Review of Palaeobotany and Palynology 110 (2000) 111–138
www.elsevier.nl/locate/revpalbo
Vegetation stability in the Southeastern Brazilian coastal area
from 5500 to 1400 14C yr BP deduced from charcoal analysis
Rita Scheel-Ybert *
UMR 5059, Université Montpellier II. Institut de Botanique, 163 rue Auguste Broussonnet, 34090 Montpellier, France
Received 6 May 1999; received in revised form 10 December 1999; accepted for publication 1 February 2000
Abstract
Charcoal analysis of six shell mounds showed that no major changes of the mainland vegetation ecosystem have
taken place along the southeastern Brazilian coast (22°53∞–22°57∞S, 42°03∞–42°33∞W ) from 5500 to 1400 14C yr BP.
These shell mounds have been occupied by sedentary fisher–gatherer–hunters. Charcoal fragments retrieved from
vertical profiles in the archaeological sites were examined; taxonomic determinations were based on a reference
collection of charred woods and a program for computer-aided identification. Charcoal assemblages of all the studied
sites present taxa from various restinga vegetation types, mangroves, xeromorphic coastal forest, and inland Atlantic
Forest. The restinga ecosystem, characteristic of the Brazilian coast, is associated with sandy beach ridges; the restinga
forest was much more abundant during the studied period than nowadays. The charcoal assemblages represent mainly
the local vegetation; a regional reconstruction depends on the study of numerous sites. In the Cabo Frio region, open
restinga taxa are more abundant in the Sambaqui do Forte, while forest elements are more important in the Sambaquis
Salinas Peroano and Boca da Barra. The sites studied in the Arraial do Cabo (Sambaqui da Ponta da Cabeça) and
in the Saquarema regions (Sambaquis da Pontinha and da Beirada) show that open restinga formations were locally
predominant. A comparison of multivariate analysis applied to both charcoal assemblages and to phytosociological
data of the extant vegetation showed a good correspondence between the charcoal spectra and the present vegetation.
The high taxonomic diversity of archaeological charcoal samples and numerous fragments showing traces of decay
before charring suggests that aleatory gathering of dead wood constituted the main source of firewood for fisher–
gatherer–hunters populations. Condalia sp. was probably selected for cultural reasons.
The only significant fluctuations on the charcoal spectra relate to the mangrove vegetation. Two relatively humid
episodes (recorded from ca. 5500 to 4900/4500 and from ca. 2300 to 2000 14C yr BP), intercalated by two episodes of
increased dryness with increased lagoon salinity (from ca. 4900/4500 to 2300 and from ca. 2000 to 1400 14C yr BP)
were recorded in the Cabo Frio region. The changes in mangrove vegetation cannot be attributed to sea-level
variations, for the three regressive and the two transgressive episodes identified for the Brazilian coast during this
period are not in phase with the development of mangroves. The stability of the mainland vegetation ecosystem is
probably due to the edaphic character of the coastal environments, which makes coastal formations much more
resistant to climatic variations and less sensitive to climatic change. We propose that this environmental stability was
a decisive factor in the maintenance of the fisher–gatherer–hunter sociocultural system. © 2000 Elsevier Science B.V.
All rights reserved.
Keywords: archeological site; Brazil; charcoal analysis; Holocene; palaeoclimate; palaeoenvironment
* Fax: +33-4-67-54-35-37.
E-mail address: [email protected] (Rita Scheel-Ybert)
0034-6667/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved.
PII: S0 0 3 4 -6 6 6 7 ( 0 0 ) 0 0 00 4 - X
112
R. Scheel-Ybert / Review of Palaeobotany and Palynology 110 (2000) 111–138
1. Introduction
The palaeoenvironmental history of Brazil is
still poorly known, but various studies have
demonstrated that significant climatic changes
affected this country during previous millennia
(e.g. Soubiès, 1980; Servant et al., 1989; Absy
et al., 1991; Ledru et al., 1995; Ferraz-Vicentini
and Salgado-Labouriau, 1996; Behling, 1997;
Ybert et al., in press). Climatic changes led to
vegetational modifications in all studied regions.
The Brazilian coast of the southeastern portion
of the Rio de Janeiro state is presently characterized
by geological, climatic and ecological features that
allow it to be considered as an important area for
palaeoenvironmental study (Martin et al., 1988;
Tasayco-Ortega, 1996). Modifications on the lagoon
sedimentation in several lagoonal cores suggest
regional climatic variations during the Holocene
(Tasayco-Ortega, 1996). This author proposes five
dry episodes, attributed to a reinforcement of semiarid conditions (3600–3500, 3100–3000, 1200–1100
and 600–500 14C yr BP ), and two rainy episodes
(2300–2100 and 700–600 14C yr BP). The main
objective of the present study was to reconstruct the
palaeoenvironmental evolution of the plant cover in
this region during the last 6000 years.
Charcoal analysis of six archaeological sites and
of a modern surface sample of another site from
this region are presented. This discipline allows a
good reconstruction of the local woody vegetation,
and comparison with phytosociological data is
quite direct. We will refer to it from this point on
as anthracology (from the Greek word for charcoal,
anthrakos). Many anthracological studies, mostly
in temperate and Mediterranean climates, have
already demonstrated the value of these analyses
for palaeoecological reconstruction (e.g. Vernet
and Thiébault, 1987; Vernet, 1992; Hopkins et al.,
1993; Badal et al., 1994; Figueiral, 1995; Heinz
and Thiébault, 1998), but these studies are still
very rare in the tropics, where floristic diversity is
far greater than in temperate regions.
2. Regional setting
Four of the seven studied sites are situated in
the Cabo Frio region (22°53∞S, 42°03∞W ) ( Fig. 1):
the Sambaqui do Forte is localized between the
Itajuru Channel and the Atlantic Ocean, on the
west margin of the channel and the north side of
Praia do Forte; the Sambaqui Salinas Peroano,
Sambaqui Boca da Barra and Sambaqui do Meio,
on the east side of the Itajuru Channel, are situated
on low inland crystalline hills, no more than 500 m
apart from each other. One site, the Sambaqui da
Ponta da Cabeça is situated at the Arraial do Cabo
Peninsula (22°57∞S, 42°14∞W ), on a crystalline hill
near the Massambaba Beach. Two sites, Sambaqui
da Beirada and Sambaqui da Pontinha, are situated
at the Saquarema region (22°55∞S, 42°33∞W ), landwards on the Pleistocene beach ridge between the
Saquarema Lagoon and the ocean, ca. 400 m from
each other.
The climate of most of the coastal zone of Rio
de Janeiro State is tropically wet, hot and rainy in
summer with a mild dry season in winter, labeled
Aw in Köppen’s classification (Barbiére, 1984). In
the Saquarema region, the mean annual temperature is 24–26°C, and the precipitation is
1000 mm/year (Sá, 1992). The Cabo Frio and
Arraial do Cabo region present a particularly dry
climate, BSh in Köppen’s classification, that is a
variant of the semi-arid hot climate due to upwelling of cold waters along the coast, which reduces
local precipitation. The mean annual temperature
is 25°C, and the precipitation rarely exceeds
800 mm/year (Barbiére, 1984).
The regional geology consists of a Precambrian
crystalline basement of granite-gneiss rocks, with
local alkaline intrusions. The Cabo Frio and
Arraial do Cabo region presents three distinct
physiographic units: (1) sandy coastal plain (beach
ridges, dunes) and lowlands (tidal areas, lagoons,
alluvial deposits); (2) low hills of the Búzios, Cabo
Frio and Arraial do Cabo peninsulas and coastal
islands; and (3) inland hills up to ca. 500 m. Only
units (1) and (3) are present in the Saquarema
region.
