74 THE BIOGEOGRAPHY OF ISLAND LIFE
dispersal and those largely dismissive of the
biogeographical significance of such processes,
favouring tectonic scenarios of barrier creation,
involving the splitting of ranges (vicariance
events). Work reviewed in this and the previous
chapter on the environmental history of islands and
on molecular phylogenies, argues for both sets of
processes playing a role, but in particular provides
overwhelming evidence for long-distance dispersal
being a biogeographically significant process.
In some cases, island forms may have changed
less than the mainland lineages from which they
sprung, or have persisted while the mainland pop-
ulation has failed, in which case they may be
termed palaeoendemics. However, where novelty
has arisen predominantly in the island context, they
are called neoendemics, although it should be rec-
ognized that this is to divide a continuum. Recently,
molecular phylogenetic analyses have begun to
transform our understanding of island lineages and
their relationships to continental sister clades, as
illustrated herein by extensive reference to the
complex colonization and evolutionary history of
the Atlantic islands of Macaronesia.
In this chapter we have avoided discussion of the
mechanisms of evolutionary change on islands,
although noting some correlates with high
proportions of endemics in island biotas. With
increasing isolation, island size, and topographic
variety, the number and proportion of endemic
species increases. Endemism within a taxon appears
to be at its greatest in regions near the edge of the
effective dispersal range. Thus, although a high
proportion of its plants and birds are endemic,
Hawaii has only three skinks, none of which is
endemic. Their colonization may perhaps have been
human-assisted and fairly recent: Hawaii is thus
beyond the effective dispersal range for skinks. In
contrast, the less remote islands of the Pacific Ocean
evidence a high proportion of endemism of skinks
and, on still less remote islands, of rodents. Ancient
‘continental’ island fragments (e.g. Madagascar) can
also be rich in endemics, although this can reflect
varying contributions from (1) stranded ‘vicariant’
forms and (2) the operation of long-distance
dispersal over long periods of time (e.g. particularly
significant for the New Zealand flora).
Some lineages have done particularly well on
oceanic islands, land snails and fruit bats among
them. Particular genera, such as the Hawaiian
Drosophila(fruit flies) and Cyrtandra(a plant exam-
ple) have radiated spectacularly. Terrestrial mam-
mals have made a good showing only on less
remote islands, often those insufficiently remote to
have escaped the attention of Homo sapiensat an
early stage, and many have thus failed to survive to
historical times. New discoveries continue to be
made of living (extant) and fossil (extinct) island
forms and this serves as a cautionary note, such
data must be appraised carefully before we can
attempt to explain current biogeographical patterns
in terms of evolutionary theory.