[naturenews] from [nature.com]
[naturenews]
Published online 31 January 2010 | Nature | doi:10.1038/news.2010.45
News
Something rotten in the state of palaeontology
Interpretations of fossil record fail to account for decay.
Daniel Cressey
{{Characteristic features of the amphioxus change during decay.}
Mark Purnell, Rob Sansom, Sarah Gabbott, University of Leicester}
A collection of decomposing bodies is challenging the interpretation of the fossil record, and could force researchers to rethink the evolution of vertebrates.
An assumption underlying the interpretation of many fossils is that as the animals rot they lose their identifying characteristics in a random order, says a team of palaeontologists from the University of Leicester, UK. By studying how animals decay, they have shown that this assumption is flawed.
"The issue is: is a character not present because it was never there or is it that the character was there and had rotted away. In cases where all we have are decay resistant structures we have to be very careful," says study author Mark Purnell.
In a series of experiments published online today in Nature1, Purnell and his colleagues Robert Sansom and Sarah Gabbott found that the features that are most important in deciding where to place an animal on the tree of life are lost first. In practical terms, this means that as something decays it seems to retreat through its own phylogeny and descend the tree (see Nature's video).
This non-random decay may not have been adequately accounted for by some researchers. "In some organisms that have been interpreted there has been way too much speculation and the data have been extended beyond what is scientifically acceptable," says Purnell.
{{“Some fossils have clearly been over-interpreted.”}
Philippe Janvier
National Museum of Natural History, Pari}
s
Philippe Janvier, a palaeontologist at the National Museum of Natural History in Paris, agrees. "Some fossils have clearly been over-interpreted."
The finding is of particular relevance to the fossils of ancient chordates — animals that develop a characteristic rod-like support structure known as a notochord at some point during their lives. These animals have provided the only direct information on the origin of vertebrates. As many of them don't have hard skeletons, our knowledge of them comes mostly from the rare fossils that preserve soft tissue, and interpreting these fossils is fraught with difficulty.
Although their study centres on chordates, the researchers say that a similar process might be occurring in other fossils classified through preserved soft tissue, and preliminary data gathered by Purnell and his colleagues show that this decay bias could be "widespread".
Chordate versus vertebrate
Modern chordates are a broad phylum of animals that includes three subphyla: cephalochordates such as the fish-like lancelets, or amphioxus; urochordates such as sea squirts; and vertebrates such as lamprey and humans. The common ancestors of these three modern groups are known as stem chordates. From these animals evolved the ancestors of the three modern groups, respectively stem cephalochordates, stem urochordates and stem vertebrates.
To get a handle on how these stem animal fossils should be interpreted, Purnell and his colleagues killed adult amphioxus (Branchiostoma lanceolatum) and juvenile lampreys (Lampetra fluviatilis) and studied them at various points during their decay.
The team found that as these modern animals decay, the features that evolved more recently, and therefore distinguished them from their ancestors, rotted first. The last things to decay were features such the notochord.
This means that as they rot, the lamprey and the amphioxus seem to slip from their 'crown' positions at the top of the phylogenetic tree back down to 'stem' versions of their actual species and eventually look like the common ancestor of them all.
The team says that their findings also apply to key fossils that have been described as early chordates — such as Metaspriggina and Cathaymyrus. These fossils cannot be reliably labelled as either chordate or vertebrate 'stem' animals, they say, as they could actually be the decayed remains of either.
Warning signs
Last year, Purnell and Philip Donoghue, a palaeontologist at the University of Bristol, UK, wrote a review paper warning of the dangers of not taking decay into account when interpreting fossils2. This latest evidence supports their concerns, and more care is clearly needed, they say.
"It's certainly going to annoy a lot of palaeontologists who have rather blithely interpreted fossils," says Donoghue. "A bunch of fossils we thought were primitive vertebrates actually now fall into a dustbin and tell us nothing about the evolution of vertebrate characters."
Although Purnell disagrees that there will be any fossils that are completely useless, Janvier also says that some fossils will now "have to just be left aside".
References
1. Sansom, R. S. , Gabbott, S. E. & Purnell, M. A. Nature advance online publication doi:10.1038/nature08745 (2010).
2. Donoghue, P. C. & Purnell, M. A. BioEssays 31, 178-189 (2009).
[naturenews]
Published online 31 January 2010 | Nature | doi:10.1038/news.2010.45
News
Something rotten in the state of palaeontology
Interpretations of fossil record fail to account for decay.