The coastal plain presents two series of lagoons
isolated by two sand barriers. The hypersaline
Araruama Lagoon is the largest in the region. It
is connected to the sea by a narrow passage, the
Itajuru Channel, but the tides are not strong
enough to result in the turnover of the lagoon
waters. Fresh water comes almost exclusively from
R. Scheel-Ybert / Review of Palaeobotany and Palynology 110 (2000) 111–138
113
Fig. 1. Geographical location of the studied sites. (1) Sambaqui do Forte; (2) Sambaqui Boca da Barra; (3) Sambaqui do Meio; (4)
Sambaqui Salinas Peroano; (5) Sambaqui da Ponta da Cabeça; (6) Sambaqui da Beirada; (7) Sambaqui da Pontinha.
rain water streaming, for there are no important
rivers in the region (Barroso, 1987). The very
shallow Saquarema Lagoon is connected to the
sea by a transient passage. It is strongly influenced
by the alternation of continental and marine
waters, and its salinity may vary from 0 to 35‰
(Moreira and Carmouze, 1991).
The palaeogeographical development of these
lagoons is closely related to Quaternary sea-level
fluctuations ( Turcq et al., in press). During the
Holocene, the sea level rose progressively to the
present mean level, which was reached by 7000
14C yr BP (Fig. 2) (Martin et al., 1979/80, 1997;
Turcq et al., in press). Subsequently, three main
transgressive episodes occurred (7000–5100, 3900–
3600 and 2700–2500 14C yr BP), intercalated by
three regressive episodes (5100–3900, 3600–2700
and after 2500 14C yr BP). It was not possible to
reconstruct a record of the relative sea-level curve
for the sector between Sepetiba bay and Cabo
Frio, but it is assumed to be comparable to other
studied sectors ( Turcq et al., in press).
2.1. Vegetation
Plant associations in this region vary according
to physiographic conditions and distance from the
ocean. The land–sea interface, especially along the
edges of rivers and lagoons, is covered by floristically poor mangrove forests and saltwater marshes.
Avicennia schaueriana and Laguncularia racemosa
are common species. Rhizophora mangle, less tolerant to high salinity, is rare.
The restinga ecosystem, characteristic of the
Fig. 2. Curve of relative sea level variation on the Rio Doce coastal plain, Espı́rito Santo (after Martin et al., 1997).
114
R. Scheel-Ybert / Review of Palaeobotany and Palynology 110 (2000) 111–138
Brazilian coast, is associated with sandy beach
ridges. This complex ecosystem includes different
vegetation types, from sparse open plant communities, herbaceous and scrub formations (‘open restinga’) to dense evergreen forest (‘restinga forest’).
Each of these vegetation types occupies a welldefined relief configuration, producing zonation of
the ecosystem. Low areas between the beach ridges
and dune slacks support marshy vegetation. Scrub
communities usually grow inland on the external
sand barrier, on dunes or in low areas. The remnant restinga forests, growing on the innermost
sand barrier, are ca. 8 m high with emergent trees
that reach 15–25 m, a dense understory and many
epiphytes (Araujo and Henriques, 1984; Araujo,
1997).
Xeromorphic formations on hillsides near the
ocean from Cabo Frio Island to Búzios Peninsula
are unique along the coast and present many
endemic species (Araujo, 1997). Columnar cacti
give a characteristic appearance to the low thickets
on the steep slopes facing the sea. Xeromorphic
forest, up to ca. 8 m high, has a well-developed
understory (FEEMA, 1988). Farther from the
ocean, low mountains support forests similar in
composition to those of the Atlantic rain forest
(Rizzini, 1979).
2.2. Prehistoric occupation
The most ancient vestiges of human occupation
of the Brazilian coast are the sambaquis, shell
mounds constructed by fisher–gatherer–hunter
populations who have inhabited the coast from at
least 7100 to 500 14C yr BP (Gaspar, 1996).
Archaeological deposits are constituted by mollusk
shells and fish bones, frequently alternating with
sandy layers. Most archaeological layers contain
artifacts, burials and hearths. Habitation structures are occasionally found. The composition of
sedimentary layers varies from one site to another,
but the manufactured artifacts found in these
horizons are similar in all of them (Gaspar, 1992).
The sites usually stand out distinctly on the landscape. They are generally established on the margins of large water bodies, presenting a set of
various ecological habitats in the surroundings
(the ocean, lagoons, rivers, restingas, mangroves,
forests…). These populations were probably sedentary (Gaspar, 1992) and have occupied the same
sites for a very long time. In the studied sites, the
period of occupation varies from about 500 to
over 3000 years ( Table 1).
3. Material and methods
Charcoal fragments were retrieved from vertical
profiles along the entire sambaqui height. Each
sample consists of a sediment layer of 2 m width,
50 cm depth and 10 cm thick. Sediment was drysieved in the field using 4 mm meshes. The charcoal
from the residues was recovered in the laboratory
with a flotation device ( Ybert et al., 1997). One
exception is the Sambaqui da Ponta da Cabeça,
for which charcoal fragments were collected during
the archaeological excavation, and plant remains
were hand-sorted.
Charcoal fragments were examined under a
reflected light brightfield/darkfield microscope on
transversal, tangential longitudinal and tangential
radial sections along fresh hand-made fractures.
Systematic determination was made by comparing
the anatomical structure with that of known wood
samples and with descriptions and photographs
from the literature (Record and Hess, 1943;
Metcalfe and Chalk, 1950; Détienne and Jacquet,
1983; Mainieri and Chimelo, 1989 and papers on
wood anatomy). As the wood anatomy of tropical
plants is poorly known, especially in our study
region, assembling a reference collection of charred
wood was indispensable. We have assembled 1840
identified samples (920 species, 410 genera, 101
families), obtained from wood collections and field
sampling. Charcoal identification was facilitated
by the development of a program for computeraided identification, coupled to a databank of
anatomical features from extant and fossil charcoal
(Scheel-Ybert et al., 1998).
All charcoal fragments measuring over 4 mm
were analysed. Smaller fragments are normally
impossible to identify. Conventionally, results are
presented either by weight or by number of charcoal pieces. In this paper, count is the only unit
115
R. Scheel-Ybert / Review of Palaeobotany and Palynology 110 (2000) 111–138
Table 1
Radiocarbon dates from the studied sites: conventional ages and calendar ages with 2s of confidence interval (95%)
Site
Level
Conventional age
Calibrated age
Forte
Level IVa
270–320 cmb
240–250 cmb
150–160 cmb
Level III a
Level II a
40–50 cmb
Level I a
0.70 mc
130–140 cmd
100–110 cmb
20–30 cmb
90–100 cmc
20–30 cmb
160–170 cme
70–80 cmb
Level IVf
Level III f
Level II f
Level I f
Level IVf
Level III f
Level II f
5520±120 BP
5270±80 BP
4910±55 BP
3815±50 BP
4330±140 BP
3940±140 BP
2320±55 BP
2240±70 BP
5180±80 BP
4340±70 BP
4490±40 BP
1830±45 BP
3760±180 BP
1430±55 BP
3270±70 BP
2080±40 BP
4520±190 BP
4300±190 BP
4160±180 BP
3800±190 BP
2270±190 BP
1810±50 BP
1790±40 BP
6180–5630
6190–5760
5720–5480
4340–3980
4820–4070
4300–3550
2360–2140
1990–1670
5700–5320
5040–4650
5280–4870
1820–1570
4540–3580
1380–1180
3630–3270
2110–1880
5240–4190
4860–3870
4720–3710
4250–3290
2750–1750
1820–1540
1730–1540
Meio
Salinas Peroano
Boca da Barra
Ponta da Cabeça
Beirada
Pontinha
cal
cal
cal
cal
cal
cal
cal
cal
cal
cal
cal
cal
cal
cal
cal
cal
cal
cal
cal
cal
cal
cal
cal
BP
BP
BP
BP
BP
BP
BP
BP
BP
BP
BP
BP
BP
BP
BP
BP
BP
BP
BP
BP
BP
BP
BP
Material
Laboratory number
Shells
Charcoal
Charcoal
Charcoal
Shells
Shells
Charcoal
Shells
Shells
Charcoal
Charcoal
Charcoal
Charcoal
Charcoal
Charcoal
Charcoal
Shells
Shells
Shells
Shells
Charcoal
Charcoal
Charcoal
Bah
GifA-98348
Ly-8467
Ly-8466
Bah
Bah
Gif-11038
Bah
Beta-84809
Gif-8454
Gif-11042
Gif-11041
Beta-83827
Gif-11043
Beta-84332
Gif-11045
Bah-1651
Bah-1647
Bah-1646
Bah-1765
Bah-1764
Gif-8683
Gif-8682
a Kneip (1980).
