paraphyly/polyphyly

[Philip Cantino]

Taxacomers,

R. de Jong wrote:
>I have always wondered why some people make such a fuss of distinguishing
>between paraphyly and polyphyly. If a paraphyletic group is defined as a
>group not including all descendants of a common ancestor, so is a
>polyphyletic group. If we agree that an orchid and an elephant have a common
>ancestor (be it far back in history) than the group orchid+elephant is a
>group that does not include all the descendants of their common ancestor.

The distinction between paraphyly and polyphyly is clearer if one adopts
Farris's (1974) definitions (see Table 3.1 of E. O. Wiley's 1981 book,
"Phylogenetics").  These are (slightly paraphrased): A paraphyletic group
includes a common ancestor and some but not all of its descendants.  A
polyphyletic group is one that excludes the most recent common ancestor of
the group.

Using de Jong's example, a group comprising an orchid and an elephant and
nothing else would be polyphyletic, because the most recent common ancestor
is excluded from the group.  To be paraphyletic, the group would have to
include an orchid, and elephant and the extinct and presumably unicellular
eukaryote that is the most recent common ancestor of plants and animals.

There are cases, however, when it is much more difficult to decide whether
a previously circumscribed taxon is paraphyletic or polyphyletic in light
of a new phylogenetic hypothesis--i.e., by delimiting the taxon on a
cladogram.  To make this decision using Farris's definition, one has to
assess whether the immediate common ancestor of the group would be assigned
to the group if it were known.  This assessment is based in part on the
character states that the hypothetical ancestor is inferred to have (by
parsimony-based mapping on the cladogram), but it also sometimes requires
guesswork about whether the taxonomists who recognize the taxon would have
included the ancestor in the group if it existed (alive or preserved as a
fossil).  The latter is necessarily a very subjective decision.  In these
cases, at least, I agree with de Jong that distinguishing paraphyly from
polyphyly has little value, and we are better off just referring to such
groups as nonmonophyletic.

Phil



Philip D. Cantino
Department of Environmental and Plant Biology
Ohio University
Athens, OH 45701-2979
U.S.A.

Phone: (740) 593-1128
Fax: (740) 593-1130
e-mail: cantino at ohio.edu