Skip Navigation

Spider webs untangle evolution

November 1, 2004 By Roxanne Khamsi This article courtesy of Nature News.

Similarity of construction shows 'convergent evolution' applies to behaviour.

Please log in to rate this page.

View Comments

The biologist Stephen Jay Gould famously proposed that if we could "rewind the tape" of evolution and play it again, chance would give rise to a world that was completely different from the one we live in now. But the concept that chance reigns supreme may ring less true when it comes to complex behaviours.

A study of the similarities between the webs of different spider species in Hawaii provides fresh evidence that behavioural tendencies can actually evolve rather predictably, even in widely separated places.

Todd Blackledge of the University of California, Riverside, and Rosemary Gillespie, of the University of California, Berkeley, studied species of Tetragnatha spiders on different Hawaiian islands. The spiders' webs vary significantly, with tissue-like 'sheet webs', disorganized cobwebs and spiral-shaped 'orb webs' as three of the most common types.

Each species had its own characteristic type of web. But the scientists found that in several cases, separate species of Tetragnatha spiders on different islands constructed extremely similar orb webs, right down to the number of spokes, and the lengths and densities of the sticky spiral that captures bugs.

Was this an example of similar environments producing the same complex behaviour, or did the spiders with corresponding webs share a common ancestor?

Branching out

To find out, Blackledge and Gillespie turned to evolutionary trees. They constructed a hypothetical tree in which each type of web evolved only once, so all the spiders weaving that web type were related. Then they compared it with the evolutionary tree suggested by the spiders' DNA.

The tree that linked spiders through their web-constructing behaviour proved "highly improbable", the researchers found, as it was very complicated, and contradicted the relationships suggested by their DNA. "There were too many extra steps," says Blackledge.

The researchers conclude that the web types must have evolved independently, driven by matching environmental conditions on the different islands. They publish their results in the Proceedings of the National Academy of Sciences1.

"The next step would be to figure out what those environmental factors are," says Blackledge. He suggests that the type of prey the spiders are trying to catch is likely to be a major factor.

"It's likely that similar forest types support similar mixes of prey, which could elicit similar web structures," agrees Jonathan Coddington, an entomologist with the Smithsonian Institution's National Museum of Natural History.

Behavioural spin

Previous research has found that physical traits, for example legs or wings, can arise independently in similar environmental conditions. And various groups have looked at the evolution of simple behaviours, such as where species locate themselves within a habitat, like a branch or lake.

"I and others have written about many examples of convergent evolution, which is when you play the tape of life several times and get the same result," says Paul Harvey, who heads the zoology department at the University of Oxford.

But the evolution of complex behaviours is less well understood. The researchers believe that predictable evolutionary convergence of behaviour applies far beyond spiders, and happens more often then some believe. "I would say it's fairly common," Blackledge says, adding that island ecosystems will be the easiest place for scientists to look.

Whether or not the spider-web discovery becomes the strongest case for predictable behavioural evolution depends on how common the phenomenon is. As Harvey explains: "Some become textbook examples and some don't."


  1. Blackledge T. A. & Gillespie R. G. Proc. Natl Acad. Sci. USA , published online doi:10.1073/pnas.0407395101 (2004).


User Tools [+] Expand

User Tools [-] Collapse

Pinterest button


Please log in to add this page to your favorites list.

Need Assistance?

If you need help or have a question please use the links below to help resolve your problem.