Tracing ancestral biogeography of Sonneratia based on fossil pollen and their probable modern analogues

Abstract

Extant tropical mangrove Sonneratia is assigned in the monogeneric subfamily of Lythraceae. There are still some debates on the early fossil pollen of Florschuetzia, which is well accepted as ancestral to Sonneratia. This paper re-assesses palynological interpretations on the historical biogeography of the genus, based mainly on the updated fossil pollen records of Paleogene through Quaternary and their probable modern analogues.
Florschuetzia was extensively documented from the late Eocene to middle Miocene in paleotropics around the Tethyan region. According to the geological age of the fossil pollen and their mophological assessments, ancestral Sonneratia migrated from the center of origin in southeastern Asia probably during early Eocene, and radiated and expanded northward to China and Japan, southward to Australia, and westward to east Africa. Until the warmer period of the early middle Miocene (Langhian), Sonneratia had the largest geographical range suggested by abundant fossil pollen from southern mainland China and southwestern Japan, out of latitudinal limit of this extant genus. Quaternary glaciations, especially the Last Glacial Maximum (LGM), played a significant role in shaping the current biogeography of Sonneratia. However unequivocal assignments of early Florschuetzia and the associated variants to the evolved Sonneratia remain an issue due to the lack of intensive morphological comparisons from different fossil sites. Thus, we examined the evolutionary trends of the extinct genus Florschuetzia towards Sonneratia on the basis of our synthesis on the updated published data and our recent pollen morphological investigation.

Full text

Available
online
at
www.sciencedirect.com
ScienceDirect
Palaeoworld
22
(2013)
133–143
Tracing
ancestral
biogeography
of
Sonneratia
based
on
fossil
pollen
and
their
probable
modern
analogues
Limi
Mao a,∗,
Swee
Yeok
Foong b
aState
Key
Laboratory
of
Palaeobiology
and
Stratigraphy,
Nanjing
Institute
of
Geology
and
Palaeontology,
Chinese
Academy
of
Sciences,
Nanjing
210008,
China
bSchool
of
Biological
Sciences,
Universiti
Sains
Malaysia,
Penang
11800,
Malaysia
Received
19
May
2013;
received
in
revised
form
20
August
2013;
accepted
5
September
2013
Available
online
14
September
2013
Abstract
Extant
tropical
mangrove
Sonneratia
is
assigned
in
the
monogeneric
subfamily
of
Lythraceae.
There
are
still
some
debates
on
the
early
fossil
pollen
of
Florschuetzia,
which
is
well
accepted
as
ancestral
to
Sonneratia.
This
paper
re-assesses
palynological
interpretations
on
the
historical
biogeography
of
the
genus,
based
mainly
on
the
updated
fossil
pollen
records
of
Paleogene
through
Quaternary
and
their
probable
modern
analogues.
Florschuetzia
was
extensively
documented
from
the
late
Eocene
to
middle
Miocene
in
palaeotropics
around
the
Tethyan
region.
According
to
the
geological
age
of
the
fossil
pollen
and
their
morphological
assessments,
ancestral
Sonneratia
migrated
from
the
center
of
origin
in
southeastern
Asia
probably
during
early
Eocene,
and
radiated
and
expanded
northward
to
China
and
Japan,
southward
to
Australia,
and
westward
to
east
Africa.
Until
the
warmer
period
of
the
early
middle
Miocene
(Langhian),
Sonneratia
had
the
largest
geographical
range
suggested
by
abundant
fossil
pollen
from
southern
mainland
China
and
southwestern
Japan,
out
of
latitudinal
limit
of
this
extant
genus.
Quaternary
glaciations,
especially
the
Last
Glacial
Maximum
(LGM),
played
a
significant
role
in
shaping
the
current
biogeography
of
Sonneratia.
However
unequivocal
assignments
of
early
Florschuetzia
and
the
associated
variants
to
the
evolved
Sonneratia
remain
an
issue
due
to
the
lack
of
intensive
morphological
comparison
from
different
fossil
sites.
Thus,
we
examined
the
evolutionary
trends
of
the
extinct
genus
Florschuetzia
towards
Sonneratia
on
the
basis
of
our
synthesis
on
the
updated
published
data
and
our
recent
pollen
morphological
investigation.
©
2013
Elsevier
B.V.
and
Nanjing
Institute
of
Geology
and
Palaeontology,
CAS.
All
rights
reserved.
Keywords:
Sonneratia;
Florschuetzia;
Biogeography;
Fossil;
Pollen
1.
Introduction
The
well
described
Cenozoic
fossil
Sonneratia
and
its
ances-
tor
Florschuetzia
include
the
wood
from
the
Paleocene–Pliocene
of
southeastern
Asia
(Ramanujam,
1956;
Shallom,
1963;
Rao
and
Ramanujam,
1966;
Chitaley,
1968;
Awasthi,
1969;
Biradar
and
Mahabale,
1973;
Kramer,
1974;
Shete
and
Kulkarni,
1982;
Bande
and
Prakash,
1984,
1986;
Lakhanpal
et
al.,
1984;
Mehrotra,
1988;
Vozenin-Serra
et
al.,
1989;
Guleria,
1991;
Srivastava,
2008),
in
the
middle
Eocene
of
Libya
(Louvet,
1970),
in
the
Oligocene
of
Austria
(Hofmann,
1952)
and
Japan
(Srivastava
and
Suzuki,
2001),
and
an
exceptional
form
in
the
late
Cretaceous
of
Uzbekistan
(Shelomentseva,
1992);
the
root
(Chitaley,
1968)
and
leaf
(Ambwani,
1991)
from
the
Paleocene
∗Corresponding
author.
