Nature Blog Network

Wednesday, October 24, 2007

Firefly Double Whammy: Evolution and Costs of Light

Um... no, not that firefly, although suspenders should totally make a comeback... NOT! How about this one:

Photo from Encyclopaedia Britannica.

Author's note: Fireflies will always hold a special place in my heart. Every summer when I was kid I would sit outside in Iowa and watch them, collecting them in my hands so I could peek inside with one eye to see if the lightning bug (as we called them in Midwest) was flashing. I look forward even more now to the day when my children can stay up late enough to watch and collect fireflies themselves. This post is dedicated to my beautiful swedish wife, who discovered fireflies for the first time when we moved east out of California 3 years ago. It is so great to see how excited she gets when they come out each summer!

Blogging on Peer-Reviewed Research
Fireflies are considered all part of the family of beetles called Lampyridae. Appropriately, this means light. The luminescence is the result of a chemical reaction with the substrate Luciferin and the enzyme Luciferase. The happens in two steps and is energetically expensive (i.e. requires ATP and oxygen). This is a slow reaction, which is why we can see it decay after the initial flash, that is controlled in specialized organ in its abdomen, called the lantern naturally. It sometimes is mistakingly thought of as a bacterial symbiosis like in many marine organisms, but for fireflies (along with many other bioluminescent critters) this is purely a chemical phenomenon.

There are about 2000 species of Lampyrids globally, with about 120 of those in North America. Stanger-Hall et al. recently reported the first phylogeny of North American fireflies in the journal Molecular Evolution and Phylogenetics. Their study also has interesting results for the evolution of light signals in this family. First the basics though, which also have importnat implications. They used nuclear (18S) and mitochondrial genes (16S and COI) to form consensus trees of 27 species of North American fireflies from 17 genera, including one species and genus that was recently placed outside of the family Lampyridae (Pterotus). They found that, in fact, North American fireflies do not form a monophyletic group. This lends support to the hypothesis that firefly diversity has been helped by multiple invasions. Furthermore, they determined that the present taxonomic classification into subfamilies and tribes is not supported by the molecular data, while Pterotus and another genus placed outside of the family are actually nested within the Lampyridae.

Figure 1 from Stanger-Hall et al. 2007. Carbon dust drawings done by Laura Line.

The real interesting result of this study has to do with evolution of the signaling. Branham & Wenzel (2003) report that lampyrid larvae possess of constant, but faint glow in larval light organs on the abdomen. These authors suggest, based morphological character analysis, that light production evolved early, predating the Lampyridae, and was retained from the larval form. This is supported by observational evidence that adults vary widely in light production and light organ placement and use. Typically, lampyrids use their light displays as sexual signals to attract a mate. But more basal family members use pheromones. Citing a chapter by Lloyd (1997), Stanger-Hall et al. describe 3 mating signal systems in the 120 North American species of firefly:
(1) Chemical signals (pheromones): ‘‘dark fireflies’’ (e.g. Ellychnia, Pyropyga, Lucidota) produce no light as adults and are active during the day; they release chemical signals to attract mates. (2) Glows (continuous light signals): ‘‘glowworm fireflies’’ (e.g. Microphotus, Phausis, Pleotomodes) tend to have larvae-like females who spend the day in underground burrows and emerge at night, emitting a continuous glow. This glow (short distance) in combination with pheromones (long distance) attracts males who will fly towards the glow, but usually do not signal themselves. (3) Flashes (short intermittent light signals): ‘‘lightningbug fireflies’’ (e.g. Photinus, Photuris, Pyractomena) are the most commonly observed. They are active at dusk or in the dark and both males and females use species-specific light signals to communicate with each other in an interactive visual morse-code that identifies the species and the sex of the signaler. Some genera (e.g. Pleotomus) and individual species within genera (e.g. Phausis reticulata) may represent intermediate stages in signal evolution (e.g. Pleotomus males glow when disturbed).
In terms of the evolution of signaling, they found no clear patterns. Whether they used flashes, glows or pheromones all the species in the phylogenetic tree were intermingled.
If you click on the above figures from Stanger-Hall et al., you can view them larger but at this size the take-home message is clear. The tree on the top is color-coded by sexual signal modes. Green signifies flashes, orange is glow, grey means using pheromones and weak glows, while black is only pheromones. It is obvious that glows and flashes have multiple origins in the North American firefly fauna.

