Homology and Homoplasy 437later, Mivart [1871] had withdrawn his opposition, cited Lankester’s homogeny
and homoplasy favorably, and reduced his own classes of homology to three.^16
7.1 Homoplasy vs Homology
Homoplasy is now contrasted with homology. “Homology and homoplasy are terms
that travel together; homoplasy being close to, but not quite, the inverse of ho-
mology. If homology is “the same thing”... homoplasy is theappearance of
“sameness” that results from independent evolution” [Wake, 1996, xvii].
The common basis for considering features as homoplastic is their independent
evolution one from the other. However, homoplasy is a portmanteau term for
classes of similarity otherwise subsumed under terms such as convergence, par-
allelisms, reversals, rudiments, vestiges and atavisms.^17 But these terms include
features that may or may not be present (i) in all individuals within a popula-
tion, all populations of a species, or all species within a genus, (ii) in all or any
individuals of the most recent ancestor or more basal taxa, (iii) as fully functional
structures or behaviours, or (iv) in taxa that share a common evolutionary history.
This is as wide a set of categories as you will find anywhere in biology. Does homo-
plasy really encompass them all or is our understanding of homoplasy misplaced,
incomplete or perhaps even wrong headed?
In the balance of this entry I outline in brief the classes of similarity that tradi-
tionally are grouped as homoplasy, these being convergence, parallelism, reversals,
atavisms, rudiments and vestiges (see [Hall, 2003a] for a more in depth discus-
sion and examples). I then align these classes of homoplasy with homology and
demonstrate the developmental and phylogenetic continuities that underlie the
apparently dichotomous categories homology and homoplasy.
8 CONVERGENCEConvergence is the appearance of similar features in independent lineages. ‘In-
dependent’ is usually taken to mean not having a shared recent ancestry, leaving
both independent and recent ill-defined. Consequently, convergence is often de-
fined operationally by an example, such as the evolution of insect, bird and bat
wings [Shubinet al., 1997; Hall, 2005a].
Because they are independently evolved, we expect convergent featuresnot
to share similar developmental pathways or mechanisms. Given the single tree
(^16) Others proffered other terms to deal with the application of homology in particular fields of
biology or for particular situations:homogenesisfor the similarity of offspring to their parents
[Mitchell, 1910],homogeneticor normal morphogeny [morphology], in contrast to convergent
morphogeny or mere convergence [Willey, 1911],homogenicfor genes with single alleles [Fisher,
1928] andhomogeneticfor the pairing in hybrids of chromosomes from one of the original ances-
tors [Waddington, 1939].
(^17) See Patterson [1982; 1988], Wake [1991], McShea [1996] Sanderson and Hufford [1996], Meyer
[1999] and Hall [1999a; 2002a; 2003a] for discussion of various types of homoplasy.