Morphometric Data - An Invariant Approach to Statistical Analysis of Shapes - page 25

homologous should be reserved for the analysis of features that are
shared due to common descent (see below). We suggest the term corre-
spondence be substituted for all other relationships so as not to confuse
phylogeny with other bases for comparison. Phylogenetic history is
known for relatively few organisms, and other important issues need
to be evaluated (see below). We offer the following outline of useful
terms as a tentative nomenclature and a basis for further discussion.
The proposed categories of landmark correspondence are not mutually
exclusive.
a) Homologous landmarks
Homologous landmarks are those landmarks chosen to represent fea-
tures on organisms that are similar due to a phylogenetic relationship.
The organisms being compared thus share a common ancestor, and the
feature under study is present in all organisms under consideration
due to each inheriting it from the common ancestor. This meaning
refers to the phylogenetic, evolutionary, or historical definition of
homology (Wagner, 1989). If the organisms being compared represent
members of a single species, then corresponding landmarks are homol-
ogous by definition because all members of a species descended from a
common ancestor. But if organisms from different species are being
compared, the landmarks used can only be considered homologous if
the phylogenetic history of the organisms is known. Lack of this type
of information forces the researcher to consider alternate terms when
referring to landmarks (see Wagner, 1989; VanValen, 1982; Muller and
Wagner, 1996).
b) Structurally corresponding landmarks
Structurally corresponding landmarks are those landmarks chosen to
represent structurally similar features on organisms. Structural corre-
spondence of parts is relatively easy to establish morphologically and
is a more conservative definition of relationships, especially when the
phylogenetic history of the organisms under consideration is not
known. An example would be a comparison of the femur of a dog and
the femur of a dinosaur; two organisms separated greatly in terms of
evolutionary time and phylogeny, but with limbs that show basic struc-
tural correspondence. Femora of these two taxa share many
distinguishing features, and comparable landmark data can be collect-
ed from the two species for that reason. Landmarks identified on