In the Sierra Nevada Mountains of California, bioluminescent millipedes scatter upon the forest floor and on a moonless night, “resemble a starry sky” (Tiemann, 1969). The most remarkable feature of these millipedes is their ability to glow at a wavelength of 495 nm. This is only one of two known instances of bioluminescence in the entire millipede class Diplopoda. Bioluminescence in millipedes is restricted to only eight species of the genus Motyxia, which are endemic to a very small area in the Sierra Nevada Mountains, and Paraspirobolus lucifugus (a distantly related species in the order Spirobolida) from Japan, Taiwan and the Federated States of Micronesia.

The function of luminescence in Motyxia remained uncertain. Authors speculated that the role of the emitted light may be a nocturnal warning signal to announce the presence of a cyanide-based chemical defense, as reflected color does in closely related millipedes in Appalachia. Still, others suggested that the luminescence serves no adaptive function.

Our current research program aims to discover the evolutionary circumstances under which this unique adaptive innovation arose. The first part of the project consists of molecular phylogenetics as a foundation to address monophyly of the genus, and as a basis for new species descriptions. Did luminescence evolve multiple times, and under what circumstances? The second part of the project is a field test of luminescence and whether it functions as a nocturnal warning signal. Currently, we are using next generation transcriptome sequencing to understand the DNA-level differences between luminescent and closely related non-luminescent taxa.

[Above, the brightest species Motyxia sequoiae photographed in its own light (174 s exposure) from Giant Sequoia National Monument, California.]

Molecular phylogenetics of Motyxia and relatives (Xystodesmidae)

^ Phylogeny of Motyxia species and close relatives. Bioluminescence has a single evolutionary origin in the millipede order Polydesmida (arrow).

In this figure, starbursts indicate presence and relative intensity of bioluminescence, circles denote absence of bioluminescence.  Ancestral origin of bioluminescence is indicated by a solid arrow at the base of the Motyxia clade. Weak bioluminescent intensity of Motyxia pior emphasized with a triangle. Motyxia sequoiae display the brightest bioluminescence of the genus. Phylogeny estimated with every known species of Motyxia (spare one) and representatives from five closely related tribes.

Testing bioluminescent aposematism in millipedes

^ Results of a field experiment to test the prediction that luminescent millipedes are attacked less often than non-luminescent millipedes.

In this figure, the bar graphs show the experiment’s results: the proportion of millipedes attacked versus luminescence. Non-luminescent millipedes were attacked more often than luminescent millipedes. Images show rodent incisor marks in clay millipedes, and live millipede (arrow) with anterior segments 1–14 missing after predator attack.

Figures adapted from: Marek et al. (2011) Bioluminescent aposematism in millipedes. Current Biology, 21, R680-R681.    [Open access]

This research is supported by a National Science Foundation Phylogenetic Systematics grant to Paul Marek (DEB-1119179)

Found a bioluminescent millipede? Found.html

Bioluminescent millipedes

Visit our new home at Virginia Tech @

Aposematism is a biological phenomenon where a feature (appearance, sound, smell) deters predators because it denotes something noxious, or unpleasant. Aposematism has long fascinated naturalists and biologists because it provides a straightforward and elegant example of evolution by natural selection.

E.B. Poulton coined “aposematism” in the book The Colours of Animals in 1890.  Any signal can be aposematic, and it doesn't have to be garish, or even conspicuous. However often times in nature, aposematic signals are conspicuous. There's empirical evidence that suggests conspicuous aposematic signals work better because they are quickly learned, easy to remember, and so different looking as to reduce errors in discriminating against edible prey. Learning (at some point in establishing avoidance) is a significant component of aposematism. Predators, over time, learn that some feature denotes something unpleasant. And learning a signal that is associated with something unpleasant is very rapid. In some instances, after some time, avoidance becomes fixed. This is referred to as innate avoidance. (It's been suggested that humans have an innate, or unlearned, avoidance to snakes as a result of an ancient evolutionary association with them.)

We study blind aposematically-colored millipedes, like the one shown above, which provide an ideal model system to investigate the ecological role of aposematism since their appearance is only viewed by other organisms, for example their predators.

Here is an outline research projects that I currently lead, and which reflect our research program’s focus on the evolutionary ecology of warning coloration and mimicry.

Above, Apheloria montana from Virginia generates cyanide to defend itself. Each individual can secrete 18 times the amount necessary to kill a pigeon-sized bird. Apheloria montana displays conspicuous black and yellow coloration to warn predators to stay away.