prey: with very few species of land birds, no land
mammals, and a small land area, it is doubtful
whether a predator population could maintain
itself, were one to reach the island (Williamson
1981). This interpretation suggests that impoverish-
ment among one group may lead through food-
chain links to disharmony in another.
The notion of disharmony has been criticized by
Berry (e.g. 1992) who has written:
It is a somewhat imprecise concept indicating that the
representation of different taxonomic groups on islands
tends to be different to that on the nearest continent, and
carrying the implication that there is a proper—or har-
monious—composition of any biological community.
This is an idea closer to the ‘Principle of Plenitude’ of
mediaeval theology than to modern ecology. (Berry 1992,
pp. 4–5)
Berry has a point: that island environments have
‘sampled’ distinctive subsets of the mainland pool
does not necessarily imply that there is something
wrong with island assemblages and that they are
further from balance with their physical environ-
ment than those of continents (this is a question
worth posing, but it can be distinguished as a sepa-
rate question). Nonetheless, consideration of remote
oceanic islands shows that they are distinctive, and
there seems little point in developing a longer phrase
to capture the same idea as the term disharmony.
Another classic example of the filtering out of
particular groups of animals with increased isola-
tion is provided by Fig. 3.3, which shows the distri-
butional limits of different families and subfamilies
of birds with distance into the Pacific from New
Guinea. Such patterns have sometimes been termed
sweepstake dispersal. The data in Fig. 3.3 seem to
be interpretable broadly in terms of the differing
dispersal powers of the groups of birds concerned,
illustrating that insular disharmony can be identi-
fied not only at the level of orders but also below
this at the family level (cf. Whittaker et al. 1997).
However, as pointed out by Williamson (1981),
within Fig. 3.3 there is a general thinning out of
islands and decline in size to the east. This raises
the question of whether dispersal difficulties alone
have caused the filter effects—a problem that, using
presently available data, appears insoluble (but see
Keast and Miller 1996).
The New Guinea–Pacific study provides a basi-
cally unidirectional pattern, but in other cases a
two-way filter appears to operate (Carlquist 1974),
best exemplified for certain linearly configured
archipelagos. A classic example is shown in Fig. 3.4,
which quantifies the decline in reptiles and birds of
oriental affinity and the increase in Australian
species going from west to east along the Lesser
Sunda Islands. A similar sort of two-way pattern
can also be detected in certain circumstances on
mainlands, particularly on peninsulas (e.g. Florida;
Brown and Opler 1990), but island archipelagos
generally provide the clearest filter effects.Biogeographical regionalism and the vicariance/dispersalism debate
These two-way filter effects evidence a merging
of faunas or floras between two ancestral regions.
They indicate the biogeographical imprint of
extremely distant events in the Earth’s history,
intimately connected with the plate-tectonic
processes that have seen the break-up of super-
continents, parts of which have come back into
contact with one another many millions of years
later. Early students of biogeography, notably
Sclater and Wallace, recognized the discontinu-
ities and, on the basis of the distribution patterns
of particular taxa, were able to divide the world
into a number of major world zoogeographic or
phytogeographic regions. One such scheme is
shown in Fig. 3.5.
The boundary between the Oriental and
Australian zoogeographic regions is marked by
what has long been known as Wallace’s line
(Fig. 3.6), which marks a discontinuity in the
distribution of mammals, and divides the Sunda
islands between Bali and Lombok—not an alto-
gether obvious place to draw the line from the point
of view of the present-day configuration of the
islands. Modifications, and additions in the form of
Weber’s line, distinguishing a barrier for Australian
mammals like Wallace’s line for Oriental mammals,
have also been proposed (Fig. 3.6). However, the
precise placing of lines is problematic, in part
because the area between the Asian and Australian
continental shelves (‘Wallacea’) actually contains52 THE BIOGEOGRAPHY OF ISLAND LIFE