ArticlePDF Available

Revision and Biogeography of Centrolobium (Leguminosae - Papilionoideae)

Authors:

Abstract and Figures

A taxonomic revision and biogeographic study of the genus Centrolobium (Leguminosae - Papilionoideae) is presented. Centrolobium includes important timber trees distributed disjunctly in seasonally dry tropical forests and rain forests in Central and South America, from Panama to south-eastern Brazil. It is characterized by large samaroid pods with a spiny seed case and an abundance of orange peltate glands covering the leaves and inflorescences. Taxonomic distinctions between some species of Centrolobium have been a source of confusion. Here, seven species are recognized: C. robustum, C. microchaete, C. tomentosum, C. ochroxylum, C. sclerophyllum, C. paraense, and C. yavizanum. Previously recognized varieties of C. paraense, C. paraense var. paraense and C. paraense var. orinocense, are not maintained. Phylogenetic analysis of DNA sequence data from the internal transcribed spacer region of nuclear ribosomal DNA and the plastid matK gene and trnL-trnF intron and spacer support the monophyly of the genus. Different molecular dating methods indicate that the Centrolobium crown group and lineages found to the west and east of the Andes diverged before the Pleistocene. Divergences between species occurring east of the Andes, particularly in Bolivia and south-eastern Brazil are more recent, but nevertheless unlikely to be explained by Pleistocene climatic changes.
Content may be subject to copyright.
Systematic Botany (2009), 34(2): pp. 345–359
© Copyright 2009 by the American Society of Plant Taxonomists
345
The Genus C entrolobium Mart. ex Benth— Centrolobium
(Leguminosae, Papilionoideae) is a neotropical genus of
seven species. It can be easily recognized, even when ster-
ile, by an abundance of orange peltate glands covering most
parts. The genus comprises mostly large trees to 30 m tall,
with a straight bole and broad, rounded crown of terminal
branches. The fruits are large winged samaras to 30 cm long,
with a conspicuous spiny basal seed chamber. A number of
species of Centrolobium are valued in the Neotropics for their
timber, which is durable and richly colored yellowish-orange
with patches of dark red, purple, or black.
Systematic Position— In his tribal classification of subfam-
ily Papilionoideae, Polhill (1981 ; 1994 ) placed Centrolobium in
tribe Dalbergieae, which was diagnosed by free keel petals,
staminal filaments partly fused and without basal fenestrae,
pods with specialized seed chambers, and seeds that accumu-
late alkaloids or other nonprotein amino acids. Recent mor-
phological and molecular phylogenetic studies suggest that
tribe Dalbergieae sensu Polhill is not monophyletic ( Lavin et al.
2001 , 2005 ). In a phylogenetic analysis of plastid ( trnK - matK
and trnL introns) and nuclear ribosomal (5.8S and ITS-1 and 2)
DNA sequences, Lavin et al. (2001) recovered a “dalbergioid”
clade comprising most of tribe Dalbergieae, but with mem-
bers of tribes Aeschynomeneae and Adesmieae and subtribe
Bryinae of tribe Desmodieae nested within it. This redefined
“dalbergioid” clade is characterized by short shoots, glan-
dular based trichomes, and aeschynomenoid root nodules.
Within the dalbergioid clade, Lavin et al. (2001) recovered
three distinct subclades: the Adesmia clade, the Dalbergia
clade, and the Pterocarpus clade. Centrolobium was placed in
the Pterocarpus clade, together with Pterocarpus Jacq., Tipuana
Benth., Platypodium Vog., Platymiscium Vog., Ramorinoa Speg.,
Inocarpus Forst, Paramachaerium Ducke, and Etaballia Benth.
Taxonomic History— The generic name Centrolobium was
proposed by Martius and published by Bentham (1839)
based on Nissolia robusta Vell. The taxonomy of Centrolobium
was reviewed by Rudd (1954) who recognized six species:
C. minus C.Presl, C. ochroxylum Rose ex Rudd, C. paraense Tul.,
C. robustum Mart. ex Benth., C. tomentosum Guill. ex Benth,
and C. yavizanum Pittier. Two varieties were recognized within
C. paraense Tul: var. orinocense Benth. and var. paraense .
Although Rudd’s species delimitations were largely sound,
new collections demonstrate that the characters that Rudd
used to diagnose some taxa vary continuously, such as leaf
base shape in the varieties of C. paraense , and as a result,
Rudd’s species concepts should be re-evaluated. The extra-
Amazonian Brazilian range of the genus was revised by Lima
(1985) , who described one new species, C. sclerophyllum Lima,
and redefined C. microchaete (Mart. ex Benth.) Lima, C. robus-
tum , and C. tomentosum . Here, we present a taxonomic revi-
sion of the whole genus over its entire geographical range.
Distribution of Centrolobium: Relevance to Neotropical
Biogeographic History— Centrolobium species grow prin-
cipally in areas of seasonally dry tropical forest (SDTF; as
described in Prado and Gibbs 1993 ; Pennington et al. 2000 ,
2006 ), but also in more humid forests. The genus is found in
parts of Brazil, Bolivia, Ecuador, Peru, Colombia, Venezuela,
Panama, and the Guianas ( Fig. 1 ). Centrolobium shows a mix-
ture of species distribution patterns, with both widespread
species and more geographically restricted endemics.
The present day distribution of SDTF is highly discontin-
uous in the Neotropics. It roughly corresponds to the New
World distribution of the “succulent biome” sensu Schrire
et al. (2005) . Prado and Gibbs (1993) and Pennington et al.
(2000) highlighted a series of unrelated species with widely
discontinuous distributions in the separated nuclei of SDTF
and suggested that during the Pleistocene, when climates
were drier and cooler; there was a widespread expanse of
seasonally dry vegetation in tropical South America. These
authors favored vicariance over dispersal as a more parsimo-
nious explanation for repeated patterns of disjunction within
the SDTF. They also suggested that climatic fluctuations in the
Pleistocene might have driven speciation in genera with high
endemism in the SDTF.
Revision and Biogeography of Centrolobium (Leguminosae - Papilionoideae)
Michael D. Pirie, 1, 5, 6 Bente B. Klitgaard, 2, 4 and R. Toby Pennington 3
1 Institute of Systematic Botany, University of Zurich, Zollikerstrasse 107, Zurich 8008 Switzerland
2 Department of Botany, Natural History Museum, Cromwell Road, London SW7 5BD U.K.
3 Royal Botanic Garden Edinburgh, 20a Inverleith Row, Edinburgh EH3 5LR U.K.
4 Royal Botanic Gardens, Kew, Richmond, Surrey TW9 3AE U.K.
5 Department of Biochemistry, University of Stellenbosch, Stellenbosch, Private Bag X1,
Matieland 7602 South Africa
6 Author for correspondence (mpirie@sun.ac.za)
Communicating Editor: Matt Lavin
Abstract— A taxonomic revision and biogeographic study of the genus Centrolobium (Leguminosae - Papilionoideae) is presented.
Centrolobium includes important timber trees distributed disjunctly in seasonally dry tropical forests and rain forests in Central and South
America, from Panama to south-eastern Brazil. It is characterized by large samaroid pods with a spiny seed case and an abundance of orange
peltate glands covering the leaves and inflorescences. Taxonomic distinctions between some species of Centrolobium have been a source of con-
fusion. Here, seven species are recognized: C. robustum , C. microchaete , C. tomentosum , C. ochroxylum , C. sclerophyllum , C. paraense, and C. yavi-
zanum . Previously recognized varieties of C. paraense , C. paraense var. paraense and C. paraense var. orinocense, are not maintained. Phylogenetic
analysis of DNA sequence data from the internal transcribed spacer region of nuclear ribosomal DNA and the plastid matK gene and trnL-trnF
intron and spacer support the monophyly of the genus. Different molecular dating methods indicate that the Centrolobium crown group and
lineages found to the west and east of the Andes diverged before the Pleistocene. Divergences between species occurring east of the Andes,
particularly in Bolivia and south-eastern Brazil are more recent, but nevertheless unlikely to be explained by Pleistocene climatic changes.
Keywords— Conservation assessments , Dalbergieae , molecular dating , Neotropics , phylogeny reconstruction , seasonally dry tropical forest
346 SYSTEMATIC BOTANY [Volume 34
Pennington et al. (2004) went on to test the second hypoth-
esis in four different clades with species confined to SDTF
using molecular dating approaches. They concluded that in
South America, species and clade ages largely predate the
Pleistocene and are therefore inconsistent with a Pleistocene
diversification scenario. More recent work has called into
question the existence of widespread seasonally dry for-
est in South America during the Pleistocene ( Lavin 2006 ;
Pennington et al. 2006 ). High levels of geographic structure
in phylogenies of SDTF clades indicate limited historical dis-
persal between discontinuous areas of dry forest, suggesting
that these areas have not been linked in the recent past ( Lavin
2006 ; Pennington et al. 2006 ). High geographic structure in
phylogenies of STDF taxa such as robinioid legumes contrast
with low geographic structure in phylogenies of some rain for-
est taxa such as Inga reflecting the greater current geographic
continuity of neotropical rain forest ( Lavin 2006 ; Pennington
et al. 2006 ). More dated phylogenies are needed to test poten-
tial generalities regarding the timing and geographical pat-
terns of diversification in neotropical rain forests and SDTF
( Pennington et al. 2006 ).
The aims of this paper are a) to produce a dated phylogeny
for Centrolobium to determine the timing of diversification of
the genus and to assess the extent of geographic structure in
its phylogeny; b) to test species delimitations in Centrolobium ,
with particular focus on species with disjunct distributions
across discontinuous areas of SDTF; and c) to present a tax-
onomic revision of Centrolobium , including an identification
key to the species of the genus.
Materials and Methods
Molecular Phylogenetic Reconstruction: Taxon and Molecular Marker
Sampling— This study utilized previously unpublished sequence data from
14 accessions, representing all seven species of Centrolobium . Sequences
from 16 individuals representing 10 (outgroup) taxa of the Pterocarpus
clade and Geoffroea decorticans (Gillies ex Hook. & Arn.) Burkart, the most
distant outgroup, were obtained from published ( Lavin et al. 2001 ) and
unpublished studies ( Platymiscium ITS sequences, Saslis-Lagoudakis
et al. unpubl. results). Where not all markers were available for particu-
lar outgroup accessions, those from different accessions representing the
same taxa were combined ( Appendix 1 ). Molecular markers were chosen
based on outgroup sequence availability and levels of variation within
Centrolobium and include the internal transcribed spacer regions of ribo-
somal DNA (ITS), chloroplast encoded matK (including spacer regions on
both sides of the matK exon), the trnL intron and the trnL-F spacer region.