b Scheel-Ybert (1998).
c Tenório (1996).
d Franco and Gaspar (1992).
e Tenório (unpublished).
f Kneip et al. (1991).
presented, for two reasons: (1) tropical woods
present a very wide range of density values [from
0.1 to 1.3 in dry wood (Briane and Doat, 1985)]
that may greatly distort biomass estimations based
on weight; (2) frequently, shell-mound charcoal
fragments are impregnated with carbonates that
randomly increase the sample weight.
Our interpretations are based on multivariate
analyses applied to both charcoal assemblages and
the results of phytosociological analyses of the
extant vegetation. The sample validity was tested
by the analysis of saturation and Gini–Lorenz
curves (Scheel-Ybert, 1998). In some samples, the
number of available charcoal fragments was too
low to afford consistent palaeoenvironmental interpretations. These levels are represented by light
bars in the charcoal diagrams.
4. Chronology
Many levels of the studied sites had already
been dated by archaeologists, and we have
obtained additional dates from analysed charcoal
samples. After calibration, all the chronological
sequences obtained are coherent ( Table 1). This
argument favours the utilization of archaeological
deposits for the study of palaeoenvironmental
events.
We have calibrated dates obtained from mollusk shells using a ‘global’ world ocean reservoir
age correction (Stuiver and Reimer, 1993).
However, dating of mollusk shells should be
avoided, especially in this region, where the reservoir effect can be significant due to local upwelling. Appropriate sampling methodology, like
116
R. Scheel-Ybert / Review of Palaeobotany and Palynology 110 (2000) 111–138
flotation, usually allows the recovery of enough
charcoal for dating.
5. Anthracological results
5.1. Cabo Frio region
The Sambaqui do Forte’s charcoal diagram
shows a great floristic diversity ( Fig. 3), with a
strong predominance of Myrtaceae species. This
family is present in different communities of
Brazilian vegetation, but high percentages are typical for restinga formations. Most of the 102 taxa
identified correspond to the different plant associations presently existing in the region. The most
typical for the restinga forest are Annonaceae aff.
Duguetia, Actinostemon sp., Alibertia sp.,
Guettarda aff. viburnoides, Rubiaceae type
Randia/Coutarea, Rudgea sp. and Esenbeckia sp.
These taxa can also occur in the xeromorphic
forest. Aspidosperma sp., Melastomataceae,
Condalia sp. and Pouteria sp. are typical for the
restinga forest as well as the open restinga.
Maytenus sp., Rheedia sp., Byrsonima sp., Rapanea
sp. and Sideroxylon aff. obtusifolium are characteristic of the open restinga. Gochnatia sp.,
Pachystroma sp. and Lonchocarpus sp. are typical
for the xeromorphic forest. Caesalpinia echinata
and Cupania sp. occur both in the xeromorphic
forest and in the restinga forest, but are considered
as typical for the xeromorphic forest Securinega
sp. and Sloanea sp. are known only in the Atlantic
Forest, but they can also occur in the xeromorphic
forest. Laguncularia racemosa and Avicennia aff.
schaueriana are characteristic of the mangrove
forest.
In the Sambaqui Salinas Peroano (Fig. 4), the
most important taxa are: Gochnatia sp.,
Pachystroma sp., Securinega sp., cf. Caesalpinia
echinata (xeromorphic forest), Annonaceae aff.
Duguetia, Rubiaceae type Randia/Coutarea,
Rutaceae aff. Metrodorea (restinga forest and xeromorphic forest), Condalia sp. (restinga forest and
open restinga), Sideroxylon aff. obtusifolium (open
restinga), Laguncularia racemosa and Avicennia aff.
schaueriana (mangrove); and in the Sambaqui
Boca da Barra (Fig. 5): Gochnatia sp.,
Pachystroma sp., Lonchocarpus sp., Machaerium
sp., Cupania sp. (xeromorphic forest), Annonaceae
aff. Duguetia, Actinostemon sp., Lauraceae, cf.
Piptadenia, Guettarda aff. viburnoides, Rubiaceae
type Randia/Coutarea, Rudgea sp. (restinga forest
and xeromorphic forest), Condalia sp. (restinga
forest and open restinga), Maytenus sp., Rapanea
sp., Sideroxylon aff. obtusifolium (open restinga),
Laguncularia racemosa and Avicennia aff. schaueriana (mangrove). Species of Myrtaceae are frequent
all along the diagrams, but they are less important
than in the Sambaqui do Forte.
Mangrove taxa are responsible for most of the
significant variations in these charcoal diagrams.
In the Sambaqui do Forte, mangrove elements are
abundant from the base of the sequence until ca.
4900 14C yr BP. After 4900 14C yr BP, they are
rare until ca. 2300 14C yr BP and increase again
thereafter. In the Sambaqui Salinas Peroano, they
are abundant at the base of the diagram (between
100 and 210 cm, ca. 4300/4500 14C yr BP), while
in the upper 100 cm, there is a reduction in the
percentages and an irregular presence. Higher percentages of Avicennia aff. schaueriana between 50
and 90 cm are probably not significant because of
the small number of charcoal fragments analysed.
In the Sambaqui Boca da Barra, mangrove elements are recorded in small percentages between
3700 and 1400 14C yr BP.
Mangrove evolution is different on both sides
of the channel. On its western side (Sambaqui do
Forte), mangrove vegetation reduces abruptly just
after 4900 14C yr BP; it remains rare from 4900 to
2300 14C yr BP, and an important re-establishment
is evidenced thereafter. On the eastern side of the
channel (Sambaquis Salinas Peroano and Boca da
Barra), mangrove is abundant until 4300/4500
14C yr BP and reduced thereafter. These chronological differences may be due to geomorphological
and/or sedimentological features of the channel
related to: different bathymetry; different contours
(which can lead to a stronger erosion at one side
than another); differences in the sedimentary
nature of the subsurface (sandy or clayey); differences in shore nature, which are sandy on the west
side and rocky on the east side (favouring rainwater flowing on its east side). It is likely that a
precipitation reduction would affect the west side
R. Scheel-Ybert / Review of Palaeobotany and Palynology 110 (2000) 111–130
Fig. 3. Charcoal percentage diagram of the Sambaqui do Forte. Light bars represent samples with less than 100 fragments analysed. Nsp: number of taxa; Ni: number of determinable fragments; Nt: total number of fragments analysed; Vs: sediment volume (in litres).
