A recent study of the Cambrian food web of 500 million years ago found that predators and prey organized themselves in much the same way back then as they do today. As ScienceDaily put it,

Similarities between half-billion-year-old and recent food webs point to deep principles underpinning the structure of ecological relationships, as shown by researchers from the Santa Fe Institute, Microsoft Research Cambridge and elsewhere. Analyses of Chengjiang and Burgess Shale food-web data suggest that most, but not all, aspects of the trophic structure of modern ecosystems were in place over a half-billion years ago. It was an Anomalocaris-eat-trilobite world, filled with species like nothing on today's Earth. But the ecology of Cambrian communities was remarkably modern, say researchers behind the first study to reconstruct detailed food webs for ancient ecosystems. Their paper suggests that networks of feeding relationships among marine species that lived hundreds of millions of years ago are remarkably similar to those of today. (Science Daily, May 1, 2008)

Lead researcher Jennifer Dunne identifies the shape of what we do not yet know:

... why food webs from different habitats and across deep time share so many regularities. It could be that species-level evolution leads to stable community-level patterns, for example by limiting the number of species with many predators through selective pressures that result in extinctions or development of predator defences. Or, patterns may reflect dynamically persistent configurations of many interacting species, or fundamental physical constraints on how resources flow through ecological networks.

These findings invigorate a long-standing debate about whether the history of life is constrained by inevitable patterns or whether it is largely shaped by chance events.

American paleontologist Stephen Jay Gould was fond of saying that if the "tape" of evolution were played again it would not necessarily produce anything like the life forms we see today.

Findings like those of Dunne's team suggest that Gould might be mistaken. Some patterns in the history of life seem do to recur regularly. Biochemist Michael Denton, for example, notes, "Curiously, the evolution tape has been played again, at least in part. This has occurred on several occasions." In Nature's Destiny he point sto the fact that marsupial and placental mammals developed pretty much the same ecology of life forms over time.

As Rob Kries of Microsoft Research News and Highlights explains, researcher Rich Williams and his colleagues (like Dunne) "understand that the acclaim with which their findings are greeted is unlikely to be unanimous." But Williams hopes that other researchers will come round in time because his team's study

... builds on a message from some ecologists that there are basic principles underlying the regularities observed across ecosystems,” Williams notes. “Of course, there’s also a huge amount of variability on top of that. Some of us focus on the underlying regularities, others focus on the system-specific details. We can sense conflict between those two ways of thinking, which is unfortunate, because I see them as complementary.

Article, citation, and resources:

The article is here.

Citation: Dunne JA, Williams RJ, Martinez ND, Wood RA, Erwin DH (2008) Compilation and network analyses of Cambrian food webs. PLoS Biol 6(4): e102. doi:10.1371/journal.pbio.0060102

Other resources

A graphic of the Cambrian food web is here, and a feature article about the research here.

Other Cambrian explosion stories from Design of Life blog:

Critical Decisions in Science: The Smithsonian secretary vs. the Cambrian explosion

Biology's Big Bangs

Questions in evolution: Animals suddenly appear ... and after that nothing much happens ... why?

Also: The Avalon Explosion: The dawn of life reveals another intricate puzzle