HOW FRAGILE AND INVASIBLE ARE ISLAND ECOSYSTEMS? 321disturbance, and introduction on an island-wide
basis. They also often sustain surprisingly high
human populations (Watts 1987); for example,
Cronk and Fuller (1995) note that Mauritius, with a
population of 530 people/km^2 , may be compared
culturally and historically with India, which has
only about 240 people/km^2.
●Crossroads of intercontinental trade. Particularly
in the days of sailing ships, islands have often been
used as watering points, re-supply posts, and stag-
ing posts for trade and for plant transport. As they
have often changed hands between different colo-
nial powers, they have, moreover, had opportuni-
ties to receive introductions from differing networks
of nations and biogeographical regions.
The deluge of disturbance and of alien species
which has swept across so many oceanic islands
goes some way towards explaining the apparent
susceptibility of islands to invasion. At least one
tropical biologist, who has worked for many years
on Hawaii, argues that this is the case:
Contrary to common opinion, many endemic island
species are strong competitors. They would not be elimi-
nated from their niche in the island ecosystem if it were
not for the new stress factors introduced by man. The
island species evolved with stress factors associated with
volcanism, fire in seasonally dry habitats, and occasional
hurricanes. The effects of volcanism resulted in superior
adaptation of many native species to extreme edaphic
conditions existing on volcanic rockland, where soil-water
regimes fluctuate almost instantly in direct relation to
rainfall. (Mueller-Dombois 1975, p. 364)
This general position is supported by D’Antonio
and Dudley (1995) who conclude that, with excep-
tions, island habitats are not inherently more easily
invaded than are continental habitats, but that,
nonetheless, exotics have had a greater impact on
island systems. Some more formal analysis pro-
vides a measure of support for Mueller-Dombois’s
views. It has been suggested that as a rough rule
(the tens rule), about 1 in 10 introductions into
exotic territories becomes established (Williamson
1996). The data for Hawaiian birds indicate a far
higher success rate, of over 50% of introduced
species establishing. However, Williamson (1996)
notes that the majority of introductions have
colonized disturbed lowland habitats, and that the
figure for species establishing into more or less
intact native habitats (some lowland forest,
together with upland habitat) falls at between 11
and 17%, which is not significantly out of line with
the tens rule. Studies elsewhere within very
disturbed islands, containing large numbers of
naturalized exotic species, have also reported that
those habitats that have remained largely
undisturbed quite often contain few exotics (e.g.
Watts 1970, on Barbados; Corlett 1992, on
Singapore). These examples suggest that the
susceptibility of island ecosystems to invasion is
above all a reflection of the profound alteration of
habitats. However, other case studies cited in this
chapter demonstrate that some invaders have not
required such assistance.
One of the curious features of particular inva-
sions is that invading species sometimes undergo a
pattern of increase to a peak of density, followed by
a crash, sometimes to extinction. This boom-and-
bust pattern appears to have occurred fairly often
on islands (Williamson 1996). It has, for instance,
been recorded for a number of the passerines intro-
duced to the Hawaiian islands, with their time to
extinction varying between 1 and 40 years. There
may be a number of reasons for these boom-and-
bust patterns, but Williamson favours those relat-
ing to resource deficiencies as providing the most
general explanation.
From the point of view of island biogeography
theory it is noteworthy that the introduction of a
single exotic species, such as the brown tree snake
(Boiga irregularis) on Guam and the shrub Miconia
calvescenson Tahiti, may cause the local extinction
of numerous native island species. Moreover, in the
case of Miconia, the effect operates within the same
taxon, i.e. it is a single species of plant which is
implicated in the loss of numerous native plants.
This provides an instance in which an equilibrium
model, involving turnover on a replacement basis
subject only to relaxation effects from area reduc-
tion (as discussed in Chapter 10), would be a
wholly inappropriate device for predicting species
losses.