Fig. 4. Charcoal percentage diagram of the Sambaqui Salinas Peroano (see legend to Fig. 3). The sediment volume retrieved in each level is 100 l.
pp. 117–120
Fig. 5. Charcoal percentage diagram of the Sambaqui Boca da Barra (see legend to Fig. 3). The sediment volume retrieved in each level is 100 l.
R. Scheel-Ybert / Review of Palaeobotany and Palynology 110 (2000) 111–138
121
122
R. Scheel-Ybert / Review of Palaeobotany and Palynology 110 (2000) 111–138
of the channel more intensely, due to the more
effective drainage associated with sandy soils.
The changes in mangrove vegetation cannot be
attributed to sea-level variations. The three regressive and two transgressive episodes identified for
the Brazilian coast during this period are not in
phase with the mangrove evolution (see Fig. 2).
Probably, precipitation reduction provoked a
higher salinity in the Araruama Lagoon which,
consequently, resulted in a reduction in mangrove
vegetation. This hypothesis agrees with the existence of climatic oscillations in the southeastern
Rio de Janeiro State, proposed by Tasayco-Ortega
(1996) based on sediment analysis. Our results
have been compared to his curve of changes in the
isotopic composition of carbonates in sediments
of the Araruama Lagoon ( Fig. 6). Low values of
d18O are interpreted to reflect a major influence of
rainwater, while high values signify increased salinity. The curve shows that salinity in the lagoon
was low until ca. 5000 14C yr BP. This can be
correlated to the high percentages of mangrove
elements in the base of the charcoal diagrams.
After this point, salinity is high until ca. 2300
14C yr BP. During this period, mangrove is rare
on both sides of the Itajuru Channel. A net reduction of salinity is recorded between 2300 and 2000
14C yr BP. It can be correlated to the
re-establishment of mangrove elements in the
Sambaqui do Forte after 2300 14C yr BP. This
re-establishment is certainly related to a humid
episode. We cannot estimate its duration, but the
fact that it is not recorded in the east side of the
channel suggests it was brief.
5.2. Arraial do Cabo region
Sorting of plant remains from the Sambaqui da
Ponta da Cabeça is not finished, so we cannot yet
present charcoal and palm-fruit samples for all
levels. However, analysis of the charcoal diagram
is very interesting and shows an important contribution of mangrove taxa in the upper levels.
The charcoal assemblage has a high floristic
diversity, most of the taxa presenting low percentages, with exception of Myrtaceae species and, in
the upper 30 cm of the analysed profile,
Laguncularia racemosa ( Fig. 7). Amongst the more
significant elements are Gochnatia sp., Pachystroma
sp., Lonchocarpus sp. and Cupania sp. (xeromor-
Fig. 6. Comparison of mangrove records and the stable carbonates isotopes curves (d18O and d13C ) of core B-2 from Araruama
Lagoon (after Tasayco-Ortega, 1996). Isotopic values are given in ‰ PDB. Core ages are interpolated from 14C dates obtained from
samples indicated by vertical lines.
Fig. 7. Charcoal percentage diagram of the Sambaqui da Ponta da Cabeça (see legend to Fig. 3).
R. Scheel-Ybert / Review of Palaeobotany and Palynology 110 (2000) 111–138
123
124
R. Scheel-Ybert / Review of Palaeobotany and Palynology 110 (2000) 111–138
phic forest), Annonaceae aff. Duguetia, Actinostemon sp., cf. Piptadenia, Rubiaceae type
Randia/Coutarea, Rudgea sp. (restinga forest and
xeromorphic forest), Aspidosperma sp., Tabebuia
sp., Condalia sp., Pouteria sp. (restinga forest and
open restinga), Maytenus sp., Erythroxylum sp.,
Rheedia sp., Byrsonima sp., Rapanea sp.,
Sideroxylon aff. obtusifolium (open restinga).
Fragments of palm stems are also frequent.
Mangrove elements increase significantly in the
top of the diagram (from ca. 2100 14C yr BP
onward). Archaeologists describe the first 90 cm
of this profile as being almost entirely formed from
hearths, while the upper 30 cm, composed mainly
of shell remains, are considered the apogee of the
occupation ( Tenório et al., 1992). Two hypotheses
are proposed to explain the increase of mangrove
elements: either the increase is due to an actual
increase of mangrove vegetation in the environment, or it is related to a change in the wood
utilization from the beginning to the end of the
occupation. Anthracological analysis of the missing samples and a better understanding of the
archaeological context are necessary before we can
make a conclusive decision for either possibility.
5.3. Saquarema region
Anthracological results of the Sambaquis da
Beirada and da Pontinha are presented in a single
diagram, as the two sites are close to each other,
and their occupations are separated for at least
500 years (Fig. 8). The most significant taxa identified are typical for open restinga (cf. Tapirira,
Maytenus sp., Erythroxylum sp., Rheedia sp.,
Byrsonima sp., Sideroxylon aff. obtusifolium) or
both open restinga and restinga forest ecosystems
(Aspidosperma sp., Pouteria sp., Myrtaceae spp.).
Restinga forest (Annonaceae aff. Duguetia,
Actinostemon sp., Rubiaceae type Randia/
Coutarea, Esenbeckia sp.…) and Atlantic forest
elements (Pachystroma sp.) are also present.
Mangrove taxa are rare. Sesbania sp., a ruderal
species that may be associated with the mangrove,
is rather frequent, especially in the Sambaqui
da Beirada.
5.4. Comparison of palaeoenvironment and present
vegetation
Anthracological results were compared to phytosociological data of comparable plant associations from diverse Brazilian localities using factor
correspondence analyses (FCA).
We also compared these results to the surface
charcoal sample retrieved from Sambaqui do
Meio, which is probably associated with a local
charcoal production not older than a few decades.
The taxonomic assemblage of this modern sample
is very similar to that found in the archaeological
samples ( Fig. 9). The taxa represented are typical
for restinga formations (Zollernia sp., cf.
Piptadenia, Jacaranda sp., Capparis sp., Rapanea
sp., Sideroxylon aff. obtusifolium, cf. Schinus,
Aegiphilla sp.…) and xeromorphic forest
(Pachystroma sp., Lonchocarpus sp., Machaerium
sp.…). The diversity of Myrtaceae and their overall
proportions are lower than in archaeological
samples. Mangrove elements are rare. Hibiscus cf.
tiliaceus and Sesbania sp. show high percentages.
These pioneer taxa, presently very frequent in the
region, are particularly abundant in areas where
mangrove existed previously and in marshy zones
at the margins of the lagoons. Most of the predominant taxa are typical for degraded or secondary vegetation (Schinus, Hibiscus, Sesbania,
Aegiphilla), thus corresponding to what is presently
found locally. Araucaria angustifolia represents an
imported taxon (wooden furniture, crates?).
The first FCA represents only the present-day
vegetation data. All analyses were made at the
generic level, to allow a comparison with charcoal
assemblages. The following phytosociological data
were used:
(1) Eleven studies in restinga forest: two areas
from the Jacarepiá Ecological Reserve, near
Saquarema, one presenting a well-preserved vegetation (Sá and Araujo, unpublished data), and the
other concerning a zone cleared 6 years earlier (Sá,
1993); one area from Espı́rito Santo State (Fabris,
1995); two distinct areas from Cananéia, São
Paulo State (Sugiyama, 1993); four different areas
from Ilha do Mel, Paraná State, and the global
data of this author (Silva, 1990); one area from
R. Scheel-Ybert / Review of Palaeobotany and Palynology 110 (2000) 111–130
Fig. 8. Charcoal percentage diagram of the Saquarema region. Lower part, Sambaqui da Beirada; upper part, Sambaqui da Pontinha (see legend to Fig. 3). The sediment volume retrieved in each level is 100 l.
pp. 125–128
Fig. 9. Charcoal percentage diagram of the Sambaqui do Meio surface sample (see legend to Fig. 3).
