A Tiger Beetle Aggregation

Not long ago, I received an interesting series of photographs from Joe Warfel, a nature photographer and macro specialist based in Massachussetts.  Joe traveled to Arizona last July, where he photographed an aggregation of Cicindela (Cicindelidia) sedecimpuntata (Western Red-bellied Tiger Beetle¹) near a small pool in the bottom of a dry creek bed at night.  Joe estimates that there may have been as many as 200 to 300 beetles per square meter in the aggregation, most of which were just “hanging out” and with only occasional individuals mating or feeding on moths that had been attracted to his headlamps.

¹ Found in the Sonoran and Chihuahuan Deserts of the southwestern U.S. and south through Mexico to Costa Rica. U.S. and northern Mexican populations are assigned to the nominate subspecies, while more southern populations are classified into four additional subspecies (Erwin and Pearson 2008).

Western Red-bellied Tiger Beetles are among the first tiger beetles to appear prior to the summer monsoons in the Sonoran Desert.  The species is famous for its daytime aggregations of as many as several thousand individuals, which congregate along the drying waterways and prey upon stranded tadpoles and other aquatic organisms (Pearson et al. 2006).  Joe noted that he has seen these aggregations many times before during the daytime at small pools and mudflats, with beetles usually mating and feeding frantically.  However, the aggregation shown in these photographs differs from those daytime aggregations by the relative inactivity of the beetles and the fact that they were congregated on dry ground rather than the moist areas that they frequent during the daytime.  In these respects, it seems to more resemble a communal nocturnal roost such as has been reported for several species of Odontocheila in South America.  In those cases, up to 70 beetles have been found resting on the foliage of low shrubs, apparently as an adaptation to avoid predation by multiplying chemical defense effectiveness as well as awareness of approaching enemies (Pearson and Vogler 2001 and references therein).  Cicindela sedecimpunctata is primarily a diurnal species (i.e., it is active during the daytime), though individuals are often attracted to lights at night, and adults of most diurnal species have been reported spending the night protected in burrows or under detritus and vegetation.  I am not aware of communal nocturnal roosts as a reported behavior for C. sedecimpunctata or any other North American tiger beetle species.

It is a bit ironic to think of tiger beetles – voracious predators that they are – as prey, but they must have many of their own predators to deal with since most species employ multiple antipredator mechanisms. In addition to the communal roosting behavior seen in these photos, a second antipredator characteristic exhibited by this species can be seen in their bright orange abdomen.  The abdomen is fully exposed only during flight, seemingly implying a “flash coloration” function for the bright color that disappears upon landing, momentarily confusing potential predators.  However, Pearson (1985) experimentally determined that orange abdomens in tiger beetles actually have an aposematic function in protecting them from predation against robber flies.  Most tiger beetle species with an orange abdomen also release a combination of benzaldehyde and cyanide² when captured (any tiger beetle collector is familiar with the characteristic “fruity” smell of a tiger beetle releasing benzaldehyde).  Pearson painted the abdomen of paper tiger beetles models either orange or black and endowed them with or without a drop of fresh benzaldehyde.  When presented on a tether to robber flies in the field, orange-abdomened models with benzaldehyde triggered significantly fewer attacks from robber flies than any other combination.  Interestingly however, vertebrate predators (lizards and birds) were not deterred by the defense chemicals or by the orange abdomen, perhaps explaining why only some and not all tiger beetle species produce defense chemicals and have bright orange abdomens (Pearson and Vogler 2001).

² Tiger beetles, thus, join millipedes as being among the few invertebrates that are capable of producing cyanide.

My sincere thanks to Joe Warfel for allowing me to use his photographs. More of his work can be seen at Eighth-Eye Photography.  Joe also recently had several images published in American Scientist magazine (November/December 2009 issue) for an article on harvestmen.  Check out the jaws on that juvenile!


Erwin, T. L. and D. L. Pearson. 2008. A Treatise on the Western Hemisphere Caraboidea (Coleoptera). Their classification, distributions, and ways of life. Volume II (Carabidae–Nebriiformes 2–Cicindelitae). Pensoft Series Faunistica 84. Pensoft Publishers, Sofia, 400 pp.

