OBSERVATIONS ON THE FORCING FACTORS OF ISLAND EVOLUTION 239past’). In illustration, Silvertown (2004) has argued
that genetic data for Canarian endemic plants indi-
cate that niche pre-emption by earlier colonist line-
ages may have inhibited the success of later ones.
Emerson and Kolm (2005a), on the other hand,
argue for a sort of facilitation effect, claiming that
diversification rate of a taxon is an increasing func-
tion of higher species richness of that taxon (diver-
sity begets diversity), using Canarian plant data as
one of four illustrative data sets. In practice,
Emerson and Kolm’s analysis is not based on a rate
(i.e. per unit time) estimate but on the proportion of
species that are endemic to particular islands, and
all that they really show is that richer Canarian flo-
ras also have high proportions of endemic species.
Our own unpublished analyses of the patterns of
single-island endemism suggest that the opportuni-
ties for speciation show a roughly parabolic rela-
tionship to island age, with speciation rates and
proportions of SIEs peaking on islands of young to
intermediate age (cf. Silvertown 2004), and then
declining, consistent with what we term the island
immaturity–speciation pulse model (Box 9.3). It
follows that the correlation between plant species
richness and single-island endemism reported by
Emerson and Kolm (2005a) fails to demonstrate
that greater richness of a taxon leads to greater
rates of (i.e. faster) speciation in that taxon. Both
Silvertown’s (2004) and Emerson and Kolm’s
(2005a) studies have been criticized for the logic
involved in the interpretation (see Cadena et al.
2005; Saunders and Gibson 2005; and see replies by
Emerson and Kolm 2005b; Silvertown et al. 2005),
supporting our suggestion earlier in this book that
the only certainty about claims regarding competi-
tive effects is that they will be disputed.
Despite the problems in testing such ideas,
ecological interactions including competition and
interactions across trophic levels (mutualisms,
parasitism, disease, predation, etc.) are clearly
important within island ecology and evolution.
This is demonstrated by the extent to which
remote, species-poor, disharmonic islands have
produced radiations of lineages now occupying
ecological niche space that they occupy nowhere
else, by the increased tendency towards generalist
pollinator mutualisms on particular islands, by the
ground-nesting behaviour of many oceanic island
birds (many now driven to extinction), and by
many other phenomena reviewed in these pages.
Alongside these biotic mechanisms and forcing
factors, we must also consider how the role of abi-
otic forcing factors influences rates and patterns of
evolution. This point is amply illustrated in studies
reviewed already in this chapter.
How the two sets of forces (biotic and abiotic)
interact is hard to determine with any precision.
But we might set out the following hypothetical
scenario (developed further in Box 9.3). In the earli-
est stages of the life cycle of a remote volcanic
island, opportunities for evolution are severely lim-
ited by the barren and volatile nature of the system.
As the island builds in area, elevation, and topo-
graphic complexity through time, it acquires
colonists and complex food webs of interacting
organisms. Opportunities for adaptive and non-
adaptive radiation (and for mechanisms such as
counter-adaptation to operate) increase accord-
ingly. We might therefore expect the highest rate of
evolution in this stage. As the island ages it
becomes less active volcanically and erosion, mega-
landslips, and subsidence combine to reduce its
area, topographic complexity, and elevational
range. Opportunities for speciation into vacant
niche space and into isolated habitat pockets are
reduced along with the overall carrying capacity of
the island. Eventually the island slips back into the
sea, in the tropics perhaps forming an atoll and so
sharing an increasingly cosmopolitan strandline
fauna and flora with other similar systems. The
unique forms found on the island have by this point
either colonized other islands or failed altogether.
Such a scenario might well apply to the
Hawaiian islands, which have been likened to a
conveyor belt, in which the islands undergo a pat-
tern of birth, growth, maturity, and decline. The
more recent lava flows within the youngest island
are thus, today, the sites in which novel species and
adaptations are most apparent. Kaneshiro et al.
(1995, p. 71) in their analysis of species groups
within the picture-wing Drosophilaof Hawaii make
the following observation:
Most of these species, like many other extant terrestrial
endemic fauna, show a very strong but by no means