R. Scheel-Ybert / Review of Palaeobotany and Palynology 110 (2000) 111–138
129
130
R. Scheel-Ybert / Review of Palaeobotany and Palynology 110 (2000) 111–138
Natal, Rio Grande do Norte State ( Trindade,
1991).
(2) Four studies in Atlantic Forest: two areas
from Rio de Janeiro State (Guedes, 1988; Kurtz,
1994); two areas from São Paulo State (Silva and
Leitão Filho, 1982; Negreiros et al., 1995).
Despite the floristic similarities between restinga
and Atlantic forests (Rizzini, 1979; Araujo and
Lacerda, 1987), FCA produces a clear separation
of these formations, even at the generic level. Three
groups are clearly recognized (Fig. 10): (1)
Atlantic Forest, (2) restinga forest from tropical
areas (Rio de Janeiro, Espı́rito Santo, Rio Grande
do Norte) and (3) restinga forest from subtropical
areas (São Paulo, Paraná). The dataset from the
secondary restinga forest is supplementary, that is
to say, the dataset relative position is represented
in the graphical results, but its numerical contributions are not considered for the FCA calculations.
Otherwise, it is very isolated from the other elements, because this floristic assemblage presents
high percentages of pioneer taxa and is therefore
very different from the others.
Anthracological results were then compared to
the phytosociological data of the present-day vegetation. Extant taxa absent from the fossil assemblages are supplementary, and we have grouped
together at the family level all taxa for which the
wood anatomy does not allow a better identification, e.g. Myrtaceae and Nyctaginaceae. In this
analysis, axis 2 identifies a group associating the
Sambaqui do Meio with the secondary restinga
forest and, to a lesser extent, a group associating
subtropical Atlantic Forest datasets. Axis 1 identifies one group associating all fossil samples with
the tropical Atlantic and restinga forests datasets,
and another associating subtropical restinga forest
datasets ( Fig. 11).
Archaeological charcoal samples resemble each
other more than they resemble the present vegetation types, but their resemblance to modern tropical vegetation types corroborates the robustness
Fig. 10. Factor correspondence analysis. Phytosociological studies only. R2RJ supplementary (see legend to Fig. 11).
R. Scheel-Ybert / Review of Palaeobotany and Palynology 110 (2000) 111–138
131
Fig. 11. Factor correspondence analysis. Correlation between archaeological samples and phytosociological studies. Each symbol
represents a group of superposed samples. $: Sambaqui do Forte; 6: Sambaqui Salinas Peroano; k: Sambaqui Boca da Barra;
: Sambaqui do Meio (modern sample); : Sambaqui da Ponta da Cabeça; #: Sambaqui da Beirada; &: Sambaqui da Pontinha.
M1RJ ( Kurtz, 1994), M2RJ (Guedes, 1988): Atlantic Forest, Rio de Janeiro; M1SP (Silva and Leitão Filho, 1982), M2SP (Negreiros
et al., 1995): Atlantic Forest, São Paulo; R1RJ (Sá and Araujo, unpublished), R2RJ (Sá, 1993): Restinga Forest, Rio de Janeiro;
RES (Fabris, 1995): Restinga Forest, Espı́rito Santo; RRN ( Trindade, 1991): Restinga Forest, Rio Grande do Norte; R1SP, R2SP
(Sugiyama, 1993): Restinga Forest, São Paulo; R1PR, R2PR, R3PR, R4PR, R5PR (Silva, 1990): Restinga Forest, Paraná.
of the anthracological method. The Sambaqui do
Meio modern sample is associated with the secondary restinga forest sample. Both are characterized
by pioneer taxa (especially Aegiphilla sp.), typical
for disturbed plant associations and early stages
of secondary succession.
A FCA with these two samples as supplementary confirms the high similarity between archaeological samples ( Fig. 12), for axis 1 opposes most
of them to the extant vegetation datasets.
However, the sample of the well-preserved Rio de
Janeiro restinga forest is associated with the
Sambaquis da Beirada et da Pontinha, all of them
situated in the Saquarema region. The dataset
from the Rio Grande do Norte restinga forest is
associated with Sambaquis do Forte, Boca da
Barra and Ponta da Cabeça. This forest is
described as a ‘composite habitat’ ( Trindade,
1991), which is exactly the case for archaeological
samples. The predominance of restinga forest elements in the Sambaqui do Forte makes these
samples more similar to the Rio Grande do Norte
restinga, while the levels of Sambaqui Boca da
Barra are less similar. Those of Sambaqui Salinas
Peroano, with a strong predominance of xeromorphic forest elements, form an isolated group.
The mixing of taxa from different plant associations in the archaeological samples explains why
they resemble each other more than they resemble
the present samples. Nevertheless, in spite of all
the processes that affect the charcoal sample
from the moment of wood gathering until its
analysis (physical transformations of charred
wood, post-depositional processes, charcoal frag-
132
R. Scheel-Ybert / Review of Palaeobotany and Palynology 110 (2000) 111–138
Fig. 12. Factor correspondence analysis. Correlation between archaeological samples and phytosociological studies. Sambaqui do
Meio and R2RJ supplementary (see legend to Fig. 11).
mentation, etc.), and in spite of the limited number
of charcoal fragments analysed, our results show
that the anthracological taxonomic assemblage
resembles the present-day plant associations reasonably well.
We compared the palaeoenvironmental results
of each studied site in a FCA considering all taxa
identified. It shows a strong opposition between
archaeological samples and the Sambaqui do
Meio, with a weak discrimination of sites on axis
2. A factor analysis with the Sambaqui do Meio
dataset as supplementary leads to a good separation of the different sites (Fig. 13). Axis 1 opposes
the sites from Saquarema and Arraial do Cabo
regions to those from Cabo Frio region. Axis 2
opposes the upper part of Sambaqui Salinas
Peroano to the lower part of this site and to the
other samples. Five groups can be identified:
1. All levels of Sambaquis da Beirada and da
Pontinha, where open restinga elements are
predominant; mangrove elements are rare.
2. All levels of Sambaqui da Ponta da Cabeça,
with predominance of restinga elements, but
also presenting some forest elements; mangrove
taxa are frequent, especially in the upper levels.
3. The Sambaqui Salinas Peroano, where xeromorphic forest is predominant, is separated in
two groups: the upper levels…
4. and the lower levels, where mangrove elements
are more abundant.
5. On a central position, all levels of Sambaquis
do Forte and Boca da Barra, which present
elements of the various plant associations in
rather equitable proportions. Levels where the
mangrove elements are more important have
values that are more negative on axis 2.
Sambaqui Boca da Barra, where forest elements
are predominant on open restinga elements,
shows values that are slightly more positives on
axis 1 than those of the Sambaqui do Forte.
In conclusion, axis 1 opposes open restinga
elements to xeromorphic forest elements, while
R. Scheel-Ybert / Review of Palaeobotany and Palynology 110 (2000) 111–138
Fig. 13. Factor correspondence analysis (archaeological samples
only). Sambaqui do Meio supplementary (see legend to
Fig. 11).
axis 2 separates samples with a strong contribution
of mangrove taxa from samples with a weak
contribution.
6. Discussion
6.1. Palaeoenvironmental interpretation
Our results show that various plant associations
existed in the Cabo Frio region during the Late
Holocene (from ca. 5500 to 1400 14C yr BP, or
6190 to 1180 cal yr BP). Restinga forest, open
restinga, xeromorphic forest and mangrove taxa
are present in the three studied sites in variable
proportions. Open restinga is better represented in
the Sambaqui do Forte, while closed forest formations are more important in Sambaquis Salinas
Peroano and Boca da Barra. This is a consequence
of the local site characteristics: the former sam-
133
baqui is situated near the beach, in the phytosociological domain of open restinga, while the latter
are situated on the margin of the channel, on
crystalline hills dominated by forest formations.