Pearson, D. L.  1985.  The function of multiple anti-predator mechanisms in adult tiger beetles (Coleoptera: Cicindelidae).  Ecological Entomology 10:65–72.

Pearson, D. L., C. B. Knisley and C. J. Kazilek. 2006. A Field Guide to the Tiger Beetles of the United States and Canada. Oxford University Press, New York, 227 pp.

Pearson, D. L. and A. P. Vogler.  2001. Tiger Beetles: The Evolution, Ecology, and Diversity of the Cicindelids.  Cornell University Press, Ithaca, New York, 333 pp.

Copyright © Ted C. MacRae 2010

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to Yahoo BuzzAdd to Newsvine

Email to a friend


Warning: post contains hardcore, taxonomic, sciencey geekiness!

Just as there is seasonality in the lives of insects, there is seasonality in the work of those who study them.  For the collector/taxonomist, everything revolves around the collecting season — time spent on anything else is time not available for collecting. As a result, I spend a good deal of my time during the summer in the field and on its associated planning and organizing activities, leaving the winter months for processing and identifying collected specimens, incorporating them into the permanent collection, generating reports to fulfill permit requirements, and ultimately preparing manuscripts for publication — the raison d’être.  Winter is also the time when I identify specimens sent to me by other collectors.  I do this not only because I’m such a nice guy (at least I hope I am), but also because such material often contains species I haven’t seen before or that represent new distributions and host plant associations that I can use to augment the results of my own studies.  Such work has occupied much of my time during the past several weeks, and I now find myself close to finishing the last of the nearly dozen batches of beetles sent to me since the end of last winter.

Of the three groups of beetles that I actively study — jewel beetles, longhorned beetles, and tiger beetles — it is the jewel beetles that are taxonomically the most challenging.  Tiger beetles can often be indentified in the field (especially with the publication of Pearson et al. (2006), or “the Bible” among cicindelophiles), and North American longhorned beetles have been reasonably well worked by a strong contingent of both professional and amateur taxonomists over the past several decades.  Jewel beetles on the other hand, despite their dazzling colors and popularity with collectors, continue to befuddle even the most dedicated collectors due to their extreme variability and poorly-defined species limits.  Of the 822 species and subspecies known from North America, fully three-fifths of them belong to one of just three hyperdiverse genera — Acmaeodera, Agrilus, and Chrysobothris.  No recent taxonomic treatments are available for any of these genera, thus, identifying species belonging to them requires access to primary literature, a well-represented and authoritatively-identified reference collection, and extraordinary patience!  This is particularly true of the genus Acmaeodera, the North American members of which were last treated collectively more than a century ago (Fall 1899) (at which time less than half of the current 149 species/subspecies were known to science).  The recent explosion of web-based images has helped matters (a particularly useful site for those interested in North American Acmaeodera is Acmaeoderini Orbis, with its galleries of Harvard type specimens and BugGuide photos); however, images are still lacking for many species, and others are not easily distinguished from the images that do exist.

Acmaeodera robigo Knull (Val Verde Co., Texas)

It is precisely this taxonomic challenge, however, that makes the group so interesting to me.  Opportunities for discovery abound, as basic information is incomplete or totally lacking for many species regarding their geographical ranges and life histories.  One of the species I encountered in a batch of material sent to me by cerambycid-specialist Jeff Huether contained three specimens that I eventually determined to represent Acmaeodera robigo.  Josef Knull (1954) first described this species from specimens collected at Lake Corpus Christi in south Texas, and nothing more was recorded about the species until Nelson et al. (1996) reported a single specimen cut from its pupal cell in the base of Dalea formosa (Fabaceae) at White River Lake in far northern Texas — a range extension of almost 500 miles!  Obviously, I didn’t have this species in my collection, and it was only after a series of eliminations that led me to the original description (and confirmation of my ID by Nearctic Acmaeodera-guru Rick Westcott based on the photos shown here) did I know for sure what it was.  These specimens were collected at Seminole Canyon State Historic Park, thus, extending into west Texas the species’ known range, and they exhibit variability in the elytral markings and punctation that was not noted in the original description.  While only an incremental increase in our knowledge of this species, collectively such increases lead to greater understanding of the genus as a whole, and Jeff’s generosity in allowing me to retain examples of the species increases my U.S. representation of the genus to 130 species/subspecies (87%).