DNA Extraction, PCR, and DNA Sequencing— Total genomic DNA was
extracted using a CTAB method ( Doyle and Doyle 1987 ) without isopro-
panol precipitation, thus avoiding coprecipitation of oxidized material
( Savolainen et al. 1995 ); silica dried or herbarium leaf material was homog-
enized and incubated in CTAB for 20 minutes with 2-mercaptoethanol at
65°C, followed by 90 minutes ambient mixing with a mixture of chloroform
and isoamyl alcohol. After centrifugation, the supernatant was purified
using Wizard DNA purification system (Promega, Madison, Wisconsin).
The ITS region was amplified and sequenced using the primers of
Beyra and Lavin (1999) . The matK and flanking trnK intron region was
amplified in two pieces using the primers trnK1L and 4R; and 1F and
trnK2R, and sequenced with the same primers plus 4R, 4F, 7L, and 32R
following Lavin et al. (2000) . The universal primers of Taberlet et al. (1991)
were used to amplify and sequence the trnL intron (primers C/D) and the
trnL-trnF spacer (primers E/F). PCR amplifications were performed in 25
μl reactions containing 1 μl 0.4% BSA and 1 μl DMSO under the following
cycling conditions: an initial 4 min. at 94°C; 35 cycles of 30 sec. at 94°C,
1 min. at 55°C, and 2 min. at 72°C; followed by 7 min. at 72°C. PCR
products were purified using QIAquick PCR purification kits (Qiagen,
Venlo, The Netherlands). Sequencing was performed on an ABI 3730XL
sequencer (Applied Biosystems, Foster City, California).
Phylogenetic Analysis— DNA sequences were edited and aligned
manually in SeqMan 4.0 (DNAStar Inc., Madison, Wisconsin), resulting in
alignments of 706 positions (ITS), 2,631 positions ( matK ), and 1,089 posi-
tions ( trnL-F ). Areas of the alignments where the assessment of homol-
ogy was ambiguous were excluded from the analyses. Gaps that could
be inferred unambiguously were scored as presence/absence characters,
following the “simple gap coding” method of Simmons and Ochoterena
(2000) .
Parsimony Analysis— Parsimony analyses were performed in the pro-
gram PAUP* 4.0b10 ( Swofford 2002 ), assuming unordered character state
transformation and equal character weights ( Fitch 1971 ). Shortest trees
were obtained using the ‘branch and bound’ method. Bootstrap percent-
ages were calculated as follows: 500 replicates of 100 random addition
sequences (RAS), with TBR branch swapping, saving a maximum of 50
trees per replicate. To test for conflict among data partitions, topologies
and bootstrap support were compared across data partitions. Congruence
among data partitions was further assessed with partition homogeneity
tests as implemented in PAUP* (with 100 replicates of 1,000 RAS, sav-
ing a maximum of 10 trees). Based on the results of congruence testing,
the data partitions were combined in further analyses. One accession
( C. tomentosum [e33]) with conflicting placements in the individual anal-
yses was excluded from the combined analyses. All other accessions,
including those with missing data, were included in the combined analy-
ses. Four of 26 Centrolobium accessions were lacking sequences for at least
one of the three sequenced cpDNA regions (corresponding to 12.3% of
aligned bases scored as missing data). Six Centrolobium accessions were
missing ITS sequences (13.8% of aligned bases).
Bayesian Inference— The ITS and chloroplast datasets were com-
bined in analyses using Bayesian inference, as implemented in MrBayes
version 3.1.2 ( Huelsenbeck and Ronquist 2001 ). The dataset was divided
into partitions corresponding to ITS, trnL-F and matK , and rates and sub-
stitution models were allowed to vary across the partitions. Modeltest 3.06
( Posada and Crandall 1998 ) was used to select the substitution model best
fitting each sequence data partition. Geoffroea decorticans was designated
as the single outgroup taxon. Two searches, each with four simultaneous
MCMC chains, were run for 5,000,000 generations, saving one tree per
100 generations. Convergence was confirmed by analysis of the samples
using TRACER (Rambaut and Drummond 2003). The burn-in values were
determined empirically from the likelihood values of saved trees. The
burn-in was discarded and 50% majority rule consensus trees calculated
together with approximations of the PP for the observed bipartitions from
the remaining trees. Trees have been posted to TreeBASE (study number
S2301).
Molecular Dating— The combined chloroplast and ITS data were used
in molecular dating analyses. To test whether the sequence data had
evolved in a clocklike fashion, a likelihood ratio test was performed on
the first of the most parsimonious tree topologies. Likelihoods of the data
with and without a molecular clock constraint were calculated and the
likelihood ratio statistic compared with χ
2 critical value with 20 degrees of
freedom ( Felsenstein 1981 ). The test rejected the molecular clock assump-
tion ( p < 0.001).
F ig . 1. Distribution map for species of Centrolobium . C. yavizanum =
filled circles; C. paraense = filled squares; C. ochroxylum = filled triangles
(point up); C. microchaete = pentagons; C. tomentosum = open squares;
C. sclerophyllum = filled triangles (point down); C. robustum = stars.
2009] PIRIE ET AL.: CENTROLOBIUM 347
We therefore applied two different molecular dating methods; penal-
ized likelihood (PL; Sanderson 2002a ) and the Bayesian uncorrelated
relaxed clock method ( Drummond et al. 2006 ) to correct for rate heteroge-
neity. While the latter method directly incorporates uncertainty in topol-
ogy and branch length estimation into the error margins, such uncertainty
can also be approximated when applying PL by using a variable sample of
trees as input (see below). Both methods can also estimate absolute ages by
constraining the age(s) of one or more nodes (molecular clock calibration).
In all analyses, Geoffroea decorticans and the accessions of Platymiscium ,
which form two branches of a basal trichotomy in the cpDNA and com-
bined analyses, were ‘pruned’ from the trees, leaving a newly defined
root node, corresponding to the most recent common ancestor (mrca)
of Grazielodendron and Centrolobium and to node 51 in the analysis Lavin
et al. (2005) . The age of this node was fixed at 41.9 mya following the
results of Lavin et al. (2005 ; node 51, p.579, Fig. 2 ). The fossil Tipuana ecu-
atoriana Burnham ( Burnham 1995 ) was used to impose a minimum age of
7.9 million years on the Tipuana stem node. Both calibrations are indicated
in the chronogram .
Penalized likelihood (assuming rate autocorrelation) was implemented
in the software package r8s ( Sanderson 2002b ). The optimum smooth-
ing parameter value was determined using the cross validation proce-
dure in r8s, based on the MrBayes tree with the highest likelihood score.
This value was used to rate smooth the 1,000 last sampled trees from the
MrBayes analysis (thus including variation in both topology and branch
lengths). Mean node ages and standard deviations were summarized
from the resulting ultrametric trees.
The Bayesian uncorrelated relaxed clock method was implemented
in the software package BEAST ( Drummond and Rambaut 2007 ). We
applied a lognormal distribution to the rates. The root node was effec-
tively fixed at 41.9 (mya) by the designation of a uniform prior with mini-
mal bounds. Preliminary analyses indicated that the posterior distribution
of the age of the Tipuana stem node was greatly in excess of 7.9 mya, and
this prior was therefore omitted from the final BEAST analyses. The mcmc
was run for five million generations, with sampling every 1,000 genera-
tions, after which convergence was confirmed by analysis of the samples
using TRACER (Rambaut and Drummond 2003). Burn-in values were
determined empirically from the likelihood values of sampled trees and
the burn-in from each run was excluded. The remaining trees were used
to calculate a 50% majority rule consensus tree and node ages (mean and
95% confidence limits).
Taxonomic Treatment— The taxonomic account that follows is based
on examination of 280 collections made available by the following her-
baria: A, AAU, BM, BR, E, F, GH, K, LPB, MO, NY, RB, U, US, and USZ.
A list of exsiccatae is provided in Appendix 2. Conservation status is
based on IUCN categories of threat criteria (IUCN 2001). Vegetation type
is assigned according to the following four major biomes: seasonally dry
tropical forest (SDTF), lowland rain forest, savanna, and restinga, with
more specific vegetations described as appropriate. Definitions of SDTF
and savanna follow Pennington et al. (2000) . Savanna and SDTF change
to rain forest along transitions of rainfall; rain forest grows in areas of
higher annual rainfall, but a key factor is that rainfall is less strongly sea-
sonal (with fewer entirely dry months) than in areas that support SDTF or
savanna ( Thomas and de Vansconcellos Barbosa 2008 ). Restinga is a scrub
vegetation growing on well-drained, sandy soils between the beach and
coastal rain forests of Eastern Brazil. It forms a narrow band of vegetation
ranging from a few hundred meters to a few kilometers wide ( Thomas
and de Vansconcellos Barbosa 2008 ).
Results
Phylogenetic Analysis— Parsimony analysis of matK
and trnL-F resulted in relatively unresolved topologies (not
shown), particularly in the case of trnL-F . The phylogenies
were without conflicting nodes with 70% bootstrap sup-
port (BS; data not shown). The partition homogeneity test
for the chloroplast data did not indicate significant conflict
( p = 0.37), and the data were combined in further analyses.
The monophyly of Centrolobium received high support (98%
BS), but the only internal nodes subject to 70% BS corre-
sponded to multiple accessions of C. ochroxylum and of
C. paraense , plus a strongly supported sister group relation-
ship between one of the accessions of C. microchaete (e1) and
one of C. tomentosum (e33) ( Fig. 2 ). Details of the parsimony
analyses, including numbers of variable and informative char-
acters, are presented in Table 1 . Direct comparison of analyses
of the different markers is complicated by the differing sam-
ple sizes for each region (as indicated in Table 1 ). Inclusion of
the single accession of Centrolobium yavizanum , which lacked
a matK sequence, resulted in a much larger number of most
parsimonious trees.
Analysis of ITS resulted in a consensus topology that also
included groupings of the multiple accessions of C. ochrox-
ylum and of C. paraense , but which differed in a number of
places from the cpDNA consensus ( Fig. 2 ). Comparison of
trnL-F and ITS data partitions using the partition homogene-
ity test did not indicate significant conflict ( p = 0.07), but com-
parison of matK and ITS, and of combined trnL-F and matK
with ITS, did indicate conflict (both p = 0.01). Only one topo-
logical difference received 70% BS, a sister group relation-
ship between the two accessions of C. microchaete (e1 and e6)
in the ITS tree, which conflicted with C. microchaete accession
e1 grouping with C. tomentosum accession e33 in the combined
cpDNA analysis ( Fig. 2 ). To eliminate this source of conflict,
C. tomentosum (e33) was removed from the combined analy-
ses of all of the data partitions. The partition homogeneity test
in the absence of this taxon did, however, still indicate conflict
between trnL-F / matK and ITS ( p = 0.01).