The only significant variations in the charcoal
diagrams concern mangrove vegetation ( Fig. 14).
We attribute these to climatic oscillations provoking changes in lagoon salinity. This interpretation
is corroborated by changes in isotopic composition
of sediment carbonates in the Araruama Lagoon
( Tasayco-Ortega, 1996), which confirms the existence of at least two dry episodes in the Cabo Frio
region, intercalated with brief humid episodes. In
spite of these climatic oscillations affecting mangrove vegetation, no other significant evidence of
vegetation change is evidenced.
It is expected that climatic changes produce
vegetation changes. We show that this is not the
case with regard to the coastal mainland vegetation. The restinga ecosystem is related to sandy
soils and to the geomorphologic nature of sandy
beach ridges. This implies that it is essentially an
edaphic formation, with a marked resilience to
climatic change. This is also the case of the xeromorphic forest characteristic of rocky outcrops in
the Cabo Frio region. In spite of the fact that this
forest is also related to climate parameters, edaphic
factors may certainly explain why it has not been
greatly affected by climatic oscillations.
Bissa (1998) presented a palynological study of
a peat core from a flooded forest area in the
coastal zone of São Paulo State between ca. 4300
and 1000 14C yr BP. He found no variations in the
vegetation that could be attributed to climatic
changes. This seems to confirm that coastal formations, strongly depending on edaphic conditioning,
are much more resistant to climatic variations and
less sensitive to climatic change.
The charcoal assemblage in the Arraial do Cabo
region shows that during the period of human
occupation of the Sambaqui da Ponta da Cabeça
(from ca. 3300 until after 2000 14C yr BP, or 3630–
1880 cal yr BP), open restinga formations were
locally predominant. This is also a consequence of
the site location. In spite of the fact that this site
is situated upon a crystalline hill, it is located very
near the beach and characterized by an especially
dry climate due to upwelling centred in its vicinity.
134
R. Scheel-Ybert / Review of Palaeobotany and Palynology 110 (2000) 111–138
Fig. 14. Summary charcoal diagrams of Sambaquis do Forte, Salinas Peroano, Boca da Barra, Ponta da Cabeça, da Beirada and
da Pontinha.
Mangrove vegetation, which is currently absent
from this region, was probably established on the
margins of the Araruama Lagoon during this
period.
At the Saquarema region, sandy beach ridges
were already occupied by the restinga vegetation
between 4300 and 3800 14C yr BP (4860–3290
cal yr BP) and from ca. 2300 until after 1800 14C
yr BP (at least 2750–1540 cal yr BP). Open restinga formations were predominant. The slight
increase of Atlantic Forest elements at the
Sambaqui da Pontinha’s level 100–120 cm, i.e.
around 2300 14C yr BP (2750–1750 cal yr BP), is
too small to be interpreted as significant for an
environmental change. The oscillations in the relative percentages of the vegetation types in the
Sambaqui da Beirada, as well as in some levels of
the Sambaquis da Pontinha and Salinas Peroano,
are not significant because of the excessively small
number of charcoal fragments analysed.
6.2. Cultural selection
Aleatory gathering of dead wood certainly constituted the main source of firewood for fisher–
gatherer–hunters populations. This is suggested by
the great diversity of the charcoal assemblage and
the numerous fragments that show traces of decay
or signs of insect larvae attack before charring.
However, the high frequencies of Condalia sp.
in most of the samples ( Figs. 3–8) are difficult to
explain by environmental criteria. Condalia buxi-
R. Scheel-Ybert / Review of Palaeobotany and Palynology 110 (2000) 111–138
folia, the only species that occurs in the Brazilian
coastal vegetation nowadays, is a very rare taxon
(Silva and Oliveira, 1989). Two hypotheses may
explain its high frequencies: (1) the species was
abundant in the environment, implying that the
vegetation was somewhat different from the present vegetation; (2) the species was not particularly
frequent but was selected for cultural reasons.
Both hypotheses imply that this taxon was at least
more frequent than it is nowadays, but the first
must be discarded because the floristic assemblage
of the archaeological charcoal is otherwise very
similar to the present vegetation. Frequencies of
Condalia fragments in our samples could be
explained by a cultural selection due to wood
features. It is considered as an excellent combustible and is reported to produce a blue pigment; the
root bark of some species is utilized as soap or
medicine; the plant bears edible fruits (Record and
Hess, 1943). The hypothesis of a ritual or mystical
utilization of this taxon is also possible. This would
explain why all Condalia fragments are vitrified,
suggesting that its wood was always charred green
(Scheel-Ybert, 1998).
Species selection is apparently contradictory to
a good representation of the surrounding vegetation in the charcoal fragments. However, we think
that most of the species present in the environment
can be represented in archaeological charcoal,
especially if fuel wood is obtained from dead wood
gathering. We assume that this argument justifies
that anthracological analyses are representative to
reconstruct palaeoenvironments, in spite of the
cultural selection that may cause over-representation of a few taxa. The high taxonomic diversity
of archaeological charcoal samples and the good
correspondence between the anthracological
spectra and the present vegetation corroborate this
argument.
7. Conclusions
Our results show that the analysis of tropical
archaeological charcoal provides reliable information on past environments and confirm the validity
of anthracology for tropical palaeoenvironmental
135
reconstruction. The charcoal samples’ great taxonomic diversity, the similarity of the anthracological results in neighbouring sites and the good
correspondence between the charcoal assemblages
and the present-day vegetation indicate that wood
gathering by prehistorical populations, most likely
for domestic fires, was not selective.
However, charcoal assemblages represent
mainly the local vegetation. The analysis of isolated sites provides only site information, and a
regional reconstruction depends on the study of
numerous sites distributed over a relatively large
area.
The six archaeological sites studied have been
occupied for a long time, from ca. 5500 to 1400
14C yr BP (6190–1180 cal yr BP). Analysis of over
15 500 charcoal fragments demonstrates that
during the Late Holocene, the southeastern coast
of Brazil presented different restinga vegetation
types, forest formations like the xeromorphic forest
from the rocky shore of Cabo Frio region and,
inland, the Atlantic Forest, and mangrove vegetation, especially on lagoons margins of Cabo Frio
and Arraial do Cabo regions. Restinga forest was
very frequent in this region during this period, and
open restinga formations were probably restricted
to the external beach ridges, to lower zones
between beach ridges and to lagoons edges.
The mainland coastal vegetation was not perturbed during this period, but changes in mangrove
elements were recorded. These changes are associated with two humid episodes in the Cabo Frio
region (recorded from ca. 5500 to 4900/4500 and
from ca. 2300 to 2000 14C yr BP) intercalated by
two episodes of increased dryness with increased
lagoon salinity (from ca. 4900/4500 to 2300 and
from ca. 2000 to 1400 14C yr BP). We propose
that the stability of the coastal mainland vegetation
is a consequence of its edaphic character, which
makes coastal formations much more resistant to
climatic variations and less sensitive to climatic
change.
Environmental stability has important consequences on prehistorical populations. We propose
that this has been a main factor in the maintenance
of the sociocultural system of the fisher–gatherer–
hunters that have lived on the Brazilian coast for
more than 6000 years (Gaspar, 1996).
136
R. Scheel-Ybert / Review of Palaeobotany and Palynology 110 (2000) 111–138
Acknowledgements
This work was accomplished while holding a
doctoral scholarship from the CAPES, Brazil.
