North America, Batesian mimics may have no single model.

Automimicry and Wasmannian Mimicry

Automimicry, or intraspecific mimicry, occurs when an organism is a model for itself.  In some chemically defended arthropods, an organism is not defended when it first emerges or after it has expelled its chemical defense.  However, since it is similar in appearance to other members of the population that are defended, predators are likely to avoid all organisms exhibiting those characteristics (Speed, et al., 2006).  Monarch butterflies employ automimicry, since all members of a population are not equally defended by cardenolides (Cohen, 1985).

Another form of defensive mimic system used by arthropods is Wasmannian mimicry.  Wasmannian mimicry occurs within social insects, where one organism resembles another organism, and also lives within the colony of that organism.  Several species of beetles and jumping spiders resemble ants and live within the ant colonies undetected by the ants.  These mimics gain shelter and protection from the ants (Triplehorn & Johnson, 2004).

AGGRESSIVE MIMICRY

An aggressive mimic uses deception to gain access to resources.  Aggressive mimicry may involve luring potential prey, or may ensure a predator or parasitoid is not perceived by its prey, either through camouflage or through mimicry of non-threatening organisms.

Several orders of arthropods use lures to aid prey capture.  Crab spiders and ambush bugs use flowers as lures to attract unsuspecting prey, and some mantids resemble flowers (Triplehorn & Johnson, 2004).  Photuris fireflies are known to mimic the answering signal of a receptive female Photinus firefly, preying on the male Photinus fireflies they attract (Lloyd, 1965; Triplehorn & Johnson, 2004).  Some arthropods, such as bola spiders, mimic the sex pheromones of their lepidopteran prey (Stowe, et al., 1987; Triplehorn & Johnson, 2004).  Luring signals can also be auditory, as exemplified by the Australian katydid, Chlorobalius leucoviridis (Marshall & Hill, 2009).  This katydid is capable of mimicking the species-specific clicking sounds of sexually receptive female cicadas, and thus attracts male prey of those species.

Aggressive mimicry has also evolved in species employing camouflage or resembling non-threatening species to increase prey capture rate.  For example, green lacewing, Chrysopa slossonae, larvae employ chemical camouflage by attaching tufts of wax from wooly aphids, Prociphilus sp., to their backs.  The ants attending the aphids do not recognize the camouflaged lacewing as an intruder, and the lacewing is able to feed on the aphids unmolested (Ruxton, et al., 2004).

REPRODUCTIVE MIMICRY

Reproductive mimicry is common in vertebrates such as birds, where sexually immature individuals resemble females, so they may live within the same areas as territorial males without inciting aggression.  Mimicry has also evolved where sexually mature males resemble females or immatures, allowing those males to steal copulations with females within another male’s territory.  This type of reproductive mimicry has been recorded in the marine isopod, Paracerceis sculpta.  Some males resemble females or juveniles, allowing them access to females within another male’s territory (Shuster, 1987).

ASILIDAE

Flies in the family, Asilidae, commonly known as robber flies or asilid flies, are voracious predators on other insects, often preying on organisms larger than themselves.  Asilidae await their prey from perches, then fly out to seize them in mid-air.  They are characterized by a hardened proboscis used to jab prey and inject lytic saliva.  Like spiders, digestion is external, and digested tissues are withdrawn as liquid.  Approximately 7100 species of Asilidae have been recorded, and many of them are not mimics.  However, mimicry among robber flies resembling Hymenoptera is apparent to the human observer.  The similarity in appearance between some flies and bees led to the ancient, inaccurate idea that bees spontaneously generated from decaying carcasses (Brower, et al., 1960).

The notion that Asilid flies have evolved to mimic other insects has been contentious.  Hardy (1930) pointed out that robber flies “have not reached the mimicry perfection reached by certain Syrphidae, especially Cerioides subarmata, which also folds its wings longitudinally as an Eumenid wasp, for which it is readily mistaken.”  Melin (1923) flatly disagrees mimicry exists within the Asilidae and adds that theories to the contrary are “merely a product of imagination.”  It has since been demonstrated that even imperfect mimics are taken less frequently by predators (Gilbert, 2007).  More recent evidence has supported that some asilid flies exhibit Batesian, Müllerian, and aggressive mimic systems (Brower, et al., 1960; Hespenheide, 1973; Watmough, 1974).

Camouflage among Asilidae

Few authors have theorized that asilid flies exhibit camouflage, though it is possible that twig-perching Leptogastrinae resemble extensions of the twigs, and that coastal dune denizens exhibit homochromy (Hardy, 1930).  Additionally, species inhabiting deserts tend to match the sandy background of their habitat (Hull, 1962).  These types of adaptations certainly are not unique to the Asilidae, as many organisms and their predators take advantage of homochromy.

