Page 14 - Asylv_molars_Pmax_revised_2014_01
P. 14
14
km²;
Oléron,
18
km²,
Ré,
9
km²),
suggesting
a
key
role
of
random
processes,
i.e.
founder
effect,
bottleneck
and
drift,
in
promoting
differentiation.
Yet,
random
processes
alone
should
lead
to
differentiation
on
the
various
islands
according
to
various
directions
corresponding
to
the
directions
of
the
P
matrix
matrix,
leading
to
a
similarity
in
the
directions
and
eigenvalues
of
the
P
matrix
and
of
the
inter-‐group
VCV
(Ackerman
and
Cheverud
2004).
In
contrast,
here,
in
all
the
cases
of
pronounced
differentiation,
evolution
occurred
in
a
parallel
way.
An
adaptive
component
may
be
proposed:
due
to
a
decrease
of
interspecific
competition,
niche
widening
is
likely
to
occur
on
many
islands
(Orsini
and
Cheylan
1988),
possibly
favoring
wide
molars
able
to
cope
with
a
broad
range
of
food
items.
Strikingly,
this
parallel
evolution
occurs
following
Pmax.
The
preexistence
of
widespread
variants,
easily
produced
by
modulating
the
same
genetic/developmental
networks
and
matching
the
adaptive
requirements,
may
have
facilitated
the
response
to
local
selective
forces,
even
on
a
short
time
scale
and
without
accumulation
of
important
genetic
difference.
The
channeling
role
of
Pmax
thus
provides
a
new
light
on
such
cases
of
parallel
evolution,
apparently
amazing
but
facilitated
by
the
congruence
of
reduced
population
size,
strong
selective
pressure,
and
agreement
between
the
direction
of
least
evolutionary
resistance
(Pmax)
and
the
selected
shape
change.
Developmental
mechanisms
and
functional
constraints
beyond
Pmax
Pmax
corresponds
to
a
variation
ranging
from
slender,
elongated
molars
and
to
broad,
stocky
molars.
Advances
in
developmental
biology
of
tooth
morphogenesis
provide
hints
about
the
mechanisms
that
may
underlie
such
morphological
variation.
Concerted
changes
in
the
size
of
all
cusps
may
be
mediated
by
changes
in
the
regulation
of
genes
controlling
tooth
development
(Mustonen
et
al.
2003),
a
change
relatively
‘easy’
to
produce.
This
explains
why
corresponding
variants
are
frequent
in
a
population,
and
why
this
kind
of
variation
can
be
easily
mobilized
by
selection
and
drift
to
cause
pronounced
differences
among
populations
even
in
relatively
short
time
span.
In
some
cases,
the
change
involved
is
more
localized
and
the
molar
elongation
seems
to
involve
mostly
the
anteriormost
part
of
the
tooth.
This
localized
variation
may
be
attributed
to
another
developmental
process,
namely
a
differential
incorporation
of
a
vestigial
bud
anterior
to
the
molar
field
and
reminiscent
of
a
premolar
present
in
murine
ancestors
(Peterková
et
al.
2005;
Prochazka
et
al.
2010).
This
component
of
molar
shape
variation
is
very
pronounced
in
the
house
mouse,
leading
to
the
occasional
occurrence
of
an
anterior
cusplet
(Renaud
et
al.
2011).
Such
a
cusplet
is
rare,
but