Thanks are due to Maria Dulce Gaspar, Lina
Kneip and Maria Cristina Tenório, archaeologists
responsible for the studied sites; to Cyl Farney
Catharino de Sá and Dorothy Sue Dunn de Araujo
for kindly permitting the utilization of unpublished
phytosociological data; to Jean-Louis Vernet,
Maurice Roux, Michel Servant and Jean-Pierre
Ybert for discussions and helpful comments during
the accomplishment of this work; to Thomas Van
der Hammen for reading the manuscript and providing valuable suggestions. We are also indebted
to Henry Hooghiemstra and one anonymous referee for valuable suggestions that greatly improved
the manuscript.
References
Absy, M.L., Cleef, A., Fournier, M., Martin, L., Servant, M.,
Siffeddine, A., Silva, M.F., Soubiès, F., Suguio, K., Turcq,
B., Van der Hammen, Th., 1991. Mise en évidence de quatre
phases d’ouverture de la forêt dense dans le sud-est de
l’Amazonie au cours des 60.000 dernières années. Première
comparaison avec d’autres régions tropicales. C. R. Acad.
Sci. Paris Ser. II 312, 673–678.
Araujo, D.S.D., 1997. Cabo Frio Region. In: Davis, S.D.
Heywood, V.H., Herrera-MacBride, O., Villa-Lobos, J.,
Hamilton, A.C. ( Eds.), Centres of Plant Diversity, Vol. 3,
373–375.
Araujo, D.S.D., Henriques, R.P.B., 1984. Análise florı́stica das
restingas do Estado do Rio de Janeiro. In: Lacerda, L.D.,
Araujo, D.S.D., Cerqueira, R., Turcq, B. (Eds.), Restingas
Origem Estrutura Processos. CEUFF, Niterói, Brazil,
pp. 159–193.
Araujo, D.S.D., Lacerda, L.D., 1987. A natureza das restingas.
Ciência Hoje 6 (33), 26–32.
Badal, E., Bernabeu, J., Vernet, J.-L., 1994. Vegetation changes
and human action from the Neolithic to the Bronze Age
(7000–4000 BP) in Alicante, Spain, based on charcoal analysis. Veget. Hist. Archaeobot. 3, 155–166.
Barbiére, E.B., 1984. Cabo Frio e Iguaba Grande, dois microclimas distintos a um curto intervalo espacial. In: Lacerda,
L.D., Araujo, D.S.D., Cerqueira, R., Turcq, B. ( Eds.), Restingas, Origem, Estrutura, Processos. CEUFF, Niterói,
Brazil, pp. 3–14.
Barroso, L.V., 1987. Diagnóstico ambiental da Lagoa de Araruama, RJ. Bol. FBCN 22, 30–65.
Behling, H., 1997. Late Quaternary vegetation, climate and fire
history from the tropical mountain region of Morro de Itapeva, SE Brazil. Palaeogeogr. Palaeoclimatol. Palaeoecol.
129, 407–422.
Bissa, W.M., 1998. Reconstituição dos paleoambientes de uma
planı́cie holocênica no curso inferior do Rio Ribeira Municı́pio de Iguape SP. Tese de Mestrado, Dept. Geografia
FFLCH USP, S. Paulo Brazil, 60 pp.
Briane, D., Doat, J., 1985. Guide technique de la carbonisation.
La fabrication du charbon de bois. EDISUD/Centre Technique Forestier Tropical, Aix-en-Provence, France. 178 pp.
Détienne, P., Jacquet, P., 1983. Atlas d’identification des bois
de l’Amazonie et des régions voisines. Centre Technique
Forestier Tropical, France. 640 pp.
Fabris, L.C., 1995. Composição florı́stica e fitossociologia de
uma faixa de floresta arenosa litorânea do Parque Estadual
de Setiba, municı́pio de Guarapari ES. Tese de Mestrado,
Inst. Biociências UNESP, Rio Claro Brazil, 195 pp.
FEEMA, 1988. Importância da biota da região de Cabo Frio.
Technical report. FEEMA, Rio de Janeiro. 50 pp.
Ferraz-Vicentini, K.R., Salgado-Labouriau, M.L., 1996. Palynological analysis of a palm swamp in Central Brazil. J. S.
Am. Earth Sci. 9, 207–219.
Figueiral, I., 1995. Evidence from charcoal analysis environmental change during the interval late Bronze Age to
Roman, at the archaeological site of Castro de Penices,
N.W. Portugal. Veget. Hist. Archaeobot. 4, 93–100.
Franco, T.C.B., Gaspar, M.D., 1992. O sı́tio arqueológico Salinas Peroano. In: Anais da VI Reunião Cientı́fica da Sociedade de Arqueologia Brasileira Vol. 1. SAB, Rio de Janeiro,
pp. 162–171.
Gaspar, M.D., 1992. Aspectos da organização social de um
grupo de pescadores, coletores e caçadores que ocupou o
litoral do Estado do Rio de Janeiro. In: Araújo, A.J.G.,
Ferreira, L.F. ( Eds.), Paleontologia e Paleoepidemiologia:
Estudos Multidisciplinares. Ensp Fundação Oswaldo Cruz,
Rio de Janeiro, pp. 95–109.
Gaspar, M.D., 1996. Análises das datações radiocarbônicas dos
sı́tios de pescadores, coletores e caçadores. Bol. Mus. Paran.
Emilio Goeldi, sér. Ciênc. Terra 8, 81–91.
Guedes, R., 1988. Composição florı́stica e estrutura de um
trecho de mata perturbada de baixada no municı́pio de
Magé, Rio de Janeiro. Arq. Jard. Bot. Rio de Janeiro 29,
155–200.
Heinz, C., Thiébault, S., 1998. Characterization and palaeoecological significance of archaeological charcoal assemblages during late and post-glacial phases in southern
France. Quat. Res. 50, 56–68.
Hopkins, M.S., Ash, J., Graham, A.W., Head, J., Hewett, R.K.,
1993. Charcoal evidence of the spatial extent of the Eucalyptus woodland expansions and rainforest contractions in
North Queensland during the late Pleistocene. J. Biogeogr.
20, 357–372.
Kneip, L.M., 1980. A seqüěncia cultural do sambaqui do
Forte — Cabo Frio, Rio de Janeiro. Pesquisas, sér. Antropol. 31, 87–100.
R. Scheel-Ybert / Review of Palaeobotany and Palynology 110 (2000) 111–138
Kneip, L.M., Pallestrini, L., Crancio, F., Machado, L.M.C.,
1991. As estruturas e suas interrelações em sı́tios de pescadores-coletores pré-históricos do litoral de Saquarema,
RJ. Bol. Inst. Arqueol. Brasil, sér. Ensaios 5, 1–42.
Kurtz, B., 1994. Composição florı́stica e estrutura do componente arbóreo de um trecho de Mata Atlântica na Estação
Ecológica Estadual do Paraı́so Cachoeiras de Macacu
RJ. Tese de Mestrado, Museu Nacional UFRJ, Rio de
Janeiro, Brazil, 106 pp.
Ledru, M.P., Braga, P.I.S., Soubiès, F., Fournier, M., Martin,
L., Suguio, K., Turcq, B., 1995. The last 50,000 years in the
Neotropics (Southern Brazil ): evolution of vegetation and
climate. Palaeogeogr. Palaeoclimatol. Palaeoecol. 123,
239–257.
Mainieri, C., Chimelo, J.P., 1989. Fichas de caracteristicas das
madeiras brasileiras. second ed., IPT, Sao Paulo, 418 pp.