Acmaeodera n. sp. (Santa Cruz Co., Arizona)

The opportunity for discovery is not limited to range extensions and new host records, but includes new species as well.  A few years ago I received a small lot of specimens collected in Arizona by my hymenopterist-friend Mike Arduser (hymenopterists, especially those interested in apoid bees, are excellent “sources” of Acmaeodera, which they encounter frequently on blossoms while collecting bees).  Among the material he gave to me was the single specimen shown here that immediately brought to my mind Acmaeodera rubrovittata, recently described from Mexico (Nelson 1994) and for which I had collected part of the type series.  Comparison of the specimen with my paratypes, however, showed that it was not that species, and after much combing through the literature I decided that the specimen best fit Acmaeodera robigo (despite being collected in Arizona rather than Texas and not matching the original description exactly).  This was before I had true A. robigo with which to compare, so I sent the specimen to Rick Westcott for his opinion.  His reply was “good news, bad news” — the specimen did not represent A. robigo, but it didn’t represent any known species either!  While the prospect of adding a new species to the U.S. fauna is exciting, basing a description on this single specimen would be ill-advised.  Only through study of series of individuals can conclusions be made regarding the extent of the species’ intraspecific variability and its relation to known species.  Until such specimens are forthcoming, the specimen will have to sit in my cabinet bearing the label “Acmaeodera n. sp.”  For all of you collector-types who live in or plan to visit southeastern Arizona, consider this a general call for potential paratypes!  The specimen was collected in early August on flowers of Aloysia sp. near the Atascosa Lookout Trailhead on Ruby Road in Santa Cruz Co.


Fall, H. C.  1899.  Synonpsis of the species of Acmaeodera of America, north of Mexico.  Journal of the New York Entomological Society 7(1):1–37.
[scroll to “Journal of the New York Entomological Society”, “v. 7 1899”, “Seq 12”]

Knull, J. N. 1954. Five new North American species of Buprestidae (Coleoptera). Ohio Journal of Science 54:27–30.

Nelson, G. H. 1994. Six new species of Acmaeodera Eschscholtz from Mexico (Coleoptera: Buprestidae). The Coleopterists Bulletin 48:272–282.

Nelson, G. H., R. L. Westcott and T. C. MacRae. 1996. Miscellaneous notes on Buprestidae and Schizopodidae occurring in the United States and Canada, including descriptions of previously unknown sexes of six Agrilus Curtis (Coleoptera). The Coleopterists Bulletin 50(2):183–191.

Pearson, D. L., C. B. Knisley and C. J. Kazilek. 2006. A Field Guide to the Tiger Beetles of the United States and Canada. Oxford University Press, New York, 227 pp.

Copyright © Ted C. MacRae 2010

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to Yahoo BuzzAdd to Newsvine

Email to a friend

Tragidion confusion

Back in October, I discussed a recent review of the cerambycid genus Tragidion, authored by Ian Swift and Ann Ray and published in the online journal Zootaxa.  These gorgeous beetles mimic the so-called “tarantula hawks” (a group of large, predatory wasps in the family Pompilidae) and have been difficult to identify due to poorly-defined species limits, wide range of geographic variation, unusually high sexual dimorphism, and apparent potential for hybridization in areas of geographic overlap. Swift and Ray (2008) recognized seven North American and four Mexican species, including two newly described species and another raised from synonymy. It was an excellent work that provided much needed clarity based on examination of types and included detailed descriptions and dorsal habitus photographs of all species and separate keys to males and females to facilitate their identification. Unfortunately, my summary caused some confusion regarding species that occur in the deserts of southern Arizona, southern New Mexico and western Texas. In this post, I’ll clarify that confusion and provide details for distinguishing these species.

