Page 8 - Invasive_rats_2009

1638 L. Ruffino et al.

Thus, even though distance to the continent, human
presence on islands, and intensity of human activ-
ities may have played a significant role in rat
dispersal in the past, these factors no longer limit
rat presence on Mediterranean islands. Island area
and isolation from the nearest potential source of
rats were the best explanatory factors in ship rat
presence on Mediterranean islands. The larger the
island, the more likely it is to be rat-infested. This
pattern is consistent with island biogeography
theories on island species richness (Lomolino
1982, 1984; Adler and Wilson 1985) as well as
on the probability of small mammal species occur-
ring on islands (Adler and Wilson 1985; Russell
and Clout 2004). Moreover, this area-related pattern
may be related to the highest probability of human
colonization on large islands due to highest habitat
Fig. 4 Influence of rat presence (noted P), rat control, and rat
absence (absent or eradicated) (noted C and A, respectively) on and resource diversity (Lomolino 2000), and thus to
the breeding success of the Cory’s shearwater on Western the highest probability of accidental rat introduc-
Mediterranean islands (n = 219). Breeding success is defined
tion. However, in the case of the long-invaded
as the number of fledged chicks divided by the total number of
Western Mediterranean island system, 74% of
hatched eggs
islands ranging from 1 to 5 ha support ship rat
populations and many very small islets have also
Discussion been found to be rat-infested, when close enough to
continents or other landmasses. On such very small
Ship rat distribution on Mediterranean islands islets close to a potential source, rats are known to
maintain only transient populations (Adler and
Today, few Mediterranean islands remain rat-free. Wilson 1985; Alcover 1993), facing a higher
Compared to most oceanic archipelagoes, Mediter- extinction risk due to founder effect, genetic drift,
ranean islands generally exhibit a low degree of and inbreeding depression (Heidrick and Kalinow-
geographical isolation and can be considered as a sky 2000; Frankam 2003). Unfortunately, we could
group of continental islands, surrounded by the not take into account rat population turnover on the
African and Euro-Asiatic landmasses (Greuter smallest islands in this review due to the lack of
1995). Moreover, most Mediterranean islands are available data. Distance from the nearest potential
part of wider island systems (e.g., Sardinia, Sicily, rat population source was a limiting factor for rat
Corsica, Balearic), which decreases their isolation presence only for the smallest islets and uninhabited
from the continent (Sara and Morand 2002). Thus, islands, whose isolation may limit natural dispersal
many Mediterranean islands (mostly small islands) and human transport links, thereby preventing the
are close enough to each other or to the continent external recruitment required for population persis-
to be reached by rats by natural dispersal (i.e., tence (e.g., Cheylan 1999).
B500 m). However, long-standing and intensive
human transport and trade activities within the basin
have probably largely facilitated the distribution of Ship rat impact on Mediterranean
this commensal and generalist rodent, which has Procellariiformes
rapidly become established on most Mediterranean
islands (Vigne 1992; Audouin-Rouzeau and Vigne Surprisingly, rat presence was not found to be a
1994; Masseti 1995; Bover and Alcover 2008). leading factor explaining the distribution of the four

123