Published online 17 August 2010 | Nature | doi:10.1038/news.2010.415


Attack of the ancient 'zombie' ants

Fossil leaf bears the telltale scars of insects infected by parasitic fungus.

Ant parasitised by fungusZombie ants have been around for millions of years.Hughes, D. P. et al

Researchers claim to have found the first evidence of 'zombie' ants in the fossil record. They have matched peculiar cuts on a 48-million-year-old fossil leaf with the 'death bites' made by modern ants infected by a fungal parasite. The research is published today in Biology Letters1.

The leaf was part of a group of fossilized leaves and plants unearthed recently from the Messel Pit in Germany's Rhine Rift Valley — an area famous for the discovery, in 2009, of Ida, a well-preserved primate fossil touted as a human ancestor (see 'Reunion of fossil halves splits scientists'). Initially, the fossil plants and leaves did not raise much interest and they were stored for years at the Smithsonian Institution's National Museum of Natural History in Washington DC.

The idea of examining the fossil record for evidence of the distinctive bite marks came to David Hughes, a behavioural ecologist at Harvard University in Cambridge, Massachusetts, as he was sitting in on a palaeobotany undergraduate course. He had just returned from fieldwork in southern Thailand where he had been studying fungal parasites infecting carpenter ants and controlling their behaviour. "Could this parasitic relationship have evolved much earlier in Earth's history?" he asked himself.

So Hughes talked to Conrad Labandeira, a palaeoecologist at the Smithsonian's National Museum of Natural History. It turned out that Labandeira had seen strange markings on a fossil leaf and had been puzzled by the specimen for years.

"This was a serendipitous discovery when a project on modern insect ecology crossed paths with a long-running palaeontology programme on the Messel Shale," says Labandeira.

Fossilised leaf with zombie ant 'death bites'Markings (bottom) on fossil leaves are likely to be the 'death bites' of ants infected by a fungal parasite.Hughes, D. P. et al

Deadly relationship

In Thailand, Hughes had seen leaf bites produced by carpenter ants, Camponotus leonardi, that were infected and manipulated by a fungus, Ophiocordyceps unilateralis2. In this parasite–host relationship, the fungus manipulates the insect's behaviour so that the ant bites into the underside of a leaf. The conditions under the carefully chosen leaves are optimal for fungal growth. After the ant bites into a leaf vein, the fungus grows rapidly, covering the ant in a dense mat of filamentous fungal hyphae. The fungus then releases spores to infect more ants.

Hughes doesn't think the bites could be anything else — such as the vein-cutting behaviour exhibited by some other ants or beetles during feeding — because the location on the leaf vein and shape of the marks are so unusual. "It is not normal ant behaviour to bite into the leaf vein because it has no real nutritional value to the ant and can in fact be toxic in some plant species," he says.

Other researchers agree. "The distinctive pattern of the bite marks and their association with the veins of the leaf is actually quite striking and unusual to see in fossil leaves," says Paul Kenrick, head of research at the Natural History Museum's Department of Palaeontology in London. "This find is telling us something significant about the evolution of the interactions between organisms and their dependencies." If the marks are those left by dying ants, the complex ant–parasite relationship must have evolved at least 48 million years ago, he says.

"The hunt is on to search the Messel Shale assemblage for other examples of these bite marks" says Labandeira, "There may be plenty of other specimens from this site in research institutes across the world just waiting to be viewed by a fresh pair of expert eyes." 

Kate Larkin is a British Science Association Media Fellow 2010.

  • References

    1. Hughes, D. P., Wappler, T. & Labandeira, C. C. Biol. Lett. advance online publication doi:10.1098/rsbl.2010.0521 (2010).
    2. Andersen, S. B. et al. Am. Nat. 174, 424-433 (2009). | Article | PubMed


If you find something abusive or inappropriate or which does not otherwise comply with our Terms or Community Guidelines, please select the relevant 'Report this comment' link.

Comments on this thread are vetted after posting.

Commenting is now closed.