Formerly, it was thought that two species of Tragidion inhabited this region, with populations exhibiting smooth elytra and breeding in dead stalks of Agave and Yucca (Agavaceae) representing T. armatum and those exhibiting ribbed elytra and breeding in dead branches of a variety of woody plants representing T. annulatum. This concept dates back to the landmark monograph of the Cerambycidae of North America by Linsley (1962). Swift and Ray (2008) noted that Linsley’s concept of T. annulatum was based on an erroneously labeled type specimen, and that true T. annulatum referred to populations in California and Baja California (for which other names – now suppressed – were being used). This left the AZ/NM/TX populations attributed to T. annulatum without a name. The previously suppressed name T. densiventre was found to refer to populations inhabiting lowland habitats and breeding in Prosopis and Acacia (Fabaceae), but those occurring in montane habitats and breeding in Quercus (Fagaceae) represented an as yet undescribed species, for which the name T. deceptum was given. I included Swift and Ray’s figure of T. deceptum in my post – but mistakenly included the male of T. densiventre alongside the female of T. deceptum!

This error may never had been noticed, had it not been for the discriminating eyes of BugGuide contributor, Margarethe Brummermann. Margarethe is currently collecting photographs for a field guide to Arizona beetles and had photographed a male and female of a “ribbed” species in Montosa Canyon. Using the illustration of T. deceptum” in my post, Margarethe concluded her specimens represented T. deceptum and asked me to confirm her ID. When I told her the specimens represented T. densiventre, her confusion was understandable (given that her male appeared identical to the T. deceptum” male in my post). Further query on her part prompted me to do a little digging, and I discovered my error. The figure in my post has since been corrected – both that figure and a figure from Swift and Ray (2008) showing the male and female of T. densiventre are included below, along with additional information to allow their identification.


Tragidion densiventre Casey, 1912

Tragidion densiventre was formerly synonymized under T. auripenne (a rare species known from the four corners region of northern Arizona, southern Utah, southwestern Colorado, and northwestern New Mexico). Males of T. densiventre can be distinguished by their longer antennae, tawny-tan elytra and distinctly red-brown head, legs, and scape, while females have shorter antennae and the elytra red-orange. Both males and females of this species are distinguished from T. deceptum by their five elytral costae that curve inward toward the suture and extend to near the elytral apices, as well as their relatively narrower basal black band. Females of this species may be further distinguished from T. deceptum by their all black (or nearly so) antennae. Tragidion densiventre is found predominantly in xeric lowland desert habitats in Arizona, New Mexico, and west Texas (as well as northern Mexico). Larvae have been recorded developing in dead Prosopis glandulosa and Acacia greggii, and adults have been observed aggregating on sap oozing from stems of Baccharis sarothroides (Asteraceae) and flowers of larval host plants. Although the biology of this species has not been described in detail, it is likely that the observations of Cope (1984) for T. auripenne refer to this species. This is the classic T.annulatum” commonly observed in the desert southwest.

Tragidion deceptum

Tragidion deceptum Swift & Ray, 2008

Tragidion deceptum superficially resembles T. densiventre due to its ribbed elytra; however, it is actually more closely related to the Mexican species T. carinatum. Like T. densiventre, the males exhibit longer antennae and tawny-tan elytra, while females have shorter antennae and orange-red elytra. However, the head, legs and scape of males are black, as in females of the species, rather than red-brown as in males of T. densiventre. Females exhibit distinctly annulated antennae, in contrast to the all black antennae of T. densiventre. Both males and females are distinguished from T. densiventre by the elytral costae – only four rather than five, not incurved towards the suture and extending only to the apical one-third of the elytra. In addition, the basal black band is very broad – exceeding the scutellum by 2 × its length. This species is similarly distributed across the desert southwest as T. densiventre but occurs in more montane habitats, where it breeds in recently dead branches of several species of Quercus. Like T. densiventre, adults are often found feeding and aggregating on Baccharis sarothroides, and in a few lower canyons bordering desert habitats in the Huachuca Mountains of southeastern Arizona this species and T. densiventre have been collected feeding alongside each other on the same Baccharis plants. Tragidion deceptum is one of several species in the genus (along with T. coquus in eastern North America) that have been collected using fermenting molasses traps (more on this in a future post).


