Cycloalexy in tortoise beetle larvae

One of the first insects I encountered during my visit this past November to  in Buenos Aires, Argentina were these tiny beetle larvae grouped together on a single leaf of an unidentified shrub.  The presence of fringed lateral appendages and exuvial-fecal debris masses held by caudal appendages immediately identifies them as larvae in the leaf beetle subfamily Cassidinae, known commonly in North America as “tortoise beetles” due to the appearance of the adults.  With nearly 3,000 species distributed throughout the world, tortoise beetles are easily recognizable as a group; however, species identifications can be much more difficult, especially in the Neotropics where the group reaches its greatest diversity (Borowiec and Świętojańska 2002–2011). Identification of larvae can be even more challenging, as the larvae of many species remain unknown, and I was unable to find adults in association with the larvae to aid my identification.

Anacassis sp. (poss. exarata) early-instar larvae on Baccharis salicifolia | Buenos Aires, Argentina

Nevertheless, host plant can be an important clue to leaf beetle identity, as most species in the family limit their feeding to a single plant genus or group of related plant genera. The shrub on which the beetles were feeding looked familiar to me, and while perusing a list of plants that have been recorded from the Reserve (Burgueño 2005) I had an “Aha!” moment when I spotted the asteraceous genus Baccharis. I decided the plant must represent Baccharis salicifolia because of its narrowly lanceolate, willow-like leaves with fine apical serrations (Cuatrecasas 1968) (see first photo). The only tortoise beetles known to feed on Baccharis are species in the genus Anacassis (McFadyen 1987), several species of which are known from Argentina, and one (Anacassis exarata) looking very much like the larvae in these photos.

Note the circular, heads-directed-inward orientation of all larvae around the periphery

The manner in which these early-instar (perhaps even newly hatched) larvae were feeding as a group while working their way down the length of the leaf towards its base is not something I had observed before. Larvae of most tortoise beetles are solitary feeders (Borowiec and Świętojańska 2002–2011), and I was further intrigued by the deliberate circular formation that the larvae had assumed.  The larvae around the periphery were all facing inward, tightly packed against each other and with their exuvial-fecal debris masses directed outward. Additional larvae were seen inside the circular formation. As I manipulated the leaf for photographs, the larvae would occasionally raise their debris masses up and outward, presumably a defensive reaction to disturbance and a perceived threat. It was clear to me that the larvae had deliberately “circled their wagons” for defensive purposes.

Close body contact allows exuvial-fecal debris masses to form a protective barrier against predators

In fact, this type of defensive strategy has been reported in a number of South American cassidines, as summarized by Jolivet et al. (1990), who coined the term “cycloalexy” (from the Greek κύκλος = circle, and αλεξω = defend) to describe such strategies. Cycloalexy can either be “heads in, tails out” or vice versa and is usually associated with other repellent activities such as coordinated threat movements, regurgitation, or biting. The strategy is intended to provide protection from predators such as ants and true bugs and parasitioid wasps, although some parasitoids seem to have thwarted the strategy by depositing their eggs where they can be ingested (thus avoiding direct confrontation with the prey). Cycloalexy has been described primarily among chrysomelid beetles and tenthredinoid hymenopterans (sawflies); however, examples from a few other insect orders (e.g., Diptera, Neuroptera, Lepidoptera) are known as well (Jolivet 2008).  All known cycloalexic insects are subsocial in the larval stage and often also exhibit maternal protection of eggs or newly hatched larvae.

This and several other older larvae had become solitary, presumably protected in part by greater size

In addition to this single group of early instar larvae, I noted also a few larger individuals—all of whom were feeding on the plant in a more solitary fashion. Presumably as the larvae grow larger they are more able to defend themselves, or perhaps larger larvae simply demand more “elbow room” because of the larger amounts of leaf tissue they require for feeding. If cycloalexy is beneficial for small cassidine larvae but less so for larger larvae, perhaps this behavior is actually more common than is currently realized.