Tel.:
+86
25
83282245.
E-mail
address:
lmmao@nigpas.ac.cn
(L.
Mao).
of
India;
and
the
pollen
from
the
Eocene
through
Miocene
of
southeastern
Asia
(Muller,
1964,
1978;
Germeraad
et
al.,
1968;
Awasthi,
1969;
Graham
and
Graham,
1971;
Guleria
et
al.,
1996;
Morley,
2000;
Srivastava
and
Suzuki,
2001;
Songtham
et
al.,
2005;
Srivastava
and
Guleria,
2006),
in
the
mid-Tertiary
of
the
Red
Sea
and
Nile
Delta
(Legoux,
1978;
Morley,
2000),
in
the
late
Paleogene
of
France
(Gruas-Cavagnetto
et
al.,
1988),
in
the
late
Miocene
to
recent
of
New
Guinea
(Khan,
1974),
in
the
middle
Miocene
of
Japan
(Yamanoi
et
al.,
1980;
Yamanoi,
1984)
and
China
(Lei,
1998),
in
the
late
Eocene
of
China
(Lei,
1998;
Liu
and
Yang,
1999),
and
in
the
Pleistocene
(e.g.,
Sohma,
1973;
Sun,
1991;
Zheng
and
Zhou,
1995;
Wang
and
Zhang,
1998;
Zheng
and
Li,
2000;
Kumaran
et
al.,
2005;
Rugmai
et
al.,
2008).
Accord-
ing
to
Germeraad
et
al.
(1968)
and
Muller
(1978),
the
fossil
pollen
of
Sonneratia
can
be
clearly
assigned
to
the
extant
species
from
the
early
Miocene
(22.5
Ma).
The
records
of
Sonneratia
show
a
gradual
change
from
types
referable
to
Lythraceae
and
strongly
support
the
evolutionary
divergence
of
Sonneratiaceae
1871-174X/$
–
see
front
matter
©
2013
Elsevier
B.V.
and
Nanjing
Institute
of
Geology
and
Palaeontology,
CAS.
All
rights
reserved.
http://dx.doi.org/10.1016/j.palwor.2013.09.002

134
L.
Mao,
S.Y.
Foong
/
Palaeoworld
22
(2013)
133–143
(Tomlinson,
1986).
Recently
Graham
(2013)
synthesized
fossil
records
for
the
family
Lythraceae
and
rejected
some
morpho-
types
of
Florschuetzia
and
Sonneratia
that
are
not
compared
favorably
with
accepted
family
or
generic
parameters.
Thus,
it
is
necessary
to
re-assess
pollen
records
and
the
possible
associated
variants
of
this
genus
more
prudently.
Germeraad
et
al.
(1968)
intensively
investigated
Florschuetzia
in
Borneo.
Of
these,
some
Miocene
species
can
be
clearly
attributed
to
the
extant
Sonnera-
tia,
such
as
F.
levipoli
attributable
to
S.
caseolaris.
Muller
(1969,
1978)
also
investigated
extant
Sonneratia
palynologically
and
reviewed
its
fossil
record
of
wood,
flowers,
fruits
and
pollen,
and
concluded
that
the
genus
first
occurred
probably
in
the
early
Eocene
in
Southeast
Asia.
However,
this
conclusion
does
not
agree
with
the
earliest
Sonneratia-like
pollen
of
Florschuetzia
sp.
from
the
late
Paleocene
deposits
in
south
central
France
(58.7–55.8
Ma;
Gruas-Cavagnetto
et
al.,
1988;
Plaziat
et
al.,
2001).
Moreover,
Florschuetzia
sp.
pollen
was
younger
than
the
earliest
fossil
wood
(Sonneratioxylon
preapetalum)
assigned
to
Sonneratia
from
the
Deccan
Intertrappean
Bed,
the
early
Paleo-
cene
(67.3–63.8
Ma)
of
India
(Awasthi,
1969).
Muller
(1964,
1978)
and
Morley
(2000)
emphasized
the
lengthy
presence
of
Florschuetzia
and
the
subsequent
appearance
of
Sonneratia
in
southeastern
Asia,
which,
however,
does
not
necessarily
sup-
port
a
southeastern
Asian
origin
of
the
genus
(Graham,
2013).
However,
Ellison
et
al.
(1999)
asserted
that
the
modern
distribu-
tions
result
almost
entirely
from
vicariance
events,
not
consistent
with
an
Indo-West
Pacific
(IWP)
centre
of
origin
for
mangrove
species
and
associated
gastropods.