Parsimony and Bayesian analyses of the combined cpDNA
and ITS data produced consistent results. BS and posterior
probability (PP) values are indicated on the Bayesian 50%
majority rule consensus topology ( Fig. 3 ). Of the three mark-
ers sequenced, ITS was the most variable. However, the rel-
ative contribution of informative characters from nuclear
(ribosomal) and chloroplast encoded markers was similar
(see Table 1 ), and the combined analyses included roughly
double the number of informative characters in compari-
son to the individual analyses. Nevertheless, resolution and
support values showed little improvement in the combined
analyses (compare Figs. 2 and 3 ).
Molecular Dating— r8s Analyses — The cross validation
test for PL selected an optimal smoothing parameter of 3.2,
which was used in all further PL analyses. The mean age of the
stem node of Tipuana was 28.1 mya, much older than the 7.9
mya Tipuana fossil. The mean age of the mrca of
Centrolobium
summarized across 1,000 MrBayes trees was 7.5 mya, with
a standard deviation of 0.9, a minimum age of 5.1 mya and
maximum of 11.0 mya.
BEAST Analyses Age estimations were consistent with
those derived using r8s, although the (95%) confidence inter-
vals were wider ( Fig. 4 ), thus providing a more conservative
test of biogeographic hypotheses. In the following discussion,
we will therefore refer to the results obtained using BEAST.
The Tipuana stem node was estimated at between 14.2 and
39.7 mya, also much older than the Tipuana fossil. The confi-
dence interval for the mrca of Centrolobium was between 3.5–
15.9 mya.
Discussion
Species Delimitation— We mainly agree with the species
delimitations of Rudd. The species of Centrolobium display
two broad patterns of distribution. The first, exhibited by C.
microchaete and C. paraense , is of wider distribution, in the case
of C. microchaete with populations found in different disjunct
areas. The second, exhibited by C. robustum , C. tomentosum , C.
sclerophyllum , C. ochroxylum, and C. yavizanum , is of restriction
348 SYSTEMATIC BOTANY [Volume 34
F ig . 2. Shortest trees (arbitrarily selected) from parsimony analysis of (a) nuclear ribosomal ITS and (b) chloroplast encoded matK and trnL-F data.
Bootstrap support (where > 50%) is indicated above the branches. Nodes that collapse in the strict consensus are indicated by arrows.
2009] PIRIE ET AL.: CENTROLOBIUM 349
to a relatively narrow geographic range. Some species display
relatively large amounts of morphological variation across
their ranges, e.g. Brazilian versus Bolivian populations of
C. microchaete ; and others relatively little, e.g. C. paraense .
Monophyly of Centrolobium and Species Relationships—
Analysis of both chloroplast and nuclear sequence data
yielded strong support for the monophyly of Centrolobium .
Within Centrolobium , multiple accessions of the same spe-
cies sometimes formed monophyletic groups. However, few
other conclusions regarding phylogenetic relationships could
be drawn, due to low variation in the sequence data and phy-
logenetic conflict among data partitions. With the exclusion
of just one taxon from the analyses, we were able to eliminate
conflict supported by > 70% BS. Levels of phylogenetic reso-
lution were not substantially greater in the combined analyses
relative to the individual analyses, perhaps due to persisting
phylogenetic conflict caused by incomplete lineage sorting or
hybridization.
Biogeographic History— Pennington et al. (2004 , 2006 ) noted
a preponderance of pre-Pleistocene speciation and high phy-
logenetic geographic structure in South American SDTF
clades. In Centrolobium , phylogenetic geographic structure
is limited, although the timing of lineage diversification also
appears to be largely pre-Pleistocene. Centrolobium began
diversifying a minimum of 3.5 mya. The most north-westerly
distributed species, C. yavizanum and C. ochroxylum , diverged
from a common ancestor at least two mya, before Pleistocene
climatic changes. The estimated ages of divergences between
species from Bolivia and Brazil (plus C. paraense from Brazil,
Guyana, Venezuela, and Colombia) span the Late Miocene
and Pliocene, with the margins of error for some nodes also
entering the Pleistocene. The distributions of C. yavizanum
and C. ochroxylum are separated from those of other species of
Centrolobium by the Andes Mountains, which may explain the
relatively deep phylogenetic divergence. There is little phylo-
genetic geographic structure in the species found east of the
Andes, as illustrated by the widespread, overlapping distri-
butions of C. robustum and C. microchaete and the widely dis-
junct distribution of C. microchaete .
The levels of phylogenetic geographic structure in Centrolo-
bium are lower than those observed in other South American
SDTF clades ( Pennington et al. 2004 ). This may be because
some Centrolobium species are not SDTF specialists, as they
also grow in more humid, less seasonal formations such as
the Amazon rain forest (e.g. C. paraense ) and the Brazilian
Atlantic Coastal rain forest (e.g. C. robustum, C. tomentosum ).
Lavin (2006) and Pennington et al. (2006) argued that rain
forest clades show less geographic structure than dry forest
clades because of the greater continuity of rain forest com-
pared to SDTF. The wide ecological tolerance of Centrolobium
may have allowed it greater opportunity for migration, lead-
ing to the lack of geographic structure observed in the east of
its range.
Taxonomic Treatment
Centrolobium Mart. ex Benth. , Ann. Mus. Vind. 2: 95. 1839.
Benth., in Mart. Fl. Bras. 15.1: 263–266. 1862; Rudd,
J. Wash. Acad. Sci. 44: 284–288. 1954; Lima, Arq. Jard. Bot.
Rio de Janeiro 27: 177–191. 1985.—TYPE: Centrolobium
robustum (Vell.) Mart. ex Benth.
F ig . 3. 50% majority rule consensus tree from Bayesian analysis of
combined ITS and chloroplast data. Posterior probabilities of clades
(where > 0.50) are indicated above the branches, bootstrap support from
parsimony analyses (where > 50%), below.
Table 1. Details of maximum parsimony analyses (all employing ‘branch and bound’ method as implemented in PAUP* to find all shortest trees).
Combined analyses in which taxa with missing data were excluded are indicated by “
1 ”, otherwise all taxa with at least one marker sampled were
included.
No. taxa Included chars Variable chars Pars.inf. chars Inf. indels No. trees Tree length CI RI
matK 22 2634 299 152 10 23 360 0.886 0.882
trnLF 17 1119 89 31 9 84 98 0.908 0.859
matK/trnLF 23 3753 388 183 19 3192 462 0.883 0.868
matK/trnLF 1 16 3753 349 99 17 16 409 0.890 0.815
ITS 19 673 260 164 - 24 484 0.698 0.653
Combined 23 4426 647 344 19 114 957 0.778 0.736
Combined
1 12 4426 503 160 15 4 663 0.848 0.649
350 SYSTEMATIC BOTANY [Volume 34
F ig . 4. Chronogram from BEAST molecular dating analysis. The error margins for age estimations are indicated by bars at the nodes (thicker, lighter
bars representing BEAST posterior distributions, thinner, darker bars representing the mean ± SD summarised from penalized likelihood rate smoothing of
1,000 MrBayes trees). Timescale and age calibration points A (secondary calibration from Lavin et al. [2005]) and B (fossil Tipuana [ Burnham, 1995 ]) are also
indicated, as are the timeframes of speciation within Centrolobium (large filled rectangle), and Pleistocene climatic changes (rectangle above the time scale).
2009] PIRIE ET AL.: CENTROLOBIUM 351
Trees, 4–35 m tall; trunk commonly buttressed at base, to
100 cm in diameter; bark grayish, vertically fissured, the sap
red; wood yellowish or orange with streaks of red, purple,
or black; twigs mostly densely pubescent, the twigs, leaves,
inflorescences, and fruits dotted with reddish-orange peltate
glands. Leaves imparipinnate, (5-)9–17(-23)-foliolate; stipules
usually caducous, rarely persistent, deltoid to broadly orbic-
ular, the apex acute; leaflets ovate, elliptic or oblong, sym-
metrical or asymmetrical, the base cordate to rounded, the
margins entire, the apex acute to shortly acuminate, the vena-
tion brochidodromous. Inflorescences of terminal panicles;
bracts narrowly to broadly ovate, rarely fan-shaped (only
in C. ochroxylum , Fig. 5K ), usually persistent, rarely cadu-
cous, reflexed or upright, the apex acute or obtuse; bracteoles
narrowly or broadly ovate, persistent or caducous. Flowers
10–22(-25) mm long, sweetly scented; calyx tubular, cam-
panulate or urceolate, sometimes adaxially gibbous, the base
acute, obtuse or truncate, 5-lobed, the three abaxial lobes
unequal, the apices acute or obtuse, the two adaxial lobes
fused for almost their entire length, the apex emarginate;
corolla yellow, sometimes suffused with red or violet, gla-
brous, the standard petal spatulate or obovate, with a short,
thick claw with darker central marking, reflexed, the wing
petals basally auriculate, with a long narrow claw, the keel
petals fused for 1/3 of their lower margin, the upper mar-
gin auriculate at the base; stamens 10, fused monadelphously
into an open sheath, the anthers 1–1.5 mm long, dorsifixed,
versatile. Fruits 1-few-seeded samaras, 6.8–22(-27) cm long;
seed chamber basal, subcylindrical, externally echinate; wing
distal, chartaceous or coriaceous, usually indumented and
covered with long-stalked to subsessile peltate glands; stylar
spine fused to the samara wing for a variable distance, the
free apex at a ca. 45°, 90° or 120° angle with the wing mar-
gin; spines of seed chambers sparsely to densely tomentose;
seeds 2–3 × 1–1.5 cm, oblong, testa thin, green, albuminous.
Figures 5A-V , 6A-U .
Distribution— From Panama, throughout northern South
America to Bolivia and Santa Catarina state in southern
Brazil, but largely avoiding the Amazon Basin, from sea level
to 1,000 m elevation. Figure 1 .
Habitat— Seasonally dry tropical forests to lowland rain
forests.
Key to CENTROLOBIUM Species
1. Stipules persistent, ca. 15 × 15 mm, leaf-like, broadly ovate ..................................................................................................................................... 3. C. paraense.