Martin, L., Suguio, K., Flexor, J.M., Bittencourt, A., VillasBoas, G., 1979/80. Le quaternaire marin bresilien ( littoral
pauliste, sud fluminense et bahianais). Cah. ORSTOM, sér.
Géol. 11, 95–124.
Martin, L., Flexor, J.M., Valentin, J.-L., 1988. Influence du
phénomène océanique pacifique ‘El Niño’ sur l’‘upwelling’
et le climat de la region du Cabo Frio sur la côte bresilienne
de l’Etat de Rio de Janeiro. C. R. Acad. Sci. Paris Ser. II
307, 1101–1105.
Martin, L., Suguio, K., Dominguez, J.M.L., Flexor, J.-M.,
1997. Geologia do Quaternário costeiro do litoral norte do
Rio de Janeiro e do Espirito Santo. CPRM/FAPESP, Belo
Horizonte, 104 pp.
Metcalfe, C.R., Chalk, C., 1950. Anatomy of the Dicotyledons.
Clarendon Press, Oxford. 2 vols., 1500 pp.
Moreira, A.L.C., Carmouze, J.-P., 1991. La lagune de Saquarema (Brésil ): hydroclimat, seston et éléments biogéniques
au cours d’un cycle annuel. Rev. Hydrobiol. Trop. 24,
13–23.
Negreiros, O.C., Custodio Filho, A., Dias, A.C., Franco,
G.A.D.C., Couto, H.T.Z., Vieira, M.G.L., Moura
Netto, B.V., 1995. Análise estrutural de um trecho de
floresta pluvial tropical, Parque Estadual de Carlos Botelho,
Núcleo Sete Barras (SP — Brasil ). Rev. Inst. Flor. S. Paulo
7, 1–33.
Record, S.J., Hess, R.W., 1943. Timbers of the New World.
Yale University Press, New Haven, 640 pp.
Rizzini, C.T., 1979. Tratado de fitogeografia do Brasil Vol. 2.
Hucitec, S. Paulo, 373 pp.
Sá, C.F.C., 1992. A vegetação da restinga de Ipitangas, Reserva
Ecológica Estadual de Jacarepiá, Saquarema (RJ ): fisionomia e listagem de Angiospermas. Arq. Jard. Bot. Rio de
Janeiro 31, 87–102.
Sá, C.F.C., 1993. Regeneração de um trecho de Floresta de
Restinga na Reserva Ecológica Estadual de Jacarepiá, municı́pio de Saquarema, RJ. Tese de Mestrado, Museu Nacional UFRJ, Rio de Janeiro, Brazil, 168 pp.
Scheel-Ybert, R., 1998. Stabilité de l’écosystème sur le littoral
Sud-Est du Brésil à l’Holocène Supérieur (5500–1400 ans
BP). Les pêcheurs-cueilleurs-chasseurs et le milieu végétal:
137
apports de l’anthracologie. Thèse de Doctorat, Univ. Montpellier II, Montpellier, France, 3 vols., 520 pp.
Scheel-Ybert, R., Scheel, M., Ybert, J.-P., 1998. Atlas Brasil —
Databank for Charcoal Analysis and Computerized Key to
Charcoal Determination. Version 1.8. CD-ROM, in Portuguese, English and French, 1500 entries.
Servant, M., Fournier, M., Soubiès, F., Suguio, K., Turcq, B.,
1989. Sécheresse holocène au Brésil (18°–20°S). Implications paléoclimatiques. C. R. Acad. Sci. Paris Ser. II 318,
1645–1652.
Silva, A.F., Leitão Filho, H.F., 1982. Composição florı́stica e
estrutura de um trecho da mata atlântica de encosta no
municı́pio de Ubatuba (São Paulo, Brasil ). Rev. Brasil. Bot.
5, 43–52.
Silva, J.G., Oliveira, A.S., 1989. A vegetação de restinga no
municı́pio de Maricá, RJ. Acta Bot. Brasil (Suppl. 3),
253–272.
Silva, S.M., 1990. Composição florı́stica e fitossociologia de um
trecho de floresta de restinga na Ilha do Mel, municı́pio de
Paranaguá PR. Tese de Mestrado, Inst. Biologia UNICAMP, Campinas, Brazil, 146 pp.
Soubiès, F., 1980. Existence d’une phase sèche en Amazonie
brésilienne datée par la présence de charbons dans les sols
(6000–3000 ans BP). Cah. ORSTOM, sér. Géol. 11,
133–148.
Stuiver, M., Reimer, P.J., 1993. Extended 14C data base and
revised CALIB 3.0 14C age calibration program. Radiocarbon 35, 215–230.
Sugiyama, M., 1993. Estudo de florestas na restinga da Ilha do
Cardoso, Cananéia SP. Tese de Mestrado, Inst. Biociências
USP, S. Paulo Brazil, 115 pp.
Tasayco-Ortega, L.A., 1996. Variations paléohydrologiques et
paléoclimatiques d’une région d’upwelling au cours de l’Holocène: enregistrement dans les lagunes côtières de Cabo
Frio (État de Rio de Janeiro, Brésil ). Thèse de Doctorat,
Univ. Pierre et Marie Curie, Paris, France, 321 pp.
Tenório, M.C., 1996. A contribuição da arqueologia na compreensão do desenvolvimento do mangue. Bol. Mus. Paran.
Emilio Goeldi, sér. Ciênc. Terra 8, 123–136.
Tenório, M.C., Barbosa, M., Portela, T., 1992. Pesquisas
arqueológicas no sı́tio Ponta da Cabeça, Arraial do Cabo,
Rio de Janeiro. In: Anais da IV Reunião da Sociedade de
Arqueologia Brasileira Vol. 2. SAB, Rio de Janeiro,
pp. 279–291.
Trindade, A., 1991. Estudo florı́stico e fitossociológico do
estrato arbustivo-arbóreo de um trecho de floresta arenı́cola
costeira do Parque Estadual das Dunas — Natal (RN ). Tese
de Mestrado, Dept. Botânica, UFRPE, Recife, Brazil,
168 pp.
Turcq, B., Martin, L., Flexor, J.M., Suguio, K., Pierre, C.,
Tasayco-Ortega, L., press. Origin of the quaternary coastal
plain between Guaratiba and Cabo Frio, State of Rio de
Janeiro, Brazil. In: Bidone, E.D., Knoppers, B., Lacerda,
L.D. ( Eds.), Environmental Geochemistry of Coastal
Lagoons. Case Study, Lagoons of Rio de Janeiro Coast,
Brazil.
Vernet, J.-L. ( Ed.), Les charbons de bois, les anciens écosys-
138
R. Scheel-Ybert / Review of Palaeobotany and Palynology 110 (2000) 111–138
tèmes et le rôle de l’homme, Actes du Colloque. Bull. Soc.
Bot. Fr., Actual. bot. 139 1992. 725 pp.
Vernet, J.L., Thiébault, S., 1987. An approach to northwestern
Mediterranean recent prehistoric vegetation and ecologic
implications. J. Biogeogr. 14, 117–127.
Ybert, J.P., Scheel, R., Gaspar, M.D., 1997. Descrição de
alguns instrumentos simples utilizados para a coleta e concentração de elementos fósseis de pequenas dimensões de
origem arqueológica ou pedológica. Rev. Mus. Arqueol.
Etnol., S. Paulo 7, 181–189.
Ybert, J.P., Turcq, B., Albuquerque, A.L.S., Cocquit, C., press.
Evolution paleoecologique et paleoclimatique holocene dans
la region moyenne du Rio Doce (Minas Gerais, Bresil )
déduite de l’analyse palynologique de deux carottes du lac
Dom Helvecio. Actes du Symposium Dynamique à long
terme des écosystèmes forestiers tropicaux. UNESCO, Paris.