Ida returns. "Missing Link" not so linky.
Remember Ida? (or Darwinius massillae to give her full name)? Remember how six months ago she was unveiled to the world in a whirlwind of hype? Remember how the blogging community despaired of this, while noting what a lovely fossil she was?
This week more research has been published, after the discovery of Afradapis, a related species from the Late Eocene of Egypt. Not nearly as nice as Ida its a jaw and teeth. This is not unusual- most of our ideas about mammalian evolution were based on teeth until genetics came along and morphology had to play catch-up. And when dealing with extinct groups such as adapids its the best thing we have going- teeth are the most robust parts of your body, and withstand the processes of decay and weathering the best. The fossils shows some features present in the catarrhine primates (the group that include apes and Old-World monkeys). Which is very interesting. However...
...all is not as it seems. The team did a phylogenetic analysis, of 117 living and extinct species, using 360 characters. This is only about 3 characters per taxa, and I'm sure more would be nice- but you can guarantee that not all characters will be preserved in each group (its rather hard to say what features are present in the ankle of a genus only known from its teeth for example), and its the first phylogenetic analysis I'm aware of that includes Darwinius which is a very Good Thing. Sadly for most people, its published in Nature, which means anyone who doesn't have a subscription only gets the abstract and whatever mangling the newspapers put this and the press release through.
This analysis shows that Ida, Afradapis and the other Adapoids are not stem anthropoids, and are not even haplorhines (the group that includes monkeys [and apes] and tarsiers). In fact they fall out as the sister group to the living strepsirrhines (lemurs, lorises, and bushbabies). To me thats far more interesting than if the adapids were on the line to humans, because it means the features they share with us either evolved twice (once in anthropoids and once in adapids- the position the paper's authors take) or are primitive to a wider group of primates. Without having read the paper I can't say why the author's favour one over the other.
ETA: Brian Switek over on Laelaps has a very nice post, (with a lovely drawing of Afradapis' jaw). The features that link adapids and anthropoids (but are not present in the earliest anthropoids) are details of the mandibular symphysis (the joint between the two halves of the lower jaw) and the loss of the second premolar. Neither of these features are likely to be lost and re-evolved, when evolution discards things they tend to stay discarded, and when they come back, they're usually subtley different.
Before we start questioning how effective the analysis was- and believe me scientists will, plugging more species in, pulling others out, re-examining specimens, poking holes everywhere they can to see how good this tree really is, taking the analysis at face value raises a whole series of interesting questions. If we're seeing convergence between anthropoids and adapids, then what were these two different groups doing that was similar, and if the features we are seeing are primitive, then why did lemurs and other groups lose them.
And this to me is the beauty of science. When scientists are wrong it is very often for interesting reasons- lack of data, working within a theory that could not account for new observations, etc. And when they are shown to be wrong, a whole vista of new questions open up, new ways to be interestingly wrong, but at the same time inch closer to a greater understanding of the universe. And people wonder why I love science so damn much.
Seiffert, E., Perry, J., Simons, E., & Boyer, D. (2009). Convergent evolution of anthropoid-like adaptations in Eocene adapiform primates Nature, 461 (7267), 1118-1121
I recommend reading this interview with Seiffert.
This week more research has been published, after the discovery of Afradapis, a related species from the Late Eocene of Egypt. Not nearly as nice as Ida its a jaw and teeth. This is not unusual- most of our ideas about mammalian evolution were based on teeth until genetics came along and morphology had to play catch-up. And when dealing with extinct groups such as adapids its the best thing we have going- teeth are the most robust parts of your body, and withstand the processes of decay and weathering the best. The fossils shows some features present in the catarrhine primates (the group that include apes and Old-World monkeys). Which is very interesting. However...
...all is not as it seems. The team did a phylogenetic analysis, of 117 living and extinct species, using 360 characters. This is only about 3 characters per taxa, and I'm sure more would be nice- but you can guarantee that not all characters will be preserved in each group (its rather hard to say what features are present in the ankle of a genus only known from its teeth for example), and its the first phylogenetic analysis I'm aware of that includes Darwinius which is a very Good Thing. Sadly for most people, its published in Nature, which means anyone who doesn't have a subscription only gets the abstract and whatever mangling the newspapers put this and the press release through.
This analysis shows that Ida, Afradapis and the other Adapoids are not stem anthropoids, and are not even haplorhines (the group that includes monkeys [and apes] and tarsiers). In fact they fall out as the sister group to the living strepsirrhines (lemurs, lorises, and bushbabies). To me thats far more interesting than if the adapids were on the line to humans, because it means the features they share with us either evolved twice (once in anthropoids and once in adapids- the position the paper's authors take) or are primitive to a wider group of primates. Without having read the paper I can't say why the author's favour one over the other.
ETA: Brian Switek over on Laelaps has a very nice post, (with a lovely drawing of Afradapis' jaw). The features that link adapids and anthropoids (but are not present in the earliest anthropoids) are details of the mandibular symphysis (the joint between the two halves of the lower jaw) and the loss of the second premolar. Neither of these features are likely to be lost and re-evolved, when evolution discards things they tend to stay discarded, and when they come back, they're usually subtley different.
Before we start questioning how effective the analysis was- and believe me scientists will, plugging more species in, pulling others out, re-examining specimens, poking holes everywhere they can to see how good this tree really is, taking the analysis at face value raises a whole series of interesting questions. If we're seeing convergence between anthropoids and adapids, then what were these two different groups doing that was similar, and if the features we are seeing are primitive, then why did lemurs and other groups lose them.
And this to me is the beauty of science. When scientists are wrong it is very often for interesting reasons- lack of data, working within a theory that could not account for new observations, etc. And when they are shown to be wrong, a whole vista of new questions open up, new ways to be interestingly wrong, but at the same time inch closer to a greater understanding of the universe. And people wonder why I love science so damn much.
Seiffert, E., Perry, J., Simons, E., & Boyer, D. (2009). Convergent evolution of anthropoid-like adaptations in Eocene adapiform primates Nature, 461 (7267), 1118-1121
I recommend reading this interview with Seiffert.