Batesian Mimicry in Asilidae

Asilid fly mimics, as with all Dipteran mimics, almost exclusively resemble Hymenoptera (Gilbert, 2007) and include several genera across different subfamilies of Asilidae.  Some of these mimics, such as the Laphria, strongly resemble bumblebees (Bromley, 1925), though their preferred prey are beetles.  This suggests that they are Batesian mimics of hymenoptera.  They benefit from the reduced threat of predation, but are not aggressive mimics, since their appearance may not deceive their prey.  It is unknown if the appearance of Laphria disguises them from their prey, a scenario supporting aggressive mimicry.  Bromley (1934) has pointed out that Laphria species mimic those hymenopteran species that are similar in size and geographical range:  Laphria saffrana mimics the female yellow jacket, Vespula squamosa, L. thoracica mimics Bombus impatiens, L. grossa mimics a Bombus impatiens queen, and L. lata and L. macquarti mimic Bombus pensylvanicus.  It stands to reason that these robber fly species were selected for characters resembling dangerous hymenoptera within their range through reduced predation.

Another Batesian hymenopteran mimic includes Afromelittodes, an African fly thought to mimic Megachile felina (Oldroyd & Bruggen, 1963; Londt, 2003).  The asilid genus, Pepsis, characterized by black or blue-black bodies and yellowish-brown wings, mimic tarantula hunting wasps (Hull, 1962), and may have evolved the resemblance for protection.  The bumblebee produces a conspicuous buzzing noise as a warning signal to potential predators, and some bumblebee mimics have taken advantage of it, producing a similar noise (Brower, et al., 1960; Hull, 1962).

In one of the few studies of mimicry in Asilidae, Brower et al. (1960) devised an experiment comparing the palatability of the asilid fly, Mallophora bomboides, and its model, Bombus americanus. When bees, the asilid flies, and edible dragonflies were presented to toads in a controlled experiment, toads learned to reject both the bees and the asilid flies.  Control toads that were not given the inedible bees did not reject the mimic asilid flies.  For one of the toads, an asilid fly bit the toad on the nose during the first trial.  After the bite, this particular toad refused either the mimic asilid flies or the bees, but not the edible dragonflies.  Though the experiment was not designed to test Müllerian mimicry between Mallophora asilid flies and bumblebees, this result supports it and may warrant further study.

Müllerian Mimicry in Asilidae

The relationship between some asilid mimics and their hymenopteran models may be Müllerian, since asilid flies are not defenseless (Brower, et al., 1960).  Since Müllerian complexes can be unequal and still be beneficial to all members, these complexes could have evolved, even if the hymenoptera are less palatable than the asilid flies.  The distinction between Batesian and Müllerian mimicry can be ambiguous (Gilbert, 2005), and more research is necessary to ascertain these relationships among the Asilidae.

In some Batesian and Müllerian mimicry systems, the models may simply be difficult and unprofitable to capture, and may not be chemically or mechanically defended (Ruxton, et al., 2004).  Hespenheide (1973) describes potential Müllerian mimicry between agile tree-running beetles and agile tree-perching asilid flies sharing similar color patterns.  The color patterns are avoided by predators due to the high energetic cost of prey capture, even though neither organism is distasteful.

Aggressive Mimicry in Asilidae

An aggressive mimic lures its prey through deception, or camouflages itself with the environment or a non-threatening organism to aid prey capture.  Aggressive mimicry in Asilidae has been suggested by several authors (Bromley, 1934; Londt, 2003; Watmouth, 1974), though it has never been proven.  Several asilid flies prefer to eat their model, including Mallophora orcina, which mimics the bumblebee, Bombus pennsylvanicus (Bromley, 1925), and Mallophora fautrix, which mimics Bombus sonorus (Alcock, 1974).  It is thought that the resemblance of these species to their models allows them to approach their prey undetected, increasing their capture rate.  Brower, et al. (1960) defined aggressive mimicry as “the superficial visual similarity of some stage in the life history of a predator to its prey or a parasite to its host.”  Bromley (1934) hypothesized that the evolution of model predation among hymenopteran mimicking asilid flies arose from a confusion of sexual attraction and attraction to prey.  Confused sexual attraction may have increased rates of preying on the model, which the flies already resembled for defensive purposes, and of preying on each other.  Such self-destructive cannibalism would have selected for individuals that did not cannibalize, though the attraction to the model may have remained.

