Cs. Gallardo et Pe. Penchaszadeh, Hatching mode and latitude in marine gastropods: revisiting Thorson's paradigm in the southern hemisphere, MARINE BIOL, 138(3), 2001, pp. 547-552
For much of the last century the developmental mode of marine invertebrates
, particularly of prosobranch gastropods, has been thought to reflect a lat
itudinal pattern known as Thorson's rule; the proportion of species with pe
lagic larvae decreases with increase in latitude. Although the predictions
of this rule have been criticized recently, its validity along latitudinal
transects in the world ocean still remains poorly tested. In the present wo
rk, we compare the frequency of occurrence of contrasting prosobranch hatch
ing modes (pelagic versus benthic development) along two latitudinal gradie
nts of the subequatorial coastline of South America: the southeast Pacific
and the southwest Atlantic. The results are clearly contrasting. While the
pattern observed along the Pacific (Chilean) coast fits the predictions of
Thorson's rule very well, benthic development predominates all along the At
lantic coast, even at subtropical latitudes. This difference in observed pa
tterns is attributable to the different compositions of the gastropod assem
blages on each side of the South American continent, which are determined,
in turn, by differences in the ecological conditions on the two sides of th
e continent. The scarcity of pelagic development among the Atlantic prosobr
anch gastropods reflects the near-continuous soft-bottom habitat there, and
the consequent prevalence of predaceous, soft-bottom taxa that had acquire
d in their evolution more evolved, non-pelagic patterns of development. In
contrast, the Pacific coast is much more heterogeneous, with a diverse mixt
ure of benthic habitats, including rocky substrates that, in part, support
grazing taxa. Our results indicate that two factors are important for Thors
on's rule to be valid. First, the habitat needs to include rocky substrates
, as soft-bottom habitats appear to favour non-pelagic development. Second,
a diverse assemblage of taxa need to be compared to avoid the problem of p
hyletic constraints, which could limit the evolution of different developme
ntal modes.