Re: Sobral and Langer 2008 (was pteros have lift-off)

[David Marjanovic <david.marjanovic@gmx.at>]

> Don't know about how supertrees are made, and do not fully grasp what
> an "average" would mean for a group of trees. Perhaps a majority rule
> consensus or something like that?

Sort of. When supertrees are made by an algorithm rather than by hand (and 
the latter is quite unusual), matrix representation is used: a matrix is 
made with the trees as taxa and the nodes as characters -- each node is 
coded as present or absent for each tree. Then the shortest tree is 
calculated from this matrix.

> I mean, in that study of Langer, is the only reason for pterosaurs
> being most closely related with prolacertiforms and drepanosaurids
> just that this hypothesis of relationship is recovered by most
> studies? I would in principle think not; as far as I know the only
> previous phylogenetic analysis stating that is that by Peters (2000)
> in the Rivista Italiana.

But that's probably the only one that contains any prolacertiforms except 
maybe *Prolacerta*. It could thus outweigh all others in this respect, 
because the others are coded as unknown (inapplicable), rather than absent, 
for, say, a *Cosesaurus*-*Langobardisaurus*-Pterosauria clade.