1. Stipules caducous, 2–15 × 2–4 mm, not leaf-like, narrowly ovate, deltate. .................................................................................................................................. 2
2. Calyx urceolate. Bracts deltoid ( Fig. 5P ). Leafl ets 5–11 × 2–5 cm, narrowly ovate or elliptic, coriaceous ........................................... 5. C. sclerophyllum.
2. Calyx campanulate or tubular. Bracts fan-shaped, narrowly or broadly ovate. Leafl ets (4-)6–20 × (2–)4–9(-14) cm, oblong, elliptic or ovate, charta-
ceous or/to membranaceous, rarely coriaceous. ................................................................................................................................................................... 3
3. Leafl ets membranaceous or coriaceous, the lower surface matted with twisting trichomes and dense glands. ...................................................... 4
4. Bract fan-shaped, 2–2.5 × 3–3.5 mm ( Fig. 5K ). Flower 15–18(-25) mm long. Samara covered by sparse trichomes. Western Ecuador and north
western Peru ....................................................................................................................................................................................... 2. C. ochroxylum .
4. Bract narrowly ovate, 4–5 × 3–4 mm ( Fig. 5T ). Flower 18–22 mm long. Samara covered by dense trichomes. Central eastern Brazil
............................................................................................................................................................................................................... 6. C. tomentosum.
3. Leafl ets chartaceous or coriaceous, the lower surface glabrous or with sparse straight to curly trichomes and glands. ........................................ 5
5. Flowers 10–12 mm long. Calyx lobes 2–5 mm long, the apices acute. Leafl ets usually chartaceous, rarely membranaceous. ................ 6
6. Leaves 9–13(-19)-foliolate. Leafl ets elliptic or oblong, the base asymmetric ( Fig. 5H ). Bracts ovate, 1.5–4 × 1.5–3 mm. Fruiting stipe
1–10 mm long. Eastern Brazil and eastern Bolivia .......................................................................................................... 1. C. microchaete.
6. Leaves 13–17-foliolate. Leafl ets elliptic, the base symmetric ( Fig. 5U ). Bracts narrowly ovate, 3–5 × 1.5–3 mm. Fruiting stipe 20–30 mm
long. Southern Panama and northern Colombia .............................................................................................................. 7. C. yavizanum.
5. Flowers 17–18 mm long. Calyx lobes 5–7 mm long, the apices obtuse. Leafl ets usually coriaceous, rarely membranaceous
............................................................................................................................................................................................................ 4. C. robustum.
1 . C entrolobium microchaete (Mart. ex Benth.) Lima, Arq.
Jard. Bot. Rio de Janeiro 27: 180. 1985. C. robustum (Vell.)
Mart. ex Benth. var. microchaete Mart. ex Benth. in Mart.
Fl. Bras. 15(1): 263. 1862.—TYPE: BRAZIL. Rio de Janeiro:
“prope Canta Gallo”, 1859, Th. Peckolt s.n. [in BR the col-
lection contains a label: Peckolt n. 260 ] (lectotype BR!).
Tree 7–26 m tall; trunk diameter 10–70 cm, the bark black-
ish grey with light vertical and longitudinal fissures, inflo-
rescences drying blackish brown. Leaves 9–13(-19)-foliolate;
stipules caducous, not observed; leaflets 4–13.4 × 2–4 cm,
usually elliptic or oblong, rarely ovate, the base asymmetric,
rounded, the apex shortly acuminate or acute, chartaceous or
coriaceous, both surfaces shiny green, covered with sparse pel-
tate glands and sparse trichomes, or glabrous. Inflorescence
axes, peduncles and calyces matted with dense dark brown
trichomes and peltate glands; bracts 1.5–4 × 1.5–3 mm, ovate,
persistent, reflexed, the apex acute; bracteoles 2–3 × 0.5–1.5
mm, elliptic, persistent, the apex obtuse or acute. Flowers
10–12 mm long; pedicel 2–3 mm long; calyx campanulate, the
tube 5–7 mm, the base acute or obtuse, the three abaxial lobes
2–3 mm long, the apices acute, the two fused adaxial lobes
2–3 mm long, the apex emarginate, acute; corolla reddish-yel-
low to dark tawny-brown, venation red; standard 10–11 × ca.
9 mm, suborbiculate, reflexed at anthesis, the wing petals ca.
10 × 4 mm, auriculate at the base of the blade, the keel petals
ca.10 × 4 mm, auriculate at the base of the blade, fused for 1/3
of the lower margins; stamens 9–11 mm long, fused basally for
1/3 of their length. Samara 6.8–17.5 × 2.9–8 cm; seed chamber
ca. 4 × 4 cm, the spines 6–17 × 0.5–1 mm, basal part sparsely
indumented with curly trichomes, ca. 1 mm long, and short-
stalked peltate glands, the apical part with similar trichomes,
but lacking peltate glands; wing chartaceous with indumen-
tum and glands similar to those on the seed chamber; stylar
spine fused to samara wing for 8–20 mm, the free apex 5–15
mm long, oriented at a 45° angle from the wing margin; stipe
10–15 mm long; seeds ca. 1 × 0.5 cm, oblong to triangular, the
testa glossy, light brown. Figures 5A , H-I; 6A-C.
Distribution— Eastern Brazil, from Santa Catarina to Ceará
and Minas Gerais, and eastern Bolivia in the departments of
La Paz and Santa Cruz, 30–680 m elevation. Figure 1 .
Habitat and Phenology— Rain forest and SDTF. In Brazil
it prefers riverine habitats, while in Bolivia it grows in rain
forest to SDTF. Prolific flowering from November to May in
352 SYSTEMATIC BOTANY [Volume 34
F ig . 5. A–V. Part of an inflorescence, an open calyx with detail of indumentum, a bract, and a leaflet of all Centrolobium species. A, H, I. C. microchaete
[from Mendonça et al. 136 , K]. B, J, K. C. ochroxylum [from Klitgaard et al. 421 , K]. C, L, M. C. paraense [from Jansen-Jacobs et al. 4192 , K]. D, Q, R. C. robustum
[from Corcovado 15921 , K]. E, N, O, P. C. sclerophyllum [from Occhioni 16363 , K]. F, S, T. C. tomentosum [from Hage 1646 , K]. G, U, V. C. yavizanum [from Zarucchi
et al. 4957 , K]. Scalebars: 1 mm in I, K, L, P, Q, T, V (shared: 3); 5 mm A, B, C, D, E, F, G (shared: 1); 10 mm in the leaflets of H, J, M, N, Q, S, U (shared: 2); and
5 mm in the inflorescences of H, J, M, O, Q, S, U. Drawn by Margaret Tebbs.
2009] PIRIE ET AL.: CENTROLOBIUM 353
Fig. 6. A–U. The samara seed chamber and detail of the base of a seed chamber spine; and part of the samara wing and detail of the samara wing of
all Centrolobium species. A, B, C. C. microchaete [from Saldías 377 , K]. D, E, F. C. ochroxylum [from Palacios et al. 818 , K]. G, H, I. C. robustum [from Martius
160 , K]. J, K, L. C. sclerophyllum [from Santos 4 , K]. M, N, O. C. yavizanum [from Zarucchi et al. 4957 , K]. P, Q, R. C. paraense [from Davis 786 , K]. S, T, U. C.
tomentosum [from Hatschbach & Zelma 49536 , K]. Scalebars: 1 mm in A, D, G, J, M, P, S (shared: 3); 20 mm B, E, H, K, N, Q, T (shared: 1); and 2 mm in C, F,
I, L, O, R, U (shared: 2). Drawn by Margaret Tebbs.
354 SYSTEMATIC BOTANY [Volume 34
Brazil and from March to April in Bolivia; fruit set from May
to November in Brazil, and from May to October in Bolivia.
Centrolobium microchaete is deciduous, and the leafless stage
coincides with fruit set at the end of the dry season.
Uses and Vernacular Names— Used in the manufacture
of fine furniture, laminate sheets for decorative panels, floor
panels, handicrafts, and railway sleepers ( Lorenzi 2002 ).
“Araribá” (Bahia, Santa Catarina, Parana, Rio de Janeiro and
Minas Gerais), “gororoba” (Bahia), “lei nova” (Minas Gerais),
“petimujú,” “putumujú,” “putumujú-mirim” (Bahia and Ceará),
“tarara,” “tarara amarilla,” and “madera canaria” (Bolivia).
Conservation Status— VU A1cd ~ vulnerable because of a
population size reduction of 50% over the last 10 yr due to
a decline in area of occupancy, extent of occurrence and/or
quality of habitat, and due to actual and potential levels of
exploitation.
Representative Specimens Examined— BOLIVIA. Beni: Balliv ian, Feb
1990, Smith et al. 14429 (MO). La Paz: Nor Yungas, 1 km from Buena
Vista, 5 May 1989, (fr.) Beck 17203 (K, MO, NY, LPB). Santa Cruz: Andres
Ibañez, Santa Cruz, Velasco Ave, north of Irala Ave, 2001, (fr.) Pirie &
Chatrou 1 (U).
BRAZIL. Bahia: ROAD Itacaré-Ubaitaba, 16 May 1966, Belém & Pinheiro
2219 (NY). Minas Gerais: Caratinga 10 Nov 1993, Klitgaard et al. 17 (AAU,
K, RB). Paraná: Morretes 10 Mar 1983, Hatchbach 46250 (AAU, F, K).
Santa Catarina: Morro de Fazenda, Itajai, 10 Jun 1954, Reitz & Klein 1877
(K, NY).
Notes— Centrolobium microchaete is primarily distinguished
by its small (10–12 mm) flowers vs. 12–22 mm in the other
Brazilian species. From C. robustum in particular it is fur-
ther distinguished by 9–13(-19)-foliolate leaves and 6.8–17.5
cm long samaras vs. (13-)15–19(-23)-foliolate leaves and
(10-)18–26 cm long samaras. C. microchaete is the only species
of Centrolobium found in Bolivia. Despite variation in the size
of fruits and flowers (generally larger in Bolivian material) all
Bolivian specimens share characteristics diagnostic of C. micro-
chaete , notably the persistent recurved bracts, indument type
and longer stipe found in neither C. ochroxylum nor C. tomen-
tosum (with which they have sometimes been confused).
2. C entrolobium ochroxylum Rose ex Rudd, J. Wash. Acad.
Sc. 44: 9. 1954.—TYPE: ECUADOR. El Oro: Portovelo
(near Zaruma), 6 Oct. 1918, J.N. Rose & G. Rose 23370
(holotype: US (no. 1022875)!; isotype: GH!, NY!).