Considering
very
few
fossils
of
gastropod
were
preserved
in
anoxic,
acidic,
peaty
soils
which
may
rapidly
dissolve
the
calcium
carbonate
shells
(Plaziat
et
al.,
1983),
we
focus
on
relatively
rich
plant
fossils
in
this
paper,
espe-
cially
on
some
well
preserved
pollen
grains.
We
re-assess
fossil
pollen
records
of
Florschuetzia/Sonneratia
with
an
emphasis
on
the
historical
biogeography
of
Sonneratia
based
on
the
paly-
nological
interpretation,
and
further
evaluate
the
evolutionary
trends
of
extinct
genus
Florschuetzia
towards
Sonneratia
by
reviewing
the
published
fossil
pollen
data
and
comparing
them
with
the
extant
species.
2.
General
systematics
of
extant
genus
Sonneratia
and
its
biogeography
2.1.
General
systematics
of
Sonneratia
The
extant
genus
Sonneratia
belongs
to
Sonneratioideae
(Engl.
et
Gilg)
S.A.
Graham,
Thorne
et
Reveal,
as
one
of
the
monogeneric
subfamilies
of
Lythraceae
within
the
Myr-
tales
validated
by
Graham
et
al.
(1998).
Sonneratia
and
inland
Duabanga
were
formerly
placed
in
Sonneratiaceae
independent
from
the
family
Lythraceae,
but
both
of
them
were
assigned
in
their
own
subfamily
based
on
morphological
and
molecular
studies
on
Lythraceae
sensu
lato
(Thorne,
1976,
1981,
1992;
Dahlgren
and
Thorne,
1984;
Graham
et
al.,
1993a,b).
The
phy-
logenetic
analysis
of
Sonneratia
in
the
Internal
Transcribed
Spacer
(ITS)
sequences
indicates
that
the
Sonneratiaceae
should
be
included
within
the
Lythraceae
instead
of
a
distinct
family
(Shi
et
al.,
2000).
Living
Sonneratia
consists
of
around
nine
species
(Fig.
1),
including
three
putative
hybrids,
namely
Son-
neratia
alba,
S.
apetala,
S.
caseolari,
S.
griffithii,
S.
×
gulngai,
S.
×
hainanensis,
S.
lanceolata,
S.
ovata,
and
S.
×
urama.
It
is
restricted
to
mangrove
communities,
often
forming
seaward
fringing,
and
some
of
them
have
a
very
restricted
range,
such
as
S.
griffithii
(Fig.
1e,
specimen
from
Herbarium
in
Prince
of
Songkla
University,
Thailand).
Species
in
Sonneratia
are
notable
for
their
large
showy
flowers
with
numerous
red
or
white
stamens
(Fig.
1)
and
the
berry-shaped
fruit
seated
on
a
persistent
calyx
with
6–8
erect
pointed
lobes
(Duke,
2006).
They
are
recognized
by
the
tall
conical
pneumatophores
aris-
ing
from
horizontal
roots.
Key
to
these
species
was
described
in
the
early
publications
(Tomlinson,
1986)
and
recent
illustra-
tion
books
for
the
field
guide
(Duke,
2006;
Wang
and
Wang,
2007);
Wang
and
Chen
(2002)
studied
the
systematics
and
biogeography
of
the
Sonneratiaceae
in
China.
However,
the
systematic
descriptions
of
some
species
probably
need
to
be
revised
to
take
into
account
their
geographical
variations,
such
as
S.
caseolaris
and
hybrids
of
S
×
gulngai,
S.
×
urama
and
S.
×
hainanensis.
Fig.
1.
Showy
flowers
of
living
Sonneratia
and
the
specimen
of
Sonneratia
griffithii.
(a)
Sonneratia
alba.
(b)
S.
apetala.
(c)
S.
caseolaris.
(d)
S.
×
gulngai.
(e)
S.
griffithii,
specimen
from
PSU-Herbarium,
Thailand,
photo
by
Jarearnsak
Sae
Wai.
(f)
S.
×
hainanensis.
(g)
S.
ovata.
(h)
S.
×
urama,
reproduced
from
Duke
(2006).
(i)
S.
lanceolata,
reproduced
from
Duke
(2006).

L.
Mao,
S.Y.
Foong
/
Palaeoworld
22
(2013)
133–143
135
2.2.
Biogeography
of
modern
Sonneratia
Sonneratia
is
an
endemic
genus
in
the
Indo-West
Pacific
(IWP)
region
(Fig.
2),
which
is
the
frontal
element
in
the
estuary
mangrove
swamps
or
tidal
inlets
and
bays
from
the
coastal
trop-
ical
East
Africa
to
Indo-Malaysia,
southern
China,
New
Guinea,
Australia,
and
islands
in
the
Western
Pacific
(Duke,
2006).
Son-
neratia
grows
mostly
along
the
banks
of
tidal
rivers,
creeks,
and
within
the
sheltered
bays
of
offshore
islands
and
reef
cays.
It
is
a
typical
constituent
of
mangrove
communities
throughout
its
geographical
range.
Mangrove
distributions
are
influenced
not
only
by
long
distance
dispersal
and
establishment
success,
but
also
by
geo-
graphical
conditions
and
past
changes
in
these
conditions
(Duke
et
al.,
1998).