The tree on the bottom is the same tree, but this time orange branches are for both flashes and glows and black branches are for pheromones only. The asterisks denoted light signal origins (orange) or losses (black) while the letters A & B represent two possible evolutionary scenarios.
"Scenario A, light signals originated once in ancestral adult lampyrids, and were subsequently lost nine times. Scenario B, ancestral lampyrids used pheromones as sexual signal, and the transition to sexual light signals evolved four times independently, followed by four losses. There are at least two other possible 10-step scenarios (multiple gains and losses), but neither is favored by any weighting where losses are considered as likely or more likely than gains. The color-coding of the branches reflects scenario B."
Video firefly larva from Thailand. Note the constant glowing in the posterior segment.

This all provides evidence that supports the hypothesis that the lampyid fauna of North America has invaded the continent multiple times with multiple origins (i.e. Europe or Asia). While scenario A allows for only a single origin of light production followed by 9 losses, scenario B is more parsimonious - requiring fewer steps. It seems to me, and I think there are papers out there that might back it up, that new gains or more rarer than losses. It would be interesting to see a meta-analysis of studies that combined morphology and molecular character data to study if indeed parsimony won out over intuition. If anyone knows of such studies, drop me a line.

Another study published in American Naturalist last month by Woods et al. 2007 studied the energetic costs and the risk of predation associated light production. This video that I couldn't figure out how to embed (might need subscriber access) describes their research (contributed by the authors as additional material with their paper). They measured energetic costs using open flow respirometry, which measures carbon dioxide production, during flashes and when at rest. They found that individual fireflies did have a significant increase in metabolic rate during light production compared to being at rest, but it was less than the rate when they were walking and not flashing. To test the hypothesis that maintaining just the bioluminescent capability has energetic costs, they also tested two other lampyrid species that are diurnal and do not produce light. Standardizing for body, they found no significant differences in the metabolic rate of lampyrids that are capable of producing light and those that have no such capability.

The next step was to see if light production incurred any costs in terms of increased predation. Woods et al. 2007 set up a experiment with arrays of sticky-trap cups with flashing LEDs that simulated the mating signals of Photinus greeni and sticky-trap cups without any light (Figure 1b from Woods et al. 2007, left). P. greeni is chemically defended against would-be predators by synthesizing steroidal pyrones. But there is one predator that is unbothered by this defense and in fact sequesters the compound for use in its own defense. This is another species of firefly called Photuris versicolor that hunts Photinus species by tracking their mating signals and captuing grounded males in the midst of getting it in on. In fact out 218 individuals of Photuris trapped, only 4 were caught on non-flashing traps! Interestingly, 96% of all trapped Photuris were females.

The take home message? Its not energetically expensive to make flash your stuff but you'll attract the wrong company at times!
"Every single night, male fireflies are out there flying a fine line between sex and death. For us, it definitely rivals the most exciting television thriller! So, next time you're outside on a summer night take a moment to admire the firefly romance and risk that’s playing out all around you."-Sara Lewis, Professor of Biology at Tufts University (quoted from press release on

Branham, M.A., J.W. Wenzel (2003). The origin of photic behavior and the evolution of sexual communication in fireflies (Coleoptera: Lampyridae). Cladistics 19(1), 1–22. doi:10.1111/j.1096-0031.2003.tb00404.x

Lloyd, J.E. (1997). Firefly mating ecology, selection and evolution. In: Choe, J.C., Crespi, B.J. (Eds.), Evolution of Mating Systems in Insects and Arachnids. Cambridge University Press, London, pp. 184–192.

Stanger-Hall, K.F., J.E. Lloyd, D.M. Hillis. (2007) Phylogeny of North American fireflies (Coleoptera: Lampyridae): Implications for the evolution of light signals. Molecular Phylogenetics and Evolution, 45, 33-49. doi:10.1016/j.ympev.2007.05.013

Woods Jr., W. A., H. Hendrickson, J. Mason, S.M. Lewis. (2007) Energy and predation costs of firefly courtship signals. American Naturalist, 170, 702-708.

1 comment:

  1. Hi, nice firefly video you have there. I am linking it to

    FYI.It is a firefly larva or a larivform female of the Lamprigera sp. Not a trilobite beetle.



Note: Only a member of this blog may post a comment.