Tree 6–30 m tall; trunk diameter 15–60 cm, sometimes but-
tressed to ca. 1 m tall; bark pale brown with grey patches
to whitish buff, smooth or lightly vertically and longitu-
dinally fissured, the slash exuding a cream/pink sap, the
twigs and the inflorescences drying blackish brown. Leaves
(7-)9–13(-17)-foliolate; stipules ca. 2 × 2 mm long, deltoid,
caducous; leaflets 6–20 × 5–14 cm, ovate or elliptic, the base
symmetric, cordate or rounded, the apex shortly acuminate,
membranaceous, both the surfaces covered with sparse to
dense curly trichomes, the lower surface also with dense pel-
tate glands. Inflorescence axes, peduncles and calyces with
dense dark brown (nearly black) trichomes and dense pel-
tate glands; bracts 2–2.5 × 3–3.5 mm, fan-shaped, caducous,
reflexed, the apex rounded; bracteoles 1.5–2 mm × 0.8–1.25
mm, narrowly ovate, the apex acute, usually caducous.
Flowers 15–18(-25) mm long; pedicel 4–6 mm long; calyx cam-
panulate, 12–16 mm long, the tube ca. 8 mm long, the three
abaxial lobes 5–8 mm long, the apices acute, the base obtuse
or acute, the two fused adaxial lobes ca. 5 mm long, the apex
emarginated, acute; corolla yellow, venation dark orange,
the standard petal ca. 15 × 12 mm, suborbicular, reflexed by
anthesis, with a central dark orange or pale yellow blotch,
the wing petals ca. 15–4 mm long, blade oblong, auriculate at
the base of the blade, and with the basal ½ covered in sculp-
turing, the keel petals ca. 15 × 5 mm, the blade triangular,
auriculate at the base of the upper margin, fused for ¼ of the
lower margins; stamens ca. 15 mm long, fused basally for ½
their length. Samara 10.5–27 × 4.1–10 cm; seed chamber 3–5
cm in diameter, the spines 15–40 × 0.5–1 mm, the basal parts
sparsely indumented with straight trichomes ca. 0.1 mm long
and subsessile peltate glands; wing chartaceous, glabrous
but with glands as on the seed chamber; stylar spine fused to
samara wing for 3–14 mm long, the free apex 4–16 mm long,
at a 120° angle with the wing margin; stipe 6–10 mm long;
seeds 2–3 × 1–1.5 cm, oblong, testa thin, green. Figures 5B ,
J-K; 6D-F.
Distribution— Ecuador, west of the Andes and the adjacent
Tumbes area in Peru at 0–350 m. Figure 1 .
Habitat and Phenology— The species is found in SDTF
on a mixture of soils. It is sometimes cultivated as a hedge-
row tree. Flowering has been recorded in October and fruit
set in September, February, May, and June. Centrolobium
ochroxylum is a deciduous species in which the leafless stage
coincides with fruit set, and trees develop new leaves while
flowering.
Uses and Vernacular Names— It is cultivated as a hedge tree
and as coffee shade, and for timber used in house, boat, and
furniture construction ( Little and Dixon 1969 ). “Amarillo”,
“amarillo de Guyaquil”, “amarillo lagarto”, “araribá” ( Rudd
1954 ; Acosta-Solis 1947 ).
Conservation Status— Centrolobium ochroxylum is in the
Red List of endemic species of Ecuador 2000 ( Neill 2000 ),
even though no threat category was attached to it in that pub-
lication. In a study that identifies biodiversity conservation
priorities for coastal northern South America, Peralvo et al.
(2007) calculated the percentage decline in potential distribu-
tion (km
2 ) to be 82.2% for C. ochroxylum , but still no category
was attached. The conservation status of the species is, based
to the available information, judged to be Near Threatened
(NT).
Representative Specimens Examined— ECUADOR. Cañar: La
Troncal, 2, Mar 1996, Neill et al. 10527 (MO). El Oro: Bosque Petrificado
Puyango, el Mirador, Klitgaard et al. 421 (AAU, K, LOJA, NY).
Esmeraldas: Anchayacu, 9 Nov 1994, Pennington et al. 15012 (K, NY).
Guayas: Cerro Azul, 18 Mar 1980, (fl.) Dodson et al. 9651 (US, MO). Loja:
Hacienda Banderones, 8 Jun 1997, Klitgaard et al. 551 (AAU, K, LOJA,
NY). Los Ríos: Canton Vinces, Jauneche Forest, 1 Oct 1979, Dodson
et al. 8646 (MO). Manabi: Jama Canton, 18 Dec 1998, Neill et al. 11644
(K, MO, NY). Napo: Estación Experimental INIAP-Payamino, 3 km north
of Coca, 22 Sep 1985, (fr.) Palacios et al. 818 (K, MO, AAU, NY).
PERU. Cerro de Pasco, Central Selva, 17 May 1985, Hartshorn et al. 2763
(MO). Tumbes: Matapalo, Zarumilla, 7 Feb 1993, Diaz et al. 6606 (MO).
Notes— Centrolobium ochroxylum , the only species occurring
in Ecuador, and just ranging into adjacent northwestern Peru,
is easily distinguished by its large (10–27 cm long) samaras
in which the stylar spine is at a 90–120° angle with the wing
margin (in the remaining species it is mostly at a 45° to 90°
angle).
3. C entrolobium paraense Tul., Arch. Mus. Par. 4: 87. 1844.—
TYPE: GUYANA. Pirara, 1841, Schomburgk 314 (holotype
P!; isotypes: BM!, F!, G n.v., K!)
Centrolobium paraense var. orinocense Benth., in Mart. Fl. Bras.
15.1: 266. 1862. C. orinocense (Benth.) Pittier, Bol. Tecn.
Minist. Agric. 5: 123 1944.—TYPE: VENEZUELA. Bolivar:
“prope Angustura” (Ciudad Bolívar), May 1851, Purdie
s.n. (holotype: K!, NY photo of type no. 2678).
2009] PIRIE ET AL.: CENTROLOBIUM 355
Centrolobium patinense Pittier, J. Wash. Acad. Sci. 5: 470.
1915.—TYPE: PANAMA: Darién, Punta Patiño, Pittier
6611 (holotype: US!).
Trees 4–30 m tall; trunk diameter 15–100 cm; bark red-
dish-brown with fine vertical fissures, the slash exuding
reddish sap; young twigs and inflorescences drying a green-
ish-cream color. Leaves (5-)9–11-foliolate; stipules ca. 15 × 15
mm, broadly ovate leaf-like, usually persistent, rarely cadu-
cous; leaflets 7–17 × 4–8 cm, ovate to broadly ovate, the base
usually symmetric, subcordate or rounded, the apex acumi-
nate or obtuse, membranaceous, the upper surface glabrate
or with dense curly trichomes, the lower surface venation
covered with indumentum similar to that of the upper sur-
face, the lower surface lamina with dense peltate glands.
Inflorescence axes, peduncles and calyces with dense rusty
brown, curly trichomes and dense peltate glands; bracts
5–10 × 5–10 mm, broadly ovate, persistent, the apex acum-
i nate; bracteoles ca. 10 × 5 mm, broadly ovate, persistent.
Flowers 12–19 mm long; pedicel 5–8 mm long; calyx cam-
panulate, the tube ca. 9–10 mm long, the base acute, the three
abaxial lobes 6–8 mm long, the two fused adaxial lobes ca.
6 mm long, the apex emarginated, obtuse; corolla dark yel-
low to orange, the venation red, the standard petal 12–14 ×
12–14 mm, orbicular, reflexed by anthesis, with central
bright red to purple blotch, the wing petals 12–15 × 5–5 mm,
blade ovate, auriculate at the base of the upper margins and
with the outer surface covered in sculpturing, the keel pet-
als 14–15 × 5 mm, the blade boomerang-shaped, auriculate
at the base of the upper margins, fused for 1/3 of the lower
margins; stamens 12.5–13 mm long, fused basally for ¼ to ½
of their length. Samara 12–22 × 4–8 cm; seed chamber 2–4.5
cm in diameter, the spines 10–30 × 1–1.5 mm, the basal part
densely indumented with curly trichomes ca. 1 mm long and
subsessile peltate glands, the apical part similar to basal part
or glabrate; indumentum dense with ca. 0.5 mm long, curly
trichomes and subsessile peltate glands; wing chartaceous,
with indumentum and glands as of the seed chamber; stylar
spine fused to samara wing for 15–25 mm long, the free apex
7–12 mm long, often at a 90° angle rarely at a 45° angle with
the wing margin; stipe 10–20 mm long; seeds not observed.
Figures 5F , S-T; 6P-R.
Distribution— Centrolobium paraense is distributed along the
Caribbean coast of Guyana, Venezuela and adjacent Colombia
and Panama, stretching into Amazonian Brazil reaching as
far south as Rio Branco in Roraima, at altitudes between 50
and 350 m above sea level. It remains unclear if the species is
native to or introduced into Trinidad. Figure 1 .
Habitat and Phenology— Centrolobium paraense is found in
a wide range of habitats from being a dominant species in
the SDTF of El Sombrero-Chaguramas-Tamanaco in Guarico,
Venezuela to the savannas of Guyana and tropical rain for-
est in Brazil. Flowering specimens were collected in March
(Brazil), otherwise from May to September. Fruit sets from
February to August The species is often described as drop-
ping most or all of its leaves between February and April.
Uses and Vernacular Names— “Pau Raihna” (Brazil);
“Kartang” (“Makusi”), “redwood” or “shipuradai” (“wapisi-
ana”) (Guyana); “baluster”, “cartán” or “colorado” (Venezuela);
“Colorado” (Colombia); “porcupine tree” (Trinidad).
Conservation Status— VU A3cd ~ vulnerable because of a
population size reduction of 30% over the next 10 yr due to
a decline in area of occupancy, extent of occurrence and/or
quality of habitat, and due to actual and potential levels of
exploitation.
Representative Specimens Examined— BRAZIL. Roraima: Rio Branco,
Boa Vista, Estação ecologica de Maracá, 8 Mar 1987, (fr.) Lewis 1420
(K, NY).
COLOMBIA. Atlántico: Barranquilla, Jan 19, 38, Elias 1580F (A, NY,
US).
GUYANA. Rupununi: Bushisland, Dadanawa, 9 Jun 1995, (fl.) Jansen-
Jacobs et al. 4012 (K, NY, U, US).
PERU. Madre de Dios: Cocha Cashu, Manu National Park, 31 Aug
1989, Foster 9940 (F, MO).
PANAMA. Darien: Punta Patiño, Darien, Pittier 6611 (US).
TRINIDAD. Port of Spain, Botanic Garden 10 Aug 1923, Fairchild 2844
(NY, US).
