Hawaiian islands, over 100 species of Cyrtandraof
the widespread family Gesneriaceae (Otte 1989). In
general, the greater the taxonomic distinction
between organisms, the more ancient their differen-
tiation, although calibrating the timescale is diffi-
cult, particularly for fossil organisms.
In recent decades the methods of traditional evo-
lutionary systematics have been challenged by the
development of phylogenetic systematics, or
cladistics, which presumes to supply a more objec-
tive means of quantifying the relatedness of a taxo-
nomic group. This involves scoring the degree to
which different organisms share the same uniquely
derived characteristics that other taxa outside the
group do not possess. These characteristics are then
used to construct a cladogram, or phylogeny(phy-
logenetic tree), which is a branching sequence set-
ting out the most parsimonious (‘simplest’) model
for the relationships between taxa, i.e. it provides a
hypothesis for the evolutionary development of a
lineage. The data employed may be morphological
or genetic (i.e. based on DNA sequences). The
papers in Wagner and Funk (1995) provide exam-
ples of both as applied to Hawaiian plants and ani-
mals, as well as an excellent explanation of the
methods of cladistics.
Historical biogeographers also make use of what
are termed area cladograms. An area cladogram is
constructed by first determining the phylogeny and
then replacing the species names with the geo-
graphic location in which those species are found.
This new tree in effect provides a series of hypothe-
ses of the sequence of dispersal and vicariance
(population subdivision by barrier formation) invol-
ved in the evolutionary development of the line-
age, emphasizing the role played by geological/
environmental history in lineage development.
Although the proponents of phylogenetic
systematics are often forthright in praising the
approach and condemning traditional systematics,
it should be recognized that the decisions as to
which characters to include in the phylogeny, and
the rules used in its construction, are the choice of
the user. The phylogeny thus constructed is there-
fore merely one approximation, albeit hopefully a
basically reliable one, to the history of events that
produced the lineage. Once such a phylogeny has
been constructed, it can then be used to explore the
likely sequence of interisland and intraisland speci-
ation events within archipelagos, or between
islands and mainlands.
The most exciting opportunities have been
opened up by advances in molecular biology and
genetics and these methods have been seized upon
in island evolutionary studies. Such studies benefit
from use of the ‘genetic clock’ which, on the
assumption of a more or less constant rate of
accumulation of mutations and thus of genetic
differences between isolates, allows estimates of the
dates of lineage divergence (Box 8.1). An excellent
illustration is provided by the studies of Thorpe
et al. (1994) on the colonization sequence of the
western Canarian lizard, Gallotia galloti, in which
the direction and timing of colonization as postu-
lated by genetic clock analyses (nuclear and mito-
chondrial DNA divergence) was shown to be
entirely compatible with the independently
derived geological data for the timing and
sequence of island origins.
It should be noted that these molecular clocks
may not always run to time. A key problem is
exemplified by Clarke et al.’s (1996) study of two
species of land snails (Partula taeniataandP. sutu-
ralis) on the island of Moorea (French Polynesia). It
was found that although the two species exhibit
significant morphological and ecological differ-
ences, they are genetically close. Clarke et al. attrib-
uted this closeness to ‘molecular leakage’ or
‘introgression’: the convergence of genetic structure
through occasional hybridization. Such introgres-
sive hybridizationis now considered to be quite
common in particular island lineages (Clarke and
Grant 1996). Even low rates (as low as 1 in 100 000)
may be enough to upset the phylogenetic trees and
molecular clocks upon which scenarios of island
evolution discussed in Chapter 9 now rest (Clarke
et al. 1996; Zink 2002). For further insights into these
and other methodological problems of biogeogra-
phy and systematics, see Myers and Giller (1988),
Thorpeet al. (1994), Ridley (1996), the exchange
between Herben et al. (2005) and Silvertown et al.
(2005); and for some reassurance about the use of
molecular clocks, see Bromham and Woolfit (2004)
and Box 8.1.198 SPECIATION AND THE ISLAND CONDITION