Page 10 - Fiorentino_alii_2008

818 V. FIORENTINO ET AL.

THE EVOLUTION OF GLOBULAR AND ally attributed to transport by humans (Welter-
KEELED FLAT SHELLS Schultes, 1998) when populations are distributed
near human settlements (Uit de Weerd, Schneider &
Land snail shells usually tend to be either high spired
or discoidal, with rare globular shapes inbetween Gittenberger, 2005). Human-aided dispersal of some
Sicilian Marmorana populations is supported by
(Teshima et al., 2003). It has been claimed that mor-
phological variation may arise from selection in dif- various observations. First, all populations belonging
to the quite easily distinguished ‘M. muralis’ are
ferent environments (Engelhard & Silk, 1994; Welter-
Schultes, 2000) or may be a product of population renowned for being widespread in Sicily in ruderal or
human habitats. The limestone substrate where Mar-
history (Gould & Woodruff, 1990; Davison & Clarke,
2000). morana live was widely used as building material,
which was transported over large distances in the
Different explanations have been given for the
evolution of flat keeled shells. Cook & Pettitt (1979) Mediterranean. Snails may therefore have been
transported with the limestone blocks, as hypoth-
stated that keeled shells may be more resistant to
crushing than rounded forms, whereas Solem & esized for others groups of snails (Welter-Schultes,
1998; Schilthuizen et al. 2004; Uit de Weerd et al.,
Climo (1985) found that a peripheral keel is associ-
ated with sheltering habitats on open ground. More- 2005). ‘Marmorana muralis’ from Florence is grouped
with those from Trapani and Marsala, two major
over, it has been hypothesized that keeled shells are
an adaptation to limestone substrates (Alonso et al., western Sicilian harbours and human settlements, on
the trade routes to northern Italy. The surroundings
1985) or to movement through rock crevices on hard
substrates (Teshima et al., 2003). Cain & Cowie of Trapani and Marsala are inhabited by ‘M. muralis’
(1978) suggested that flat shells are adapted for populations. This confirms that the phenotype cannot
be the result of adaptation to ruderal habitat (also
crawling on horizontal surfaces.
All Marmorana populations are confined to lime- home to other shell phenotypes).
Regarding the Favignana population, which is
stone substrates. Globular and flat shells are both
present without any clear evidence of environmental grouped with eastern Sicilian populations, it is diffi-
cult to explain its origin in term of Sicilian paleogeog-
constraints (Fiorentino et al., 2008). Interestingly,
according to Elejalde et al. (2005), the evolution of raphy. However, Favignana, famous for its trade in
fish and quarried stone and the largest island of the
keeled flat shells does not always occur as an adap-
tation to limestone substrates and ‘all theories about Egadi Archipelago, was a Punic colony, that traded
since the fifth century BC (Racheli, 1986). Further-
the evolution of this shell form are basically specula-
tive’. Similar to Marmorana from Sicily, the genus more, the first and largest Greek colonies were in
eastern Sicily. In the absence of more detailed data, it
Iberus studied by Elejalde et al. (2005) has globular to
flat-shelled populations that cross, producing fertile is therefore possible that colonization by Marmorana
followed ancient trade routes. To verify this hypoth-
hybrids. Intermediate shelled populations are also
known to occur in nature. esis, future research should investigate major ancient
commercial routes in relation to the erratic distribu-
From our analysis, it emerges that globular and flat
shapes may have been present during radiation of the tion of Italian ‘M. muralis’.
group. If flat keeled shells are associated with shel-
tering habitats on open ground, as postulated by
TAXONOMIC IMPLICATIONS
Solem & Climo (1985), then the two shell shapes
(globular and flat keeled) may have adapted to differ- Although we expected to observe at least five species
ent rock microhabitats in the two extreme habitats (Manganelli et al., 1995; Bank, 2007), corresponding
to the main shell morphs, the analysis demonstrated
occurring in a quilt pattern during the Quaternary
glacial phases: steppe-like habitat on the continental local differentiation, parapatric distribution, and com-
pletely incongruent patterns between morphology and
shelf and forested areas elsewhere (Agnesi et al.,
2000). This hypothesis is difficult to verify with the phylogeny, which disagreed with current taxonomy. In
the present situation, defining the taxonomic rank
present state of the art.
and status of the various clades identified in the
present study is an extremely difficult task, except for
the Marettimo population, which molecular analysis
DISPERSAL EVENTS: ON THE ROOTS OF ANCIENT showed to be quite different from the other Sicilian
TRANSPORTS Marmorana. The distinction of Marmorana of Maret-
Given the low vagility of land snails, long-distance timo is also confirmed by morphology: the shape of
dispersal generally occurs through passive transport. its love-dart is quite different from that of the other
Because of their size and habitat, rafting and trans- Sicilian populations studied so far; the dart is
port by birds are often excluded. Dispersal is gener- regarded as significant for distinguishing Marmorana

© 2008 The Linnean Society of London, Biological Journal of the Linnean Society, 2008, 94, 809–823

Downloaded from https://academic.oup.com/biolinnean/article-abstract/94/4/809/2701256
by guest
on 18 November 2017