Cope, J. 1986. Notes on the ecology of western Cerambycidae. The Coleopterists Bulletin, 38:27–36.

Linsley, E. G. 1962. The Cerambycidae of North America. Part III. Taxonomy and classification of the subfamily Cerambycinae, tribes Opsimini through Megaderini. University of California Publicatons in Entomology, 20:1-188, 56 figs.

Swift, I. and A. M. Ray. 2008. A review of the genus Tragidion Audinet-Serville, 1834 (Coleoptera: Cerambycidae: Cerambycinae: Trachyderini). Zootaxa, 1892:1-25.

Copyright © Ted C. MacRae 2009

Add to FacebookAdd to NewsvineAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to Ma.gnoliaAdd to TechnoratiAdd to Furl

New species and a review of the genus Tragidion

ResearchBlogging.orgSpecies of Tragidion are among the larger and more attractive cerambycids in North America, making them popular among collectors. Their bright orange and black coloration clearly functions in mimicking spider wasps (family Pompilidae) in the genera Pepsis and Hemipepsis – the so-called “tarantula hawks.” Unfortunately, species of Tragidion have been difficult to identify due to a high degree of morphological similarity between species, wide range of variation across geographic areas within species, unusually high sexual dimorphism and dichromatism, and apparent potential for hybridization in areas of geographic overlap. This has confounded efforts to delimit species boundaries, resulting in a confusing assortment of names whose proper application has eluded even the most esteemed of North America’s cerambycid taxonomists. Recently, some much needed clarity was provided by Ian Swift and Ann M. Ray in the journal Zootaxa. Their taxonomic review of Tragidion – the first systematic treatment of the entire genus – recognizes seven species in North America and another four restricted to Mexico. Two species – T. agave from California and Baja California and T. deceptum from montane areas of the southwestern U.S. and northern Mexico (both pictured) – are described as new, and a third – T. densiventre from desert areas of the southwestern U.S. and northern Mexico – is raised from synonymy under T. auripenne. Four new synonymies are also proposed, and dorsal habitus photographs and a key to all species are provided. Life history information is limited for most species of Tragidion. One species – T. coquus – occurs broadly across the eastern and central U.S., where it breeds in a variety of dead hardwoods, especially oak. Several species occur in the southwestern U.S. and northern Mexico – some are found in xeric lowland desert habitats, where they breed in dead branches of Prosopis glandulosa and Acacia greggii (T. densiventre) or dead flower stalks of Yucca and Agave (T. agave and T. armatum), while a fourth (T. deceptum) is found in more montane habitats mining the heartwood of recently dead branches of Quercus. Adults of another species in California and Baja California, T. annulatum, are strongly attracted to brushfires and burning vegetation, and individuals have been observed landing on still-burning and smoldering shrubs, causing their legs and abdomens to melt to the surface of the branches. At several post-burn sites, the melted bodies of this species were common on the charred branches of their hosts, and females have been observed ovipositing on woody shrubs that have been burned. This species likely plays an important role in the decomposition of burned woody material in coastal areas of California. The remaining U.S. species – T. auripenne – is known from only a handful of specimens collected in xeric habitats in the Four Corners region of the southwestern U.S. It’s life history, as well as those of the four strictly Mexican species, remains essentially unknown.

Tragidion agaveTragidion agave Swift & Ray 2008, ♂ & ♀ – California & Baja California. Tragidion deceptumTragidion deceptum Swift & Ray 2008, ♂ & ♀ – southwestern U.S. and northern Mexico.

REFERENCE: Swift, I., Ray, A. M. (2008). A review of the genus Tragidion Audinet-Serville, 1834 (Coleoptera: Cerambycidae: Cerambycinae: Trachyderini) Zootaxa, 1892, 1-25