Rodent phylogeny

[Dorothee HUCHON]

>    There have long been suspicions and suggestions that
>Order Rodentia might be polyphyletic, and Caviomorpha seems
>to be the group most likely in need of being split off from
>rodents.  I was busy with bacteria at the time, but a few
>years ago there was some new evidence published concerning
>this issue, but I don't recall where.  Could someone
>refresh my memory about the caviomorph debate (anything on
>the internet about this?).





Some precision about the rodent relationships.

Rodentia is the most diverse mammalian order, and it phylogeny is still
completely unclear.
        Morphologically the monophyly of Rodents is mainly based on their ability
to gnaw which is characterised by the presence of one incisor by half jaw.
Lagomorph are also gnawing mammals but they possess 2 incisors by half jaw
(2 incisors is the ancestral mammalian state). Other characters supporting
rodent monophyly have been evidenced in Luckett and Hartenberger 1993 (J.
Mammal. Evol. 1 127-147)
If the monophyly of rodent is well accepted from the morphological point of
view relationships between families are completely unclear, especially
because many convergent evolution occur at the morphological level during
the history of the rodents (review in Jeager 1988 in the Phylogeny and
classification of the tetrapods Vol2 mammals (MJ. Benton eds Clarendon
press oxford) 177-199 ). Rodents are currently divided in two suborders the
sciurognathi and the hystricognathi according to the insertion plan of the
incisors relative to the angle of the jaw. Currently the sciurognath state
is ancestral and for this reason it is recognised that sciurognathi may be
a paraphyletic group (i.e. the Ctenodactylidae (the gundi family) have been
proposed to be either the sister clade to all other rodent or the sister
clade of hystricognathi). Hystricognathi monophyly is well accepted but
relationships between families is debated.
        Currently the morphological consensus could be described by a
multifurcation including the 9 main sciurognath clade
(sciuridae+Aplodontidae, Castoridae, Gliridae, muridae,
Geomyidae+heteromyidae, Dipodidae, Pedetidae, Ctenodactylidae,
anomaluridae), and the monophyletic hystricognath. ie. Hartenberger 1998
C.R. Acad. Sci. Paris. Science de la Terre et des planètes 326 pp439-444

Molecular studies could have been a solution to solve the phylogeny of
rodent but instead of solving the uncertainties they complicated the
question suggesting a possible rodent polyphyly. Last complete
mitochondrial studies of rodents (i.e. Reyes et al. 1998 Mol Biol Evol
15:499-505) divide rodent between two clades, a mice and rat clade, and  a
guinea pig and dormice (gliridae Sciurognathi) clade which sister clade is
the rabbit. These molecular studies have been largely criticised because
they include few species and it have been suggested that a larger taxonomic
sampling could lead to evidence the rodent monophyly. Rodent polyphyly is
supposed to be the result of long branch attraction. Currently only few
studies include a large rodent taxonomic sampling (e.g. Nedbal et al. J.
Mammal. Evol 3:201-237, Huchon et al. Mol Biol Evol 16:577-589) but they
failed to evidence either rodent monophyly or paraphyly (rodent families
and mammalian order diversify in the same multifurcation). Currently the
question of rodent phylogeny is still largely debated and will need to be
investigate with longer sequences and large taxonomic sampling.

About rabbit..., this is another complex question.



______________________________________________

Dorothee HUCHON
Lab. de Paleontologie, Paleobiologie et Phylogenie - CC064
Institut des Sciences de l'Evolution (UMR 5554 CNRS)
Universite Montpellier II
Place E. Bataillon
34 095 Montpellier Cedex 5 - France

Tel. = 33 4 67 14 48 63                  ____()
Fax = 33 4 67 14 36 10              ____|_____°\#
Email = huchon at isem.univ-montp2.fr        ^  ^ #
____________________________________________