[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Re: Are dinosaurs really reptiles?

Both synapsids and reptiles (sauropsids according to some) are known from
the Upper Carboniferous at around (if I remember correctly) 310-315 million
years ago. 

About 312, close to the base of the Upper Carboniferous.

They were distinct lineages by then, so the split happened by
that time at the latest.

At the very latest, because the sauropsid with that age is *Hylonomus*, which is probably more closely related to the diapsids than even the turtles are (see tree posted a few days ago). The theropsid (synapsid) with the same age, *Protoclepsydrops*, is so poorly preserved could be anything; the next youngest theropsid (*Archaeothyris*) is already an ophiacodontid, so not right at the base either. In other words, the origin of amniotes is several branching points removed from these fossils.

The Scottish sites from the middle Early Carboniferous have not yielded amniotes so far, despite containing the terrestrial-looking *Lethiscus*, *Westlothiana* and *Casineria*, so that probably means that there weren't any amniotes 335 Ma ago yet. That said, *Casineria* -- originally, like *Westlothiana* before it, announced as the world's oldest amniote -- is poorly preserved (most notably, the skull is missing), has only received a brief description, and has IIRC never been included in a phylogenetic analysis!

The closest relatives of Amniota -- *Solenodonsaurus* and Diadectomorpha -- are only known from layers younger than the oldest known amniotes. So, they don't help either. The next closest relatives are the lepospondyls, which are common 335 Ma ago (*Lethiscus* and most likely *Westlothiana* belong there).

In short, whenever you see a molecular divergence date estimation paper that uses the origin of amniotes as a calibration point, you know the authors don't know what they are talking about. For reasons of tradition (copying from each other and ultimately from Kumar & Hedges 1998, AFAIK), this calibration point is immensely popular, despite being rubbish.