VENEZUELA. Anzoategui: Bergantín, 14 Mar 1945, Steyermark 61487 (F,
MO). Aragua: Entre Villa de Cura y San Juan de los Morros, 26 Dec 1923,
Pittier 11358 (NY, US). Bolivar: Monteco, 4 Aug 1978, Liesner & Gonzales
6000 (AAU, MO, NY). Guarico: Calabozo, Jun 1969, (fl.) Aristeguieta 7155
(MO, NY, US). Portuguesa: Guanare, 13 Mar 1982, (fr.) Liesner et al. 12654
(MO). Sucre: Fila de Guayuta, 18 Aug 1973, (fl.) Steyermark et al. 107725
(F, MO, NY). Zulia: Santa Rosa, 21 Aug 1967, (fr.) Steyermark & Fernandez
99547 (F, NY, U, US).
Notes— Centrolobium paraense is easily distinguished from
all other Centrolobium species by its large leafy stipules that
leave a large scar when dropping off and by the large, broadly
ovate bracts and bracteoles (5–10 × 5–10 mm).
Bentham (1862) described two varieties of C. paraense : var.
paraense and var. orinocense . The latter was subsequently raised
to the status of species by Pittier (1944) . Variety orinocense was
distinguished from the typical variety by glabrescent leaf-
lets with cordate bases vs. leaflets with dense indumentum
and rounded bases. In our studies these differences are, how-
ever, not upheld as leaves posses a combination of cordate
and rounded leaflet bases, and as the amount of indumentum
seems to be correlated with the age of the leaves.
4. C entrolobium robustum (Vell.) Mart. ex Benth., Ann.
Wien Mus. 2: 95.1839. C. robustum var. macrochaete Mart.
ex Benth. Mart. Fl. Bras. 15(1): 263. 1862. Nissolia robusta
Vell. Fl. Flum.: 298. 1825.—TYPE: BRAZIL. Rio de Janeiro:
“habitat in silvis maritimis”, (lectotype: Fl. Flum. Icon. 7:
tab 85. 1827[1831].
Centrolobium minus C. Presl., Bot. Bemerk.: 62. 1844.—TYPE:
BRAZIL. Rio de Janeiro: “Habitat ad Rio de Janeiro
Brasiliae”, “prope urbem in valle Larangeiras”, April
1834 [seen on BR specimen], Luschnath s.n. [labelled
Martii Herb. Fl. Bras. n. 160 on specimen in K] (holotype:
PR? n.v.; isotypes: BR!, K!, MO!).
Centrolobium minus C. Presl. var. longo-stipitata N. F. Mattos,
Loefgrenia 78: 6. 1983.—TYPE: BRAZIL. Rio de Janeiro:
“Habitat in Estado do Rio de Janeiro, Jardim Botânico”,
Nov. 1969, Mattos 494 (holotype: RB n.v.).
Tree (7-)15–30 m tall; diameter unknown; bark lightly fis-
sured; twigs and inflorescences drying blackish brown,
glabrate. Leaves (13-)15–19(-23)-foliolate; stipules cadu-
cous, not observed; leaflets 4–12(-18) × 1.5–6(-9) cm, elliptic
or oblong, the base usually asymmetric, cordate, or rounded,
the apex cuspidate, coriaceous, both the surfaces glabrous or
with sparse curly trichomes, the lower surface with dense ses-
sile peltate glands. Inflorescence axes, peduncles and calyces
glabrate or with sparse rusty brown curly trichomes; bracts
3.5–6.5 × 2.5–4 mm, ovate, persistent, reflexed, the apex
acuminate; bracteoles 2.8–3.1 × 0.6–0.9 mm, ovate, the apex
acute, persistent. Flowers 17–18 mm long; pedicel 6–11 mm
long; calyx tubular, the tube ca. 8 mm long, the base obtuse,
the three abaxial lobes 5–7 mm long, the apices obtuse, the
two fused adaxial lobes ca. 7 mm long, the apex emarginate,
356 SYSTEMATIC BOTANY [Volume 34
obtuse; corolla dark yellow to orange suffused with dark red,
the venation red, the standard petal 15–16 × 11–12 mm, orbic-
ular, reflexed by anthesis, with central bright red to purple
blotch, the wing petals 15–16 × 5–6 mm, blade ovate, auricu-
late at the base of the upper margins and with the outer sur-
face covered in sculpturing, the keel petals 15–16 × 5–6 mm,
blade boomerang-shaped, auriculate at the base of the upper
margins, fused for 1/3 of the lower margins; stamens 12.5–
13 mm long, fused basally for ¼ to ½ of their length. Samara
(10-)18–26 × 5–8 cm; seed chamber 3–3.5 cm in diameter, the
spines 15–35 × 0.5–1 mm, the basal part densely indumented
with curly trichomes ca. 1 mm long and short-stalked peltate
glands, the apical part shiny rusty brown, glabrous; wing
chartaceous with indumentum and glands as those of the
seed chamber; stylar spine fused to the wing for 20–30 mm,
free apex ca. 10 mm long, at a 45° angle with the wing margin;
stipe 15–20 mm long; seeds not observed. Figures 5D , Q-R;
6G-I.
Distribution— Centrolobium robustum occurs in the states of
Bahia, Rio de Janeiro and São Paulo in Brazil, from 0–750 m.
Figure 1 .
Habitat and Phenology— The species occurs in coastal rain
forest and in coastal scrub. Flowering specimens were col-
lected in November, fruiting in November, March, July, and
August.
Uses and Vernacular Names— “Araribá”, “araribá amarelo”,
“iriribá” and “erarobá” (Rio de Janeiro and São Paulo; Lima
1985 ), “lei nova”, “putumuju” (Brazil in general; Rudd 1954 ).
Conservation Status— CR A2cd ~ critically endangered
because of a population size reduction of 80% over the last
10 yr due to a decline in area of occupancy, extent of occur-
rence and/or quality of habitat, and due to actual and poten-
tial levels of exploitation.
Representative Specimens Examined— BRAZIL. Bahia: Itabuna, 10
Jul 1964, (fr.) Silva 58331 (NY, US). Rio de Janeiro: Estrada Dona Castorina,
18 Nov. 1985, Lima et al. 2622 (K). São Paulo: 4 Nov 1968, Koortfor s.n.
(NY).
Notes— Presl (1844) distinguished Centrolobium robustum
from C. minus by its larger samaras. Lima (1985) placed, how-
ever, C. minus in synonymy under C. robustum , arguing that
the type specimen of C. minus represents a specimen with
immature fruits.
The differences between C. robustum and the most similar
species C. microchaete are described in the notes of the latter
species.
Centrolobium robustum is distinguished from C. tomentosum
by its glabrous to glabrescent leaves and inflorescences (vs.
densely indumented, also at maturity), the ovate, relatively
small bracts and bracteoles (vs. narrowly ovate), and the
samara with a stylar spine at a 45° angle with the wing margin
(vs. at a 90° angle). The species differs from C. sclerophyllum by
the relatively large membranaceous to coriaceous leaflets (vs.
small rigid-coriaceous leaflets in C. sclerophyllum ), the ovate
bracts and bracteoles (vs. deltate), and the (10-)18–26 cm long,
indumented samara (vs. 10–12 cm long, glabrescent).
It has not been possible to locate the specimen cited as
the type of C. minus var. longo-stipitata Mattos (= Mattos 494 ).
The diagnosis of C. minus var. longo-stipitata specifies, how-
ever, the fruiting stipe length to 1.8–2.5 cm which, along with
the distribution, fits well with the range as below for C. robus-
tum : 1.5–2.0 cm. The only other diagnostic feature of C. minus
var. longo-stipitata is a leaf character which is too variable to
be reliable.
5. C entrolobium sclerophyllum Lima, Arq. Jard. Bot.
Rio de Janeiro 27: 182. 1985.—TYPE: BRAZIL. Espírito
Santo: Reserva Florestal da C.V.R.D., Linhares, próximo
a estrada 161, talhão 604, 20/11/1973, J. Spada 205 (holo-
type: RB !, isotypes: COL n.v., CVRD n.v., RBR n.v.).
Tree 6–30 m tall; trunk diameter 15–70 cm, the bark lightly
fissured; twigs and inflorescences drying reddish brown.
Leaves (11-)13–17(-19)-foliolate; stipules caducous, not
observed; leaflets (3-)5–9(-12) × (1.5-)2–4(-5) cm, ovate, the
base asymmetric or symmetric, cordate or rounded, the apex
long-acuminate, coriaceous, the upper surface glabrous, dark
green, the lower surface rusty brown, glabrous and with
dense short-stalked peltate glands. Inflorescence axes, pedun-
cles and calyces matted with rusty brown curly trichomes;
bracts 3–5 × 1.5–3.5 mm, deltate, persistent, reflexed, the
apex obtuse; bracteoles 1.5–3 × 0.6–1.2 mm, ovate, the apex
obtuse, persistent. Flowers 16–17 mm long; pedicel 1–2 mm
long; calyx urceolate, the tube 6–6.5 mm long, the base trun-
cate, the three abaxial lobes 3–4 mm long, the apices acute,
the two fused adaxial lobes 2–3 mm long, the apex emargin-
ate, acute; corolla yellow, the venation deep red, the standard
petal 15–16 × 11–12 mm, the wing petals 14–15 × 5–6 mm, the
blade oblong, auriculate at the base of the upper margins and
with the outer surface covered in sculpturing, the keel pet-
als 14–15 × 5–6 mm, the blade half-moon-shaped, auriculate
at the base of the upper margins, loosely adnate for 1/3 of
the lower margins; stamens 15–16 mm long, fused basally
for ½ of their length. Samara 10–12 × 3.5–4 cm; seed chamber
1.5–3 cm in diameter, the spines 7–16 × 0.5–1 mm, the basal
with sparse indumentum of curly trichomes ca. 1 mm long
and short-stalked peltate glands, the apical part shiny rusty
brown, glabrate; wing chartaceous, glabrate or with sparse
0.2 mm long, straight trichomes; stylar spine fused to samara
wing for 8–12 mm, the free apex 3–4 mm long, at a 45–90°
angle with the wing margin; stipe 20–30 mm long; seeds not
observed. Figures 5E , N-P; 6J-L.
Distribution— The species is found in the ‘Matas de tab-
uleiro’ in southern Bahia and northern Espirito Santo. A dis-
junct population occurs in the semiarid caatinga formations
of western Bahia and northern Minas Gerais, at altitudes from
400–1,000 m. Figure 1 .
