Page 11 - Delicado_Machordom_2015

D. Delicado et al. Evolutionary patterns of Pseudamnicola

3 Ma) to the ca. 10 Ma, 5 Ma and 2 Ma estimated in Deli- that such a cladogenetic event is more likely to have
cado et al. (2013). In any case, these ages indicate that their occurred by a process of colonization of several Mediterra-
phylogenetic relationships and distribution could be the nean regions during the MSC followed by isolation, rather
result of vicariant and climatic events that occurred in the than by a vicariant event produced by fragmentation of a lar-
Iberian Peninsula during the Miocene, or of a process of ger habitat. This is probably testified by the j parameter in
isolation by distance (Wright 1943) (also supported by our ancestral area estimation. The poor dispersal capability
Mantel test), or a combination of both. The Miocene was a of freshwater snails (they require permanent water courses,
crucial period for the Iberian Peninsula because of several although they may survive desiccation for several days) (Jen-
geological processes related to Alpine orogeny that criti- sen et al. 1996; Haase et al. 2010; Havel et al. 2014), and
cally affected the diversification and dispersal of many their inability to cross marine water masses make the
species of plants and animals (Joger et al. 2007; Miguel hypothesis of colonization less plausible. However, the esti-
et al. 2007; Pardo et al. 2008). Thus, the uprising of the mated periods of diversification of Pseudamnicola species are
current Iberian mountain ranges during the Alpine orogeny subsequent to when the Mediterranean peninsulas and
may have led to fragmented habitats of Corrosella and as a islands were formed and therefore, it is likely that the popu-
consequence, to allopatric speciation within the group, lations arrived after that event. Two alternative scenarios
which increased its species richness and changed distribu- may also explain the colonization process: (i) via land bridges
tion patterns. For instance, the second and last uprising of connecting several microplates, for instance between Sardi-
the Pyrenees in the North (Barbadillo et al. 1997; Hevia nia, Tunisia, Sicily and Italy (Rosenbaum et al. 2002; Goes
2004) may have influenced the split between C. astieri and et al. 2004) or between the Balearic Islands and the Iberian
C. navasiana – C. hauffei. In addition, the creation of the Peninsula (Carranza et al. 2004; Fochetti et al. 2009; L azaro
Betic Cordillera and the active plate tectonics in the south- et al. 2011), both of which were established during the Mes-
ern and eastern regions of the Alboran Sea (approximately sinian (6–5 Ma) and the Plio-Pleistocenic glaciations, or (ii)
10 Ma, according to Rosenbaum et al. 2002) may have through passive extra-aquatic dispersion via water birds.
been one of the reasons of the radiation involving southern This dispersal mechanism has been reported as the most fea-
Corrosella species. Finally, the creation of the Iberian sible explanation for the distribution pattern of other hydro-
hydrological network during the Quaternary (Vargas et al. bioids, especially for brackish genera (Wilke & Davis 2000;
1998) may have led to vicariant events that influenced the Liu et al. 2003; Haase et al. 2010; Kappes & Haase 2012).
most recent splitting of Corrosella species. Indeed, this hypothesis has recently been confirmed by
Furthermore, species of Pseudamnicola s. str. have been Wada et al. (2012), who demonstrated that gastropods can
found on the Mediterranean basin mainland, peninsulas pass through and survive the digestive system of birds and
and islands, yet only the species P. subproducta inhabits the fishes and can even subsequently produce offspring, making
Iberian Peninsula. Therefore and though with low proba- predation by birds a possible method of dispersal. Kappes &
bility values (see pie charts at nodes of Fig. S1), our results Haase (2012) proposed several active and passive dispersal
point towards the hypothesis that the diversification of mechanisms for freshwater mollusks. However, in this case,
Pseudamnicola s. str. did not occur in the Iberian Peninsula, no geographic criterion exists and multiple tectonic plates
but probably towards central Mediterranean areas (Fig. 3). are involved, therefore long-distance dispersal via birds may
The inclusion of more populations from different Mediter- be more likely hypothesis to explain this fact.
ranean localities in future analyses will help to clarify the
most likely evolutionary scenario of the Pseudamnicola s. str. Systematics
In any case, the coalescence analysis performed with the Genus Pseudamnicola Paulucci, 1878
available data shows rapid cladogenetic events whose period Type species. Bithynia lucensis Issel, 1866 (Kennard &
does not mismatch any of the hypotheses discussed above, Woodward 1926), subsequent designation.
as such events seem to have been posterior to the geologi-
cal origin of the western Mediterranean region and ante- Diagnosis. Shell ovate-conic, slightly longer than wide and
rior to Plio-Pleistocenic glaciations. Thus, the tree an aperture wider than long; female genitalia with pyriform
topologies and estimated divergences suggest the geological bursa copulatrix, pigmented renal oviduct and one elongate
events occurred during the Messinian salinity crisis (MSC) seminal receptacle; prostate gland between two and three
(between 5.96 and 5.33 Ma, according to Krijgsman et al. times longer than wide; penis broadly triangular with the
1999) as the main drivers for the diversification of Pseu- base expanded and many folds along its entire surface;
damnicola s. str. penis with a dark patch of pigment, whose extension varies
Based on both the divergence time estimate and the cur- among species, from its middle region to the tip; pig-
rent distribution patterns of Pseudamnicola s. str., we deduce mented nervous system generally elongate.

ª 2015 Royal Swedish Academy of Sciences, 44, 4, July 2015, pp 403–417 413