0198566123.pdf

biogeographers, in that we may misinterpret frag-
mentary data and reach erroneous conclusions.
Before we embark on a consideration of speciation
patterns and processes, we provide at the end of
this chapter some brief illustrations of island losses
as a cautionary note.

3.2 Species poverty

Islands typically have fewer species per unit area
than the mainland, and this distinction is more
marked the smaller the area of the island, i.e.
interarchipelago species–area curves are steeper
than curves constructed by subdividing a large
mainland area (Rosenzweig 1995). This will be
examined more fully in Chapter 4, and the follow-
ing examples should suffice to introduce the theme
here. First, Figure 3.1 illustrates that no matter
whether the Californian mainland plant data are
built up in spatially nested sets or non-nesteddata are
considered, the points lie above the regression line
for the Californian island data. A second example
(after Williamson 1981) is provided by Jarak island
in the Straits of Malacca (96 km from Sumatra,
64 km from Malaya, and 51 km from the nearest
island). This 40 ha island is forested but lacks
dipterocarps, the family that provides the most
common dominant species in the forests of Malaya,
but whose members have poor long-distance
dispersal ability. Within a 0.4 ha plot on the island,

Wyatt-Smith (1953) recorded 34 species of trees

having a trunk diameter greater than 10 cm,

compared with figures of 94 and 102 species,

respectively, for two Malayan mainland plots.

These mainland figures actually exceeded the total

spermatophyte (not just tree!) flora noted by Wyatt-

Smith in his survey of Jarak, and although it would

be a surprise if this was found to be a complete

inventory of all the species on the island (a

notoriously difficult feat to achieve), it nonetheless

gives an indication of the extent of impoverishment

of a not particularly remote island. A third example

is St Helena (15 56 S, 5 42 W), a remote island of

about 122 km^2 , and 14.5 million years old, which

has a known indigenous flora of just 60 species, of

which 50 are endemic, with 7 of them extinct, and

others endangered (Davis et al. 1994).

The successful spread of species transported to

islands such as St Helena in recent centuries

would appear to indicate that such islands are

undersaturated with species and could support

more in total (D’Antonio and Dudley 1995; Sax

et al. 2002). However, some turnover is often

involved, with endemic plant and especially

animal species becoming extinct as part of this

process. As discussed later in this volume, the

assessment of such changes in diversity is prob-

lematic, as, for instance, the losses and gains are

also tied up with other human influences, such as

forest clearance and horticultural and agricultural

SPECIES POVERTY 49

Figure 3.1Species–area curves for Californian plants,

showing the steeper slope for islands (logS 1 open

squares), compared with two alternative nested sets

from within the mainland (logS 1 dots, San Francisco area;

logS 2 , open circles, Marin County). The diamonds

represent other areas from the mainland, which lie in

neither nested set. (Redrawn from Rosenzweig 1995,

Fig. 2.9.)