Habitat and Phenology— This little-known species occurs
in SDTF. Flowering specimens were collected in February and
April, fruiting in February and July.
Uses and Vernacular Names— “Banha-de-galinha”, “pau de
sangue” and “putumujú pequeno” (Bahia) and “araribá”,
“araribá rosa” and “lei rosa” or “lei nova” (Espirito Santo).
Conservation Status— CR A2cd ~ critically endangered
because of a population size reduction of 80% over the last
10 yr due to a decline in area of occupancy, extent of occur-
rence and/or quality of habitat, and due to actual and poten-
tial levels of exploitation.
Representative Specimens Examined— BRAZIL. Bahia: Bom Jesus de
Lapa, 2 Jul 1983, Coradin et al. 6358 (K, NY). Espirito Santo: 23 Aug 1981,
(fr.) Lima 1635 (K). Minas Gerais: Itaobim, 3 Mar 1982 (fl.), Rizzini & Mattos
Filho 1557 (RB). Rio de Janeiro: Quissamá, Faz. São Miquel, 14 Feb 2007
(fl.), Lima et al. 6510 (RB).
Notes— Centrolobium sclerophyllum is readily distinguished
from the other species occurring in eastern Brazil by the urce-
olate calyx and rigid coriaceous leaflets.
6. C entrolobium tomentosum Guillem ex Benth., Hook.
Journ. Bot. 2: 66.1840.—TYPE: BRAZIL. Distrito Federal:
“Brasilia, Cachoeira dos Campos”, P. Claussen & B.
2009] PIRIE ET AL.: CENTROLOBIUM 357
Delessert 879 (lectotype: K!; isolectotypes: F!, GH!, K! (2
sheets), NY!).
Tree 10–35 m tall, trunk diameter 10–60 cm, the bark
smooth; twigs and inflorescences light orange brown. Leaves
(11-)13–17(-19)-foliolate; stipules ca. 15 × 4 mm, narrowly
ovate, caducous; leaflets 7–16(-18) × 4–8(-10) cm, oblong
or elliptic, the base asymmetric or symmetric, cordate or
rounded, the apex obtuse, coriaceous or membranaceous, the
upper surface with sparse, curly trichomes, the lower mat-
ted with golden curly trichomes and dense peltate glands.
Inflorescence axes, peduncles and calyces matted with dark
brown curly trichomes; bracts 4–5 × 3–4 mm, narrowly ovate,
caducous, upright, the apex acute; bracteoles 5–7 × 2–3 mm,
oblong, the apex acute, persistent. Flowers 18–22 mm long;
pedicel 3–5 mm long; calyx campanulate, the tube ca. 8–9
mm long, the base obtuse, the three abaxial lobes 5–8 mm
long, the apices obtuse, the two fused adaxial lobes 5–7 mm
long, the apex emarginated, obtuse; corolla yellow suffused
by red, the venation dark red, the standard petal 15–16 × ca.
14 mm, reflexed by anthesis, the wing petals 14–15 × 5–6 mm,
blade oblong, auriculate at the base of the upper margins
and with the outer surface covered in sculpturing, the keel
petals 14–15 × 5–6 mm, blade half-moon-shaped, auriculate
at the base of the upper margins, loosely adnate for 1/3 of the
lower margins; stamens 15–16 mm long, fused basally for ½
of their length. Samara 13–18(-23) × 6–8(-9) cm; seed chamber
3–5 cm in diameter, the spines 9–30 × 1–1.5 mm, the basal and
apical parts moderately indumented with curly trichomes
ca. 1 mm long and long-stalked peltate glands; wing charta-
ceous or coriaceous, with indumentum and glands as those
of the seed chamber; stylar spine fused to samara wing for
40–50 mm, the free apex ca. 20 mm long, at a 90° angle with
the wing margin; stipe 10–15 mm long; seeds ca. 20 × 10 mm,
oblong, smooth, the testa glossy rusty brown. Figures 5F , S-T;
6S-U.
Distribution— Centrolobium tomentosum has a wide geo-
graphic range in central and eastern Brazil from São Paulo
to Bahia and extending into the interior of Goiás, from 0–750
m. Figure 1 .
Habitat and Phenology— It occurs naturally in savanna and
associated gallery forest, in varzea (floodplain) forest, and in
coastal rain forest mixed with restinga; and is cultivated and
invasive in secondary forests. The species is deciduous, and
flowering was recorded in January to March, while fruit set
takes place from August to November.
Uses and Vernacular Names— The species has beautiful
wood that is used in boat and canoe construction, and for
furniture and carriages ( Lorenzi 1992 ). “Araribá”, “araribá-
rosa”, “aribá”, “araruva”, “ararauba”, “carijó”, “iriribá-rosa”,
“putumuju” (Bahia), “tipiri” (Minas Gerais).
Conservation Status— VU A3cd ~ vulnerable because of a
population size reduction of 30% over the next 10 yr due to
a decline in area of occupancy, extent of occurrence and/or
quality of habitat, and due to actual and potential levels of
exploitation.
Representative Specimens Examined— BRAZIL. Bahia: Ilheus,
Centro de Pesquisas do Cacau (CEPLAC), Pennington & Carvalho 293
(E, K). Distrito Federal: Campus de IBDF, 28 Jul 1989, França & Melo 20331
(NY). Espirito Santo: Linhares, 13 Nov 1993, Klitgaard et al. 27 (AAU, K,
RB). Gioás: Corrumba de Goiás, 25 Jan 1968, Irwin et al. 19160 (K, NY).
Minas Gerais: 1840, Claussen 879 (BM, F, GH, K, NY). Paraná: Arapoti, 28
Jan 1989, Hatchbach 52799 (NY). Rio de Janeiro: Rio de Janeiro, 13 Aug
1984, Oliviera 43 (NY). São Paulo: Bofetem, 9 Apr 1971, Gottsberger &
Gottsberger 9471 (NY).
Notes— For differences between C. tomentosum and C. micro -
chaete and C. robustum , see the notes under the latter two spe-
cies. Centrolobium tomentosum does, however, bear closest
resemblance to C. paraense . It is best distinguished by the
narrowly ovate, caducous stipules, bracts and bracteoles vs.
the longer persisting leafy stipules and the broadly ovate
bracts and bracteoles of C. paraense and the large flower size
(18–22 mm vs. 12–19 in C. paraense ).
7. C entrolobium yavizanum Pittier, J. Wash. Acad. Sci. 5: 439.
1915.—TYPE: PANAMA. Darién: Yaviza, 22 Apr 1914,
H. Pittier 6572 (holotype: US!; isotypes: BM!, GH n.v., NY!).
Tree 8–25 m tall, with buttresses up to 1.8 m; trunk dia-
meter 30–40 cm; twigs and inflorescences drying black-
ish brown. Leaves 13–17-foliolate; stipules caducous, not
observed; leaflets 6.5–12 × 4–5.5 cm, elliptic, chartaceous or
membranaceous, the base asymmetric or symmetric, cor-
date or rounded, the apex acuminate, both the surfaces with
sparse straight trichomes and lower surface rusty brown,
covered in short-stalked glands. Inflorescence axes, pedun-
cles and calyces matted with dark brown curly trichomes;
bracts 3–5 × 1.5–3 mm, narrowly ovate, persistent, reflexed,
the apex acute; bracteoles 1.5–2.5 × 1–1.5 mm, narrowly ovate,
the apex acute, persistent. Flowers 10–12 mm long; pedicel
3–7 mm long; calyx tubular, the tube 5–6 mm long, the base
obtuse, the three abaxial lobes 2–5 mm long, the apices acute,
the two fused adaxial lobes ca. 5 mm long, the apex emar-
ginated, acute; corolla yellow/orange with central portion of
standard red, the standard petal 11–12 × 9–10 mm, the wing
petals ca. 10 × 3.5 mm, the keel petals ca. 12 × 4 mm; stamens
10–11 mm long. Samara 7.2–16 × 3.4–5.8 cm; seed chamber
ca. 3 cm in dia meter, the spines 10–20 × 0.5–1 mm, the basal
part sparsely indumented with curly trichomes ca. 1 mm long
and long-stalked peltate glands, the apical part rusty brown,
shiny, glabrate; wing chartaceous, glabrate, with sparse long-
stalked peltate glands; stylar spine fused to samara wing
for 3–5 mm long, the free apex 6–14 mm long, at a 45–90°
angle with the wing margin; stipe 20–30 mm long; seeds not
observed. Figures 5G , U-V; 6M-O.
Distribution— The species occurs in Darién province of
Panama, and Bolivar, Santander, Antioquia and Boyacá
departments in Colombia, at sea level. Figure 1 .
Habitat and Phenology— Centrolobium yavizanum has been
collected in remnant primary SDTF and in disturbed areas
such as roadside vegetation and as trees remaining in pas-
tures. Flowering usually occurs from March to June, and fruit-
ing from March to July.
Uses and Venacular Names— “Guayacán jobo” (Colombia).
Conservation Status— The species has been designated
‘vulnerable’ on the IUCN red list of threatened species (Mitré
1997), due to threats of habitat loss through logging, clear cut-
ting and human settlement.
Representative Specimens Examined— COLOMBIA. Antioquia: 23 Mar
1987, Zarucchi et al. 4957 (K, MO, NY, US). Bolivar: Loba, Amargamiento
Rico, Apr 1916, Curran 483 (GH). Boyacá: Guaguaqui, 14 Jul 1917, Whitford
& Pinzon 16A (A). Choco: Léon 550 (MO). Santander: Puerto Berrio, 30 Apr
1935, Haught 1688 (NY).
PANAMA. Darién: Pico Pedenjo, 15 Mar 1968, Duke 15424 (MO). Panama:
El Llano, 1 Sep 1971, Gentry & Tyson 1731 (MO).
Notes— Centrolobium yavizanum is most similar to C. micro-
chaete (occurring in Bolivia and southern Brazil), from which
it cannot readily be distinguished in flower, though the bracts
are generally larger and ovate (vs. smaller and narrowly
ovate) and the calyx lobes acuminate rather than acute. In
358 SYSTEMATIC BOTANY [Volume 34
fruit, it can be distinguished by the greater length of the stipe
(20–30 mm vs . 10–15 mm).
Acknowledgments. MDP thanks SYNTHESYS for financial support
to visit the Natural History Museum in London and the Royal Botanic
Gardens, Kew. DNA sequencing was carried out in the labs of the Royal
Botanic Garden, Edinburgh and Utrecht University, The Netherlands. The
authors are grateful to Haroldo C. de Lima for providing the study with
Centrolobium leaf samples, to Will Goodall Copestake for sequences, and
we thank Margaret Tebbs for the beautiful line drawings. The construc-
tive and detailed comments of two anonymous reviewers are gratefully
acknowledged.