Geological
and
climatic
changes
have
greatly
influenced
the
modern
distribution
of
species
and
the
evolu-
tion
of
mangrove
habitat.
Mangrove
evolution,
diversification,
and
dispersal
apparently
were
accelerated
by
continental-drift
(Duke
et
al.,
1998).
The
most
widespread
species
in
the
genus
is
Sonneratia
alba,
from
East
Africa
to
India
and
through
Southeast
Asia
(including
southern
China
and
Indonesia)
to
the
western
islands
in
the
Pacific
Ocean
including
New
Cale-
donia,
the
Solomon
Islands
and
northern
Australia
(Fig.
2),
mostly
at
low
tidal
contours
within
the
frontal
stands
of
down-
stream,
the
lower
reaches
of
estuaries
and
offshore
island
enclaves
in
the
regions
of
moderate
to
high
rainfall
where
tidal
ranges
exceed
1
m
(Duke,
2006).
Sonneratia
caseolaris
occurs
frequently
in
the
frontal
stands
in
some
upstream
estu-
arine
situations
subjected
to
high
levels
of
freshwater
runoff,
from
the
west
coast
of
India
to
southern
China
and
the
western
islands
in
the
Pacific
Ocean,
including
New
Guinea
and
north-
ern
Australia
(Fig.
2).
Interestingly,
the
habitat
distribution
of
Sonneratia
×
gulngai,
which
is
a
hybrid
of
S.
alba
and
S.
case-
olaris,
overlaps
those
of
the
parent
species
in
that
it
is
found
in
mid
intertidal,
intermediate
estuarine
situations.
Sonneratia
ovata
prefers
a
different
habitat,
usually
occurring
at
the
high
tide
margin
in
estuaries
and
is
occasionally
found
on
the
banks
of
tidal
creeks
and
rivers
on
muddy
soils
inundated
only
by
spring
tides.
It
is
scattered
in
widely
separate
localities
from
China
and
Thailand
through
the
Peninsular
Malaysia,
the
Riau
Archipelago,
Java,
and
Borneo,
to
Sulawesi,
the
Moluccas,
and
Daru
Island
and
Milne
Bay
in
New
Guinea
(Fig.
2).
Sonner-
atia
lanceolata
occurs
at
lower
tidal
contours
in
the
upstream
estuarine
position.
It
is
relatively
unknown
beyond
Australia,
Indonesia,
and
New
Guinea
(Duke,
2006;
Fig.
2).
Sonnera-
tia
×
urama
is
the
hybrid
of
S.
alba
and
S.
lanceolata
(Duke,
1994).
This
putative
hybrid
has
been
reported
in
Australia,
Indonesia,
and
Papua
New
Guinea
(Fig.
2),
essentially
reflect-
ing
the
range
of
the
less
common
parent,
S.
lanceolata
(Duke
and
Jackes,
1987).
S.
griffithii
has
a
restricted
distribution
along
the
shores
of
the
Andaman
Sea,
northward
to
Bengal
and
south-
ward
to
the
upper
Malay
Peninsula,
Thailand
(Tomlinson,
1986;
Fig.
2).
3.
Fossil
pollen
records
of
Florschuetzia
and
Sonneratia
The
genus
Florschuetzia
has
been
named
in
memory
of
the
late
Prof.
Dr.
F.
Florschfitz,
the
founder
of
palynology
in
the
Netherlands
(Germeraad
et
al.,
1968).
Florschuetzia
is
the
pollen
from
genus
of
several
species
known
mainly
from
abundant,
varied
fossil
grains
in
the
tropics
of
Southeast
Asia,
and
also
in
the
Paleocene
of
southern
France
(Figs.
3
and
4).
There
are
seven
known
species
of
the
genus
Florschuetzia,
namely
F.
trilobata,
F.
semilobata,
F.
levipoli,
F.
meridionalis
from
Borneo
(Germeraad
et
al.,
1968),
F.
reticulata
from
Sulawesi,
Indonesia
(Sohma,
1973),
F.
minutes
from
India
(Rawat
et
al.,
1977),
and
F.
claricolpata
from
central
Japan
(Yamanoi,
1984).
F.
reticulata
and
F.
minutes
are
not
accepted
based
on
Graham’s
evaluation
(Graham,
2013).
The
former
six
species
do
not
have
a
colpate
aperture,
although
colpoid
grooves
may
occur
in
F.
trilobata.
However,
according
to
SEM
observation
by
Yamanoi,
F.
claricolpata
is
characterized
by
a
distinctly
colpate
aperture
(Yamanoi,
1984;
also
Fig.
3).
More
detailed
information
on
this
species
is
presented
in
the
discussion.
Fig.
2.
Geographical
distribution
(in
green)
of
extant
genus
Sonneratia
in
the
Indo-West
Pacific
(IWP)
region
(based
on
Duke,
2006;
Tomlinson,
1986).
(For
interpretation
of
the
references
to
color
in
this
figure
legend,
the
reader
is
referred
to
the
web
version
of
the
article.)

136
L.
Mao,
S.Y.
Foong
/
Palaeoworld
22
(2013)
133–143
Fig.
3.
Fossil
pollen
of
Florschuetzia,
Trilobapollis
and
Quaternary
sub-fossil
of
Sonneratia.
(a)
F.
semilobata,
holotype,
early
Miocene,
Borneo,
23–24
␮m.
(b)
F.
claricolpata,
holotype,
middle
Miocene,
Japan,
25–34
␮m.
(c)
Florschuetzia
sp.
2,
middle
Miocene,
Taiwan
(China),
35
␮m.
(d)
F.
trilobata,
holotype,
Oligocene,
Borneo,
28–35
␮m.
(e)
F.
meridionalis,
holotype,
Miocene,
Borneo,
35–60
␮m.
(f)
F.
levipoli,
holotype,
Miocene,
Borneo,
30–50
␮m.
(g)
Trilobapollis
ellipticus,
ancestral
Florschuetzia
or
variants?
Paleocene,
Leizhou
Peninsula,
China.
(h)
Florschuetzia
sp.
1,
late
Paleocene,
France,
26
␮m.
(i,
j)
Early
Pleistocene
sub-fossil
from
Leizhou
Peninsula,
southern
China,
Sonneratia
sp.
3
(cf.
S.
alba).
(k–n)
Late
Quaternary
sub-fossil
from
Cambodia,
scale
bar
=
10
␮m;
(k,
l)
Sonneratia
caseolaris
in
polar
view
and
equatorial
view,
respectively;
(m)
S.
alba;
(n)
Sonneratia
sp.
(cf.
S.
×
gulngai
type).
Red
arrow:
mesoporium
meridional
ridge
(pollen
diagnostic
of
S.
alba
and
S.
×
hainanensis);
blue
arrow:
structurally
discontinuous
between
equatorial
and
polar
regions;
light
green
arrow:
pseudocolpus.
Photo
sources:
Gruas-Cavagnetto
et
al.,
1988
(Florschuetzia
sp.
1);
Yamanoi,
1984
(F.
claricolpata);
Sun
et
al.,
1980
(Trilobapollis
ellipticus);
Weiming
Wang
(Florschuetzia
sp.
2);
Wang
and
Zhang,
1998
(Sonneratia
sp.
3);
Germeraad
et
al.,
1968
(all
others
except
those
otherwise
specified).
(For
interpretation
of
the
references
to
color
in
this
figure
legend,
the
reader
is
referred
to
the
web
version
of
the
article.)
Florschuetzia
trilobata
in
Central
Java
was
first
associated
with
Sonneratia
in
the
middle
Eocene
deposits
of
southeastern
Asia,
but
Muller
(1981a)
viewed
it
as
not
quite
fully
Sonneratia.
F.
levipoli
from
the
early
Miocene
of
Borneo
(ca.
19
Ma)
is
the
earliest
fossil
pollen
directly
attributable
to
living
species
S.
caseolaris,
then
followed
by
F.
meridionalis
from
the
middle
Miocene
of
Borneo
attributable
to
S.
alba
according
to
Muller
(1984)
and
Morley
(2000).
The
Miocene
pollen
of
F.
levipoli
(S.
caseolaris)
in
Borneo
marks
the
dawn
of
extant
Sonneratia
as
an
obligate
mangrove
genus,
while
earlier
Florschuetzia
trilobata
being
described
from
fresh-water
and
saline
deposits
(Graham,
2013).
The
earliest
record
of
Florschuetzia
pollen
attributable
to
Sonneratiaceae
is
reported
by
Gruas-Cavagnetto
et
al.
(1988)
in
the
late
Paleocene
of
southern
France
(Thanetian,
58.7–55.8
Ma),
but
as
Graham
(2013)
pointed
out
that
it
is
difficult
to
confirm
the
Sonneratia-like
pollen
based
on
the
LM
picture
of
the
isolated
pollen
grains,
which
are
earlier
than
the
later
Florschuetzia
spp.
from
Asia
(see
question
marked
pollen
grain
in
Fig.
3h
and
Fig.
4).
Zaklinskaja
(1978)
reported
Florschuetzia
in
the
Eocene
of
West
African;
Yamanoi
et
al.
(1980)
and
Yamanoi
(1984)
studied
the
middle
Miocene
Florschuetzia
in
central
Japan;
Liu
and
Yang
(1999)
studied
the
Florschuetzia
trilobata
in
the
pollen
flora
of
southern
China,
which
suggests
mangrove
development
in
the
low
area
of
the
Bose
Basin
since
the
late
Eocene.
The
frequent
pollen
record
is
explained
by
its
current
over-
representation
reported
in
modern
pollen
assemblages
(Caratini
et
al.,
1973).
The
botanical
significance
of
Florschuetzia
trilo-
bata
is
evaluated
according
to
Germeraad
et
al.
(1968)
and
Muller
(1978,
1981b),
as
the
oldest
type,
F.
trilobata,
is
linked
with
an
intermediate
between
Lythraceae
and
its
subfamily
Son-
neratiaceae.
Although
it
is
recorded
in
the
same
period
in
early
Miocene
as
the
unambiguous
Sonneratia
pollen,
it
has
been
sug-
gested
as
an
ancestor
of
the
evolved
Sonneratia.
Germeraad
et
al.
(1968)
reported
Florschuetzia
in
assemblages
along
with
Son-
neratia
and
Rhizophora.
Consequently
it
is
suggested
that
F.
trilobata
was
probably
produced
by
a
true,
but
extinct,
man-
grove
plant
which
was
replaced
gradually
by
the
extant
species
in
the
end
of
Paleogene
and
afterwards.

L.
Mao,
S.Y.
Foong
/
Palaeoworld
22
(2013)
133–143
137
Apparently,
Sonneratia-like
pollen
of
Florschuetzia
reported
by
Gruas-Cavagnetto
et
al.
(1988)
from
the
late
Paleocene
deposits
in
south
central
France
(58.7–55.8
Ma)
needs
to
be
con-
firmed
(question
marked
in
Figs.
3
and
4),
as
well
as
F.
reticulata
(Sohma,
1973)
of
the
Quaternary
and
F.
minutus
(Rawat
et
al.,
1977)
from
the
Eocene
of
India.
Some
investigators
reported
the
early-mid
Quaternary
Son-
neratia
sp.
(cf.
alba)
in
the
coastal
areas
of
South
China
and
SCS
(Zhou,
1988;
Sun,
1991;
Zheng
and
Zhou,
1995;
Lei,
1998;
Wang
and
Zhang,
1998;
Zheng
and
Li,
2000)
(see
also
Fig.
3i,
j),
where
it
is
out
of
the
latitudinal
limits
for
this
genus.
Impacts
of
climatic
events
and
consequent
sea
level
changes
as
a
result
of
the
Quaternary
glaciations
on
the
tropical
man-
grove
habitats
are
important
in
shaping
the
current
distribution
range
of
Sonneratia.
In
the
Holocene
deposits,
mangrove
sub-
fossil
pollen
grains
are
preserved
as
perfect
as
the
living
plants’
(Fig.
3k–n).
The
pollen
grains
of
Sonneratia
spp.
were
extracted
from
the
estuary
sediments
in
the
upper
Mekong
River
delta,
Cambodia
(Li
et
al.,
2012).
Table
1
presents
detailed
fossil
information
on
Florschuetzia
and
Sonneratia,
including
some
macrofossils.
4.
Discussion
4.1.
Notes
on
Florschuetzia
claricolpata
from
Japan
and
Florschuetzia
spp.
from
southern
China
Yamanoi
(1984)
published
the
fossil
pollen
of
Florschuetzia
claricolpata
from
the
middle
Miocene
of
central
Japan
(Fig.
3b,
cf.
Fig.
3c
from
middle
Miocene
of
Taiwan),
which
is
consid-
ered
comparable
to
extant
Sonneratia
alba
(also
see
Fig.
4,
top
right).
Moreover,
F.
claricolpata
accounted
for
41%
of
the
pollen
assemblage,
with
other
mangrove
pollen
Excoecaria
(12%)
and
Bruguiera
(0.5%)
from
the
same
assemblage
(Yamanoi,
1984).
F.
claricolpata
is
described
as
heterocolpate,
isopolar,
porate,
and
rounded
triangular
in
polar
view,
which
is
very
close
to
Lagerstroemia.
Yamanoi
(1984)
discussed
morphological
over-
laps
between
the
two
highly
similar
pollen
types.
Muller
(1981b)
suggested
that,
since
some
pollen
types
of
Lagerstroemia
are
similar
to
Florschuetzia
trilobata
but
have
a
colpate
aperture,
these
genera
are
likely
to
have
a
common
ancestral
matrix.
The
absence
of
colpi
is
confirmed
in
Sonneratia
caseolaris
and
S.
ovata
with
SEM
by
Muller
(1981a).
However,
pollen
grains
of
S.
alba
and
S.
griffithii
have
a
colpal
part
though
it
is
poorly
developed
(Muller,
1969,
1981a).
To
examine
the
colpi
in
the
extant
S.
alba,
Yamanoi
(1984)
observed
100
well
expanded
pollen
grains
sampled
from
Semakau
Island
(Singapore)
using
SEM
and
calculated
the
following
frequencies
of
presence
of
colpi:
distinct
(5%),
more
or
less
distinct
(30%),
faint
(59%),
and
absent
(6%).
These
facts
and
Muller’s
phylogenetic
view
of
Lagerstroemia
and
Sonneratia
(Muller,
1981b)
suggest
that
the
presence
of
colpi
may
be
more
frequent
in
the
ancestral
populations
of
S.
alba
than
the
recent
ones.
Yamanoi
(1984)
concluded
that
F.
claricolpata
has
a
definite
affinity
to
Sonnera-
tia
and
shares
more
pollen
morphological
features
with
S.
alba,
especially
with
subtypes
D
and
F
described
by
Muller
(1969)
than
any
other
species,
and
that
a
similar
subtype
already
existed
since
the
middle
Miocene.
Sun
et
al.
(1980)
established
genus
Trilobapollis
of
the
early
Paleogene
from
Leizhou
Peninsula,
northern
South
China
Sea
(SCS),
which
resemble
Florschuetzia
trilobata
based
on
their
trilobate-shape,
except
for
the
absence
of
apertures
in
Trilobapollis.
It
is
speculated
that
Trilobapollis
could
be
an
ancestral
type
of
F.
trilobata
(Yamanoi,
1984);
or
it
may
be
a
variant
of
F.
trilobata
given
colpi
recognizable
under
SEM.
Sonneratia
first
appears
on
the
Paleocene
island
continent
of
India
(Prakash,
1960),
earlier
than
other
closely
related
pollen
of
the
region
(such
as
Santalumidites,
Sonneratioipollis,
Jugopolis
and
Florschuetzia)
that
appeared
during
Eocene
(Venkatachala
and
Kar,
1968;
Venkatachala
and
Rawat,
1972;
Rawat
et
al.,
1977).
However,
Santalumidites
from
the
Eocene
of
Australia
and
Sonneratioipollis,
Jugopolis
and
Florschuetzia
from
the
Eocene
of
India
show
no
clear
affinity
with
the
extant
Sonner-
atia
(Muller,
1981a).
Therefore,
it
seems
that
the
distribution
range
of
Florschuetzia
was
limited
along
the
equatorial
regions
of
Southeast
Asia.
F.
claricolpata
from
Japan
and
F.
trilobata,
F.
levipoli,
and
F.
meridionalis
from
the
Pearl
River
mouth
basin,
South
China
(Lei,
1998;
Liu
and
Yang,
1999),
together
with
Trilobapollis
from
early
Paleogene
(Sun
et
al.,
1980),
possibly
belong
to
an
ancestral
type
of
Florschuetzia;
all
of
them
were
distributed
beyond
the
known
geographic
range
of
Florschuet-
zia,
suggesting
coastal
environment
with
tropical
climate
in
southwestern
Japan
and
northern
SCS
during
Paleogene
through
middle
Miocene
(Yamanoi,
1984).
4.2.
Pollen
evolution
of
Florschuetzia
towards
Sonneratia
Florschuetzia
is
at
the
base
of
a
complex
of
emerging
pollen
types
that
evolved
into
pollen
forms
of
Lagerstroemia,
Sonner-
atia,
and
possibly
Trapa
and
other
genera
(Germeraad
et
al.,
1968;
Muller,
1978,
1984;
Morley,
2000;
Graham,
2013).
The
earliest
Florschuetzia
pollen
is
reported
from
the
late
Paleocene
Thanetian
(58.7–55.8
Ma)
of
south
central
France,
which
was
situated
at
the
southern
border
of
the
Tethys
(Gruas-Cavagnetto
et
al.,
1988;
Plaziat
et
al.,
2001;
Graham,
2013).
The
pollen
morphology
and
pollen
evolution
of
Florschuetzia
towards
the
extant
genus
Sonneratia
are
illustrated
in
Figs.
3
and
4.
How-
ever,
it
needs
confirmation
on
whether
or
not
there
is
an
early
form
of
Florschuetzia
with
no
distinctive
polar
caps,
prominent
mesocolpal
ridges,
and
the
well-defined
verrucate
equatorial
belt
of
later
Florschuetzia
species
(Graham,
2013).
Fossil
pollen
grains
of
Florschuetzia
are
abundant
and
highly
varied
struc-
turally
in
the
Eocene
deposits
from
southeastern
Asia
(Muller,
1981b;
Guleria
et
al.,
1996;
Morley,
2000).
The
oldest
ones
are
morphologically
more
diverse
than
later
ones
(Graham,
2013).
The
oldest
species
in
Asia
is
F.
trilobata,
which
can
be
hypothesized
as
an
evolved
Trilobapollis
ellip-
ticus,
which
was
found
from
Leizhou
Peninsula
in
the
early
Paleocene
of
China
(Sun
et
al.,
1980).
The
species
variation
and
speciation
might
be
triggered
by
extreme
climatic
events,
such
as
the
Paleocene–Eocene
Thermal
Maximum
(PETM),
which
might
have
played
a
significant
role
in
the
early
evolution
of
Florschuetzia.
F.
trilobata
is
considered
to
be
the
precursor
to

138
L.
Mao,
S.Y.
Foong
/
Palaeoworld
22
(2013)
133–143
Table
1
Fossil
Florschuetzia
and
Sonneratia
(pollen
records
updated
based
on
Graham,
2013).
Fossil
name Organ Period
Locality
Data
source
Remark
Florschuetzia
(Extinct),
see
also
Sonneratia
Florschuetzia
sp.
(as
Sonneratiaceae)
Pollen
late
Paleocene
France
Gruas-Cavagnetto
et
al.,
1988
Accepted?
Florschuetzia
trilobata
Pollen
middle
Eocene
to
early
Miocene
Borneo,
Java,
Assam,
Thailand,
China,
Red
Sea,
Nile
Delta
Germeraad
et
al.,
1968;
Muller,
1981a;
Watanasak,
1990;
Handique,
1993;
Lei,
1998;
Liu
and
Yang,
1999;
Morley,
2000
Accepted
Florschuetzia
minutus Pollen
Eocene
India
Zaklinskaja,
1978 Unconfirmed
Florschuetzia
cf.
meridionalis Pollen
Eocene
India
Rawat
et
al.,
1977 Unconfirmed
Florschuetzia
semilobata Pollen
early
Miocene Borneo,
Thailand Germeraad
et
al.,
1968;
Muller,
1981a;
Morley,
2000 Accepted
Florschuetzia
levipoli
(holotype) Pollen
early
Miocene
to
Pliocene Borneo,
China Germeraad
et
al.,
1968;
Muller,
1978;
Lei,
1998 Accepted
Florschuetzia
levipoli Pollen
late
Miocene
to
Recent New
Guinea Khan,
1974 Accepted
Florschuetzia
meridionalis
(holotype) Pollen
middle
Miocene Borneo
Germeraad
et
al.,
1968;
Muller,
1978
Unconfirmed
Florschuetzia
meridionalis Pollen
late
Miocene
to
Recent New
Guinea
Khan,
1974
Accepted
Florschuetzia
sp.
Pollen
middle
Miocene
Thailand
Songtham
et
al.,
2005
Unconfirmed
Florschuetzia
claricolpata
(holotype)
Pollen
middle
Miocene
Japan
Yamanoi,
1984
Accepted
Florschuetzia
reticulata
Pollen
Quaternary
Indonesia
Sohma,
1973
Unconfirmed
Sonneratia,
including
Sonneratioxylon;
see
also
Florschuetzia
Sonneratioxylon
turonicum
Wood
late
Cretaceous
Uzbekistan
Shelomentseva,
1992
Unconfirmed
Sonneratia
meghalayensis
Leaf
late
Paleocene
India
Ambwani,
1991
Sonneratiorhizos
raoi Root
Paleocene India
Chitaley,
1968 Unconfirmed
Sonneratiaceae
(Florschuetzia
sp.)
Pollen
late
Paleocene
France
Gruas-Cavagnetto
et
al.,
1988
Accepted?
Sonneratioxylon
preapetalum Wood Paleocene
(Danian)
to
Miocene India,
Thailand,
Sumatra Awasthi,
1969;
Biradar
and
Mahabale,
1973;
Kramer,
1974;
Shete
and
Kulkarni,
1982;
Bande
and
Prakash,
1984;
Lakhanpal
et
al.,
1984;
Mehrotra,
1988;
Vozenin-Serra
et
al.,
1989;
Guleria,
1991;
Srivastava,
2008
Accepted
but
not
all
Sonneratioxylon
aubrevevillei
Wood
middle
Eocene
Libya
Louvet,
1970
Accepted
aff.
Sonneratia:
Sonneratioipollis Pollen
Eocene
India
Venkatachala
and
Kar,
1968
Unconfirmed
aff.
Sonneratia:
Jugopollis
tetraporites
Pollen
Eocene
India
Venkatachala
and
Rawat,
1972
Sonneratia
kyushuensis Wood
early
Oligocene
Japan
Srivastava
and
Suzuki,
2001
Accepted
Sonneratioxylon
prambachense
Wood
late
Oligocene
Austria
Hofmann,
1952
Unconfirmed
Sonneratia
caseolaris
Pollen
early
Miocene
Borneo
Germeraad
et
al.,
1968;
Muller,
1978;
Morley,
2000
Accepted
Sonneratia
alba
Pollen
middle
Miocene
Borneo
Muller,
1978
Accepted
Sonneratioxylon
dakshinense
Wood
Miocene
to
Pliocene
India
Ramanujam,
1956;
Awasthi,
1969;
Mehrotra,
1988
Unconfirmed
Sonneratioxylon
dudukurense Wood Tertiary India
Rao
and
Ramanujam,
1966;
Mehrotra,
1988 Unconfirmed
Sonneratia
sp.
Pollen
late
Pleistocene
China,
Thailand,
India
Sun,
1991;
Zheng
and
Zhou,
1995;
Wang
and
Zhang,
1998;
Zheng
and
Li,
2000;
Kumaran
et
al.,
2005;
Rugmai
et
al.,
2008
Accepted
Sonneratioxylon
duabangoides Wood Paleocene India
Shallom,
1963;
Chitaley,
1968 Unconfirmed

L.
Mao,
S.Y.
Foong
/
Palaeoworld
22
(2013)
133–143
139
Fig.
4.
Tentative
interpretation
of
the
pollen
evolution
from
Florschuetzia
to
the
extant
genus
Sonneratia
as
proposed
by
Germeraad
et
al.
(1968)
and
Muller
(1984),
also
showing
fossil
pollen
of
Trilobapollis,
possible
ancestral
Florschuetzia?
and
interspecific
diagnostic
of
two
pollen
species
of
living
Sonneratia
caseolaris
and
S.
alba.
Scale
bar
=
10
␮m
(SEM
equatorial
views),
1
␮m
(SEM
detailed
polar
cap),
20
␮m
(LM).
More
species
in
this
genus,
see
Mao
et
al.
(2012).