Borowiec, L., and J. Świętojańska. 2002–2011. Cassidinae of the world – an interactive manual (Coleoptera: Chrysomelidae). [accessed 3 Dec 2011].

Burgueño, G. 2005. Manejo de la vetación en reservas naturales urbanas de la region metopolitana de Buenos Aires. Aves Argentinas, Asociación Ornitológica del Plata, Proyecto Reservas Naturales Urbanas, 16 pp.

Cuatrecasas, J. 1968. Notas adicionales, taxonómicas y corológicas, sobre Baccharis. Revista de la Academia Colombiana de Ciencias Exactas, Físicas y Naturales 13(50):201–226.

Jolivet, P. 2008. Cycloalexy. In: J. L. Capinera [Ed.], Encyclopedia of Entomology, Springer Science+Business Media B.V.

Jolivet, P., Vasconcellos-Neto, J., and Weinstein, P. 1991. Cycloalexy: A new concept in the larval defense of insects. Insecta Mundi 4(1–4) (1990):133–141.

McFadyen, P. J. 1987. Host-specificity of five Anacassis species [Col.: Chrysomelidae] introduced into Australia for the biological control of Baccharis halimifolia [Compositae]. Entomophaga 32(4):377–379.

Copyright © Ted C. MacRae 2011

13 thoughts on “Cycloalexy in tortoise beetle larvae

  1. All known cycloalexic insects are subsocial in the larval stage and often also exhibit maternal protection of eggs or newly hatched larvae

    What are the criteria for behavior that requires the categories “nonsocial”, “subsocial” and “social”? Does the lunar/tidal timing of coral breeding make them “microsocial” because it is so time limited, while self-insemination of some tapeworms qualify as “nonsocial”?

    Not trying to be an ass, just can’t figure out the check-off list.

    • The classic definition of subsociality involves parental guarding of eggs or nymphs or construction of nests provisioned with food collected by the adult. True sociality (eusociality) involves cooperative brood care with overlapping generations and reproductive division of labor. By those definitions, coral and tapeworms would both be considered nonsocial.

        • As you’ve hinted at, the delimitation between social, subsocial, protosocial, presocial, etc. can be hard to follow. As you say yourself, the classic definition of subsocial involves parental care. In this case, and other I have seen, there is no parental care. These larvae would then be simply communal.

          I would simply call them social. This follows somewhat articles by Costa and Fitzgerald (1996) and Wisclo (1997). Where both argue the use of the prefixes “sub”, “quasi”, “pre”, “proto” is pejorative and leads these insects and behaviours to be studied less.

          – Wcislo, W. (1997) “Are behavioral classifications blinders to studying natural variation?” Chapter 1, pp. 8-13 in The Evolution of Social Behaviour in Insects and Arachnids Edited by Jae C. Choe
          – Costa, J. T. and T. D. Fitzgerald (1996). “Developments in social terminology: semantic battles in a conceptual war.” Trends in Ecology & Evolution 11(7): 285-289

    • Cool – I’ll be interested in hearing if it actually is more prevalent than realized. I doubt the 1st instars of many cassidines have been observed – if they tend to be cycloalexic while young and solitary when older, that would explain a lot (well, and I guess raise a whole bunch of other cool questions as well regarding precisely when the shifts occur, what are the selective pressures, etc.).

      • That’s me, and I actually concentrate on its occurrence in Chrysomelinae, and even though I was looking for it in another subfamily, I did see many tortoise beetle larvae that have it. I wouldn’t be surprised to learn that some species are cycloalexic only when young. My previous study subject, forest tent caterpillars (Malacosoma disstria), are quite gregarious from instar 1 to 3, but instars 5 and 6 are almost solitary. In that case, they are thought to disperse in search of food.

        There are so many species that are known only from adults, no doubt there are many more with circular defence. Many project possibilities on the ecological, physiological and phylogenetic aspects!


Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s