Literature Cited
Acosta-Solis , M. 1947 . Commercial possibilities of the forests of Ecuador–
mainly Esmeraldas province . Tropical Woods 89 : 1 – 47 .
Bentham , G. 1839 . Centrolobium Mart. ex Benth . Annalen des Wiener Museums
der Naturgeschichte 2 : 95 .
Bentham , G. 1862 . Centrolobium . Pp. 263 – 264 in Flora Brasiliensis 15 part
1 fasc. 29, ed. C. F. P. v. Martius , A. W. Eichler , and I. Urban . Munich :
R. Oldenbourg .
Beyra , M. A. and M. Lavin . 1999 . Monograph of Pictetia (Leguminosae -
Papillionoideae) and a review of the Aeschynomeneae . Systematic
Botany Monographs 56 : 1 – 93 .
Burnham , R. J. 1995 . A new species of winged fruit from the Miocene
of Ecuador: Tipuana ecuatoriana (Leguminosae) . American Journal of
Botany 82 : 1599 – 1607 .
Doyle , J. J. and J. L. Doyle . 1987 . A rapid DNA isolation procedure for
small quantities of fresh leaf tissue . Phytochemical Bulletin 19 : 11 – 15 .
Drummond , A. J. , S. Y. W. Ho , M. J. Phillips , and A. Rambaut . 2006 .
Relaxed phylogenetics and dating with confidence . Public Library of
Science. Biology 4 : e88 . doi: 10.1371/journal.pbio.0040088
Drummond, A. J. and A. Rambaut . 2007 . BEAST: Bayesian evolution-
ary analysis by sampling trees . BMC Evolutionary Biology 7 : 214 .
doi:10.1186/1471-2148-7-214
Felsenstein, J. 1981 . Evolutionary trees from DNA sequences: A maximum
likelihood approach . Journal of Molecular Evolution 17 : 368 – 376 .
Fitch, W. M. 1971 . Toward defining the course of evolution: minimum
change for a specified tree topology . Systematic Zoology 20 : 406 – 416 .
Huelsenbeck, J. P. and F. Ronquist . 2001 . MrBayes: Bayesian inference of
phylogenetic trees . Bioinformatics 17 : 754 – 755 .
IUCN . 2001 . IIRL Categories . Criteria : Version 3.1. IUCN Species Survival
Commission .
Lavin , M. 2006 . Floristic and geographical stability of discontinuous sea-
sonally dry tropical forests explains patterns of plant phylogeny and
endemism . Pp. 433 – 447 in Neotropical savannas and seasonally dry for-
ests . eds. R. T. Pennington , G. Lewis , and J. Ratter . London : Taylor
and Francis .
Lavin, M. , P. S. Herendeen , and M. F. Wojciechowski . 2005 . Evolutionary
rates analysis of Leguminosae implicates a rapid diversification of
lineages during the Tertiary . Systematic Biology 54 : 575 – 594 .
Lavin, M. , R. T. Pennington , B. B. Klitgaard , J. I. Sprent , H. C. De Lima ,
and P. E. Gasson . 2001 . The dalbergioid legumes (Fabaceae): delimita-
tion of a pantropical monophyletic clade . American Journal of Botany
88 : 503 – 533 .
Lavin , M. , M. Thulin , J.-N. Labat , and R. T. Pennington . 2000 . Africa,
the odd man out: molecular biogeography of Dalbergioid Legumes
(Fabaceae) suggests otherwise . Systematic Botany 25 : 449 – 467 .
Lima, H. C. D. 1985 . Centrolobium Martius ex Bentham (Leguminosae -
Papilionoideae) estudo taxonômico das espécies Brasilieras extra
Amazônicas . Arquivos do Jardim Botânicodo Rio de Janeiro 27 :
177 – 191 .
Little Jr., E. L. and R. G. Dixon . 1969 . Árboles comunes de la Provincia de
Esmeraldas. i - xii . Rome : FAO .
Lorenzi , H. 1992 . Árvores Brasileiras vol. 1. Manual de indentificação e cul-
tivo de plantas arbóreas nativas do Brasil . Nova Odessa, Brazil : Editora
Plantarum .
Lorenzi, H. 2002 . Brazilian Trees 2. A guide to the identification and cultivation
of Brazilian native trees . Nova Odessa, Brazil : Editora Plantarum .
Mitré, M. 1997 . Centrolobium yavizanum . Pp. 351 in IUCN Red List of
Threatened Species .
Neill , D. 2000 . Centrolobium . Pp. 97 in Libro rojo de las plantas endémicas
del Ecuador eds. R. Valencia , N. Pitman , S. León-Yánez , and P. M.
Jørgensen . Quito : Publicaciones del Herbario QCA, Pontificia
Universidad Católica del Ecuador .
Pennington, R. T. , M. Lavin , D. E. Prado , C. A. Pendry , S. K. Pell , and
C. A. Butterworth . 2004 . Historical climate change and specia-
tion: neotropical seasonally dry forest plants show patterns of both
Tertiary and Quaternary diversification . Philosophical Transactions of
the Royal Society B. Biological Sciences 359 : 515 – 538 .
Pennington, R. T. , G. Lewis , and J. A. Ratter . 2006 . An overview of the
plant diversity, biogeography and conservation of neotropical savan-
nas and seasonally dry forests . Pp. 1 29 in Neotropical savannas and
seasonally dry forests , eds. R. T. Pennington , G. Lewis , and J. Ratter .
London : Taylor and Francis .
Pennington, R. T. , D. E. Prado , and C. A. Pendry . 2000 . Neotropical sea-
sonally dry forests and Quaternary vegetation changes . Journal of
Biogeography 27 : 261 – 273 .
Peralvo, M. , R. Sierra , K. R. Young , and C. Ulloa-Ulloa . 2007 . Identification
of biodiversity conservation priorities using predictive modeling: an
application for the equatorial region of South America . Biodiversity
and Conservation 16 : 2649 – 2675 .
Pittier, H. 1944 . Centrolobium orinocense (Benth.) Pittier . Boletin Tecnico.
Ministerio de Agricultura y Cria. Caracas 5 : 123 .
Polhill , R. M. 1981 . Dalbergieae . Pp. 233 – 242 in Advances in legume system-
atics, part 1 , eds. R. M. Polhill and P. H. Raven . Richmond, Surrey :
Royal Botanic Gardens, Kew .
Polhill, R. M. 1994 . Classification of the Leguminosae . Pp. 35 – 48 in
Phytochemical dictionary of the Leguminosae 1 , eds. F. A. Bisby ,
J. Buckingham , and J. B. Harbourne . London : Chapman and Hall .
Posada , D. and K. A. Crandall . 1998 . Modeltest: testing the model of DNA
substitution . Bioinformatics 14 : 817 – 818 .
Prado , D. E. and P. E. Gibbs . 1993 . Patterns of species distributions in the
dry seasonal forests of South America . Annals of the Missouri Botanical
Garden 80 : 902 – 927 .
Presl, C. 1844 . Botanische Bemerkungen . Abhandlungen der königlichen
Bohm Gesellschaft für Wissenschaft 3 : 62 .
Rambaut , A. and A. J. Drummond . 2003 . Tracer, version 1.3. Available
from http://beast.bio.ed.ac.uk/Tracer .
Rudd, V. E. 1954 . Centrolobium (Leguminosae): Validation of a specific
name and a brief review of the genus . Journal of the Washington
Academy of Sciences 44 : 284 – 288 .
Sanderson, M. J. 2002a . Estimating absolute rates of molecular evolution
and divergence times: a penalized likelihood approach . Molecular
Biology and Evolution 19 : 101 – 109 .
Sanderson , M. J. 2002b . r8s, version 1.50. Distributed by the author,
Section of Evolution and Ecology, University of California , Davis .
Savolainen , V. , P. Cuénoud , R. Spichiger , M. D. P. Martinez , M. Crèvecoeur ,
and J.-F. Manen . 1995 . The use of herbarium specimens in DNA
phylogenetics: evaluation and improvement . Plant Systematics and
Evolution 197 : 87 – 98 .
Schrire , B. D. , G. P. Lewis , and M. Lavin . 2005 . Biogeography of the
Leguminosae . Pp. 21 – 54 in Legumes of the world . eds. G. Lewis,
B. Schrire , B. Mackinder , and M. Lock . Richmond, Surrey : Royal
Botanic Gardens, Kew .
Simmons, M. P. and H. Ochoterena . 2000 . Gaps as characters in sequence-
based phylogenetic analysis . Systematic Biology 49 : 369 – 381 .
Swofford , D. L. 2002 . PAUP*: Phylogenetic Analysis Using Parsimony
(*and other methods). Version 4. Sunderland : Sinauer Associates Inc .
Taberlet, P. , L. Gielly , G. Pautou , and J. Bouvet . 1991 . Universal primers for
amplification of the three non-coding regions of chloroplast DNA .
Plant Molecular Biology 17 : 1105 – 1109 .
Thomas, W. W. and M. R. de Vansconcellos Barbosa . 2008 . Natural veg-
etation types in the Atlantic coastal forest of northeastern Brazil .
Pp. 6 – 20 in The Atlantic Coastal Forest of northeastern Brazil , ed. W.W.
Thomas . Memoirs of the New York Botanical Garden vol 100 . Bronx :
New York Botanical Garden .
Appendix 1. Names, vouchers and Genbank accessions for DNA sam-
ples used for molecular phylogenetic analyses. A code for Centrolobium
samples (following the voucher) is used in the text and figures to discern
samples of the same species. GenBank accession numbers are in the fol-
lowing sequence: trnL - F, matK, ITS, replaced with — where not sampled.
Centrolobium microchaete, (Mart. ex Benth.) Lima, Brazil, Paraná,
G. Hatchbach 46089 (F, NY), 769, -, EU401407, -; Centrolobium microchaete
(Mart. ex Benth.) Lima, Brazil, Minas Gerais, B. B. Klitgaard 17 (AAU, K),
E6, EU401419, EU401408, EU401431; Centrolobium microchaete (Mart. ex
Benth.) Lima, Bolivia, Santa Cruz, R. T. Pennington 898 (E), E1, EU401420,
EU401409, EU401432; Centrolobium ochroxylum Rose ex Rudd, Ecuador, Los
Ríos, B. B. Klitgaard 658 (AAU, K), E7, EU401421, EU401410, EU401433;
Centrolobium ochroxylum Rose ex Rudