The straw that broke Malpighiales' back

[Ken Kinman]

John,
      Although my angiosperm posts are mainly directed at the botanical community, I guess I should explain to the zoologists how I am approaching angiosperm phylogeny (and classification).  I generally approve of the APG's (Angiosperm Phylogeny Group's) approach, compared to the overly split classifications of Cronquist, Takhtajan, Thorne, Dahlgren, and Reveal.  However, in a few cases I think APG has gone overboard on some of their biggest "Orders" (and this probably alienates a lot of non-molecular botanists).  My goal has been to start with APG and reverse some of their overlumping, so that there will be at least one classification which bridges the vast gulf between those which are overlumped (APG) and the overly-split ones which many non-molecularists use.  Therefore I only list families in cases where my ordinal groupings differ from APG.

       As for the "Malpighialean" grouping in particular, it seems to be their worst case of overlumping (to the point it is probably paraphyletic or even polyphyletic), and I can only hope APG will split it even IF it is holophyletic (which I seriously doubt).  None of the characters listed at the Angiosperm Phylogeny Website (link below) for their "Malpighiales" seems particularly unique or helpful in supporting such a grouping.  The Cymothoe statement is false, since that genus also has hosts well outside of their Malpighiales (in the asterid Order Lamiales).

      I proposed what families I presently plan to include in each of the 5 separate Orders (Linales, Violales, Euphorbiales, Malpighiales, and Hypericales), and within that context I am examining characters that would group some of these to each other and to Orders outside of the "Malpighialean" grouping.  "Rosid" interrelationships are pretty obscure and messy (which is why I left those Orders uncoded in my classification here on taxacom last May; see archives), but I like a challenge so this is what I am directing my biological time towards right now.  My prediction is that the "Malpighialean" grouping will turn out to be a highly paraphyletic "grade" near the base of the Rosids, and that any molecular or morphological similarities in this "grade" as a whole are plesiomorphic.  We shall see.  Anyway, attempts to cram Rafflesiales into that already messy grouping was just too much for me to ignore the problem any longer.
               -------- Cheers,
                           Ken Kinman
Link to Angiosperm Phylogeny Website:

http://www.mobot.org/MOBOT/Research/APweb/