Daniel Cressey
{{Characteristic features of the amphioxus change during decay.}
Mark Purnell, Rob Sansom, Sarah Gabbott, University of Leicester}
A collection of decomposing bodies is challenging the interpretation of the fossil record, and could force researchers to rethink the evolution of vertebrates.
An assumption underlying the interpretation of many fossils is that as the animals rot they lose their identifying characteristics in a random order, says a team of palaeontologists from the University of Leicester, UK. By studying how animals decay, they have shown that this assumption is flawed.
"The issue is: is a character not present because it was never there or is it that the character was there and had rotted away. In cases where all we have are decay resistant structures we have to be very careful," says study author Mark Purnell.
In a series of experiments published online today in Nature1, Purnell and his colleagues Robert Sansom and Sarah Gabbott found that the features that are most important in deciding where to place an animal on the tree of life are lost first. In practical terms, this means that as something decays it seems to retreat through its own phylogeny and descend the tree (see Nature's video).
This non-random decay may not have been adequately accounted for by some researchers. "In some organisms that have been interpreted there has been way too much speculation and the data have been extended beyond what is scientifically acceptable," says Purnell.
{{“Some fossils have clearly been over-interpreted.”}
Philippe Janvier
National Museum of Natural History, Pari}
s
Philippe Janvier, a palaeontologist at the National Museum of Natural History in Paris, agrees. "Some fossils have clearly been over-interpreted."
The finding is of particular relevance to the fossils of ancient chordates — animals that develop a characteristic rod-like support structure known as a notochord at some point during their lives. These animals have provided the only direct information on the origin of vertebrates. As many of them don't have hard skeletons, our knowledge of them comes mostly from the rare fossils that preserve soft tissue, and interpreting these fossils is fraught with difficulty.
Although their study centres on chordates, the researchers say that a similar process might be occurring in other fossils classified through preserved soft tissue, and preliminary data gathered by Purnell and his colleagues show that this decay bias could be "widespread".
Chordate versus vertebrate
Modern chordates are a broad phylum of animals that includes three subphyla: cephalochordates such as the fish-like lancelets, or amphioxus; urochordates such as sea squirts; and vertebrates such as lamprey and humans. The common ancestors of these three modern groups are known as stem chordates. From these animals evolved the ancestors of the three modern groups, respectively stem cephalochordates, stem urochordates and stem vertebrates.
To get a handle on how these stem animal fossils should be interpreted, Purnell and his colleagues killed adult amphioxus (Branchiostoma lanceolatum) and juvenile lampreys (Lampetra fluviatilis) and studied them at various points during their decay.
The team found that as these modern animals decay, the features that evolved more recently, and therefore distinguished them from their ancestors, rotted first. The last things to decay were features such the notochord.
This means that as they rot, the lamprey and the amphioxus seem to slip from their 'crown' positions at the top of the phylogenetic tree back down to 'stem' versions of their actual species and eventually look like the common ancestor of them all.
The team says that their findings also apply to key fossils that have been described as early chordates — such as Metaspriggina and Cathaymyrus. These fossils cannot be reliably labelled as either chordate or vertebrate 'stem' animals, they say, as they could actually be the decayed remains of either.
Warning signs
Last year, Purnell and Philip Donoghue, a palaeontologist at the University of Bristol, UK, wrote a review paper warning of the dangers of not taking decay into account when interpreting fossils2. This latest evidence supports their concerns, and more care is clearly needed, they say.
"It's certainly going to annoy a lot of palaeontologists who have rather blithely interpreted fossils," says Donoghue. "A bunch of fossils we thought were primitive vertebrates actually now fall into a dustbin and tell us nothing about the evolution of vertebrate characters."
Although Purnell disagrees that there will be any fossils that are completely useless, Janvier also says that some fossils will now "have to just be left aside".
References
1. Sansom, R. S. , Gabbott, S. E. & Purnell, M. A. Nature advance online publication doi:10.1038/nature08745 (2010).
2. Donoghue, P. C. & Purnell, M. A. BioEssays 31, 178-189 (2009).