The preference for hymenopteran prey items among asilid mimics of hymenoptera has been documented by several authors.  O’Neill & Seibert (1996) noted that most Megaphorus prey are hymenopteran, Londt (1992) noted that Bana apicida, a bee-mimic in Namibia, prefers to prey on honeybees and similar hymenoptera, and Melin (1923) noticed that Dioctria resemble Ichneumonid wasps and tend to favor them as prey items.  Castelo & Corley (2004) reported that Mallophora reficauda, a South American grassland bee mimic, may cause up to an astounding 80% reduction in honey production during high fly-density years.  Londt (2003) compiled prey data from South Africa supporting hymenopteran preference among asilid mimics of hymenoptera.  The prey data is based on 87 prey records, though it is unknown what biases may exist within these data.  Bromley, et al. (1960) noted that Mallophora bomboides fed exclusively on its model, when this prey was prevalent.  Preference can be misinterpreted for opportunistic predators when a particular prey item is abundant or other prey items are scarce.  Further study would be necessary to establish an unquestionable preference for hymenoptera among these asilid flies.

An example from Africa presents a stronger case for aggressive mimicry.  Asilid flies in the genus, Hyperechia, resemble Xylocopa carpenter bees and will take the adults as prey (Green, 1925).  It is known that these flies oviposit near the carpenter bee burrows, and the larvae migrate into open carpenter bee brood cells prior to the bee laying its eggs and sealing the cells.  The asilid larva then prey on the bee larva.  It is thought the resemblance of the adult fly to the bee may enable the female fly to approach the bee burrows and lay her eggs (Watmough, 1974), a scenario for aggressive mimicry.  As with many possible mimicry examples, experimental evidence supporting these concepts is currently unavailable.

The observational evidence of hymenopteran prey would support the notion of aggressive mimicry among asilid flies as described by Bromley (1934).  The reproductive strategy of Hyperechia flies also fits the concept of aggressive (or parasitic) mimicry.  However, these suggestions of aggressive mimicry based on prey preference are unproven and may be unfounded.  Ruxton, et al. (2004) pointed out that “asilids rarely approach their prey in a ‘casual’ manner as if disguised,” and suggested asilid mimicry is likely Batesian.  Indeed, the evolution of defensive mimicry in these flies is easier to explain, and may be a less dubious concept than aggressive mimicry resulting from confused attraction or through deceiving prey with disguise.  Prey preference in these hymenopteran mimics may be coincidental to the evolution of defensive mimicry.

SUMMARY

Mimicry among the arthropods has been well documented for defensive, aggressive, and reproductive purposes.  Evidence suggests that many asilid flies utilize defensive mimicry, and some may employ aggressive mimicry.  Several genera of asilid flies mimic hymenoptera, and these adaptations benefit the flies by discouraging attack from predators, a case for Batesian mimicry.  Some asilid flies may also be part of Müllerian mimic complexes.  One possible Müllerian complex involves bees, which can inflict a painful sting, and the bee mimics, which can inflict a painful bite.  Another possible Müllerian complex lies within agile, tree-running beetles, and agile, tree-perching asilid flies, both of which have a high energetic cost for capture.  Asilid flies which mimic hymenoptera and also prey on them may use their disguise to capture unsuspecting prey.  One such fly, Hyperechia, preys on its model’s larvae as larvae, and it may use its adult disguise for strategic ovipositing.  However, few studies have been conducted to verify how mimicry may benefit the Asilidae, and more research is needed to advance many of these theories beyond speculation.

REFERENCES

Alcock, J. 1974. Observations on the behavior of Mallophora fautrix Osten Sacken (Diptera: Asilidae). Pan-Pacific Entomologist 50: 68-72.

Barber, J.R., & Conner, W.E.. 2007. Acoustic mimicry in a predator-prey interaction. Proceedings of the National Academy of Sciences 104(22): 9331-9334.

Bromley, S.W. 1925. The Bremus-resembling Mallophorae of the southeastern United States (Diptera, Asilidae). Psyche: A Journal of Entomology 32: 190-194.

Bromley, S.W. 1934. The laphriine robber flies of North America (Diptera: Asilidae). Unpublished doctoral dissertation, Ohio State University, Columbus, OH.

Brower, L.P., Brower, J.V.Z., & Westcott, P.W. 1960. Experimental studies of mimicry: The reactions of toads (Bufo terrastris) to bumblebees (Bombus americanorum) and their robber fly mimics (Mallophora bomboides) with a discussion of aggressive mimicry. The    American Naturalist 94(878): 343-355.

Castelo, M.K., & Corley, J.C. 2004. Oviposition behavior in the robber fly Mallophora ruficauda (Diptera: Asilidae). Annals of the Entomological Society of America 97(5): 1050-1054.

Cohen, J.A. 1985. Differences and similarities in cardenolide contents of queen and monarch butterflies in Florida and their ecological and evolutionary implications. Journal of Chemical Ecology 11: 85-103.

Dikow, T., & Londt, J.G.H. 2000. A review of Lamyra Loew (Diptera: Asilidae: Laphriinae). African Entomology: Journal of the Entomological Society of Southern Africa 8(2): 189-  200.

Gilbert, F. 2005. The evolution of imperfect mimicry. In: Fellowes, M., Holloway, G. and J. Rolff, eds. Insect evolutionary ecology. Wallingford: CABI Publishing. pp. 231-288.

Green, E.E. 1925. Asilid mimicry in Ceylon. Proceedings of the Royal Entomological Society of London 1924: 193.

Hardy, G.H. 1930. Revisional notes on the tribe Brachyrrhopalini (robber flies), with remarks on habits and mimicry. Proceedings of the Royal Society of Queensland 41(5): 59-72.

Hespenheide, H.A. 1973. A novel mimicry complex: Beetles and flies. Journal of Entomology (A)48(I): 49-55.

Hull, F.M. 1962.  Robber flies of the world. Bulletin of the United States National Museum 224 (1, 2): 1-907.

Jolivet, P. 1998. The Interrelation Between Insects and Plants.  Boca Raton: CRC Press.

Lloyd, J. E. 1965. Aggressive Mimicry in Photuris: Firefly Femmes Fatales. Science. 149: 653-654.

Londt, J.G.H. 1992. Bana, a new genus of bee-mimicking assassin flies from southern Namibia (Diptera: Asilidae: Stenopogoninae). Cimbabasia: Journal of the State Museum 13: 91-97.

Londt, J.G.H. 2003. A revision of the afrotropical genus Afromelittodes Oldroyd & Van Bruggen, 1963 (Diptera: Asilidae: Laphriinae) and discussion of its possible mimetic         resemblance to bees of the genus Megachile. African Invertebrates 44(2): 133-146.

Marshall, D.C., & Hill, K.B.R. 2009. Versatile Aggressive Mimicry of Cicadas by an Australian Predatory Katydid. PLoS ONE 4(1): e4185.

Melin, D. 1923. Contribution to the knowledge of the biology, metamorphosis and distribution of the Swedish Asilids.  Unpublished doctoral dissertation, University of Uppsala, Uppsala, Sweden.

O`Neill, K.M., & Seibert, C. 1996. Foraging behavior of the robber fly Megaphorus willistoni (Cole) (Diptera: Asilidae). Journal of the Kansas Entomological Society 69: 317-325.

Oldroyd, H., & Bruggen, A.C. van. 1963.  A new South African bee mimic (Diptera: Asilidae).  Journal of the Entomological Society of Southern Africa 26(1): 190-193.

Ritland, D.B., & Brower, L.P. 1991. The viceroy butterfly is not a Batesian mimic. Nature 350:  497-498.

Rowland H.M., Ihalainen E., Lindstrom L., Mappes J., & Speed, M. 2007. Co-mimics have a mutualistic relationship despite unequal defences. Nature 448: 64-67.

Ruxton G.D., Sherratt T.N., & Speed, M.P. 2004. Avoiding attack: the evolutionary ecology of crypsis, warning signals, and mimicry. Oxford: Oxford University Press.

Shuster, S. 1987. Alternative reproductive behaviors: three discrete male morphs in Paracerceis sculpta, an intertidal isopod from the northern Gulf of California. Journal of Crustacean Biology 7(2): 318-327.

Speed, M.P., Ruxton, G.D., & Broom, M. 2006. Automimicry and the evolution of discrete prey defences. Biological Journal of the Linnean Society 87:3, 393-402

Stowe, M.K., Tumlinson J.H., & Heath, R.R. 1987. Chemical mimicry: bolas spiders emit components of moth prey species sex pheromones. Science 236: 964-967.

Triplehorn, C., & Johnson, N. 2004. Borror and Delong’s Introduction to the Study of Insects 7th ed. Belmont, California: Thomson Brooks/Cole.

Turner, J.R.G. 1984. Darwin’s coffin and Doctor Pangloss: do adaptionist models explain mimicry? In: Shorrocks, B. (ed.) Evolutionary Ecology.  Oxford: Blackwell Scientific.  pp. 313-361.

Watmough, R.H. 1974. Biology and behaviour of carpenter bees in southern Africa. Journal of the Entomological Society of southern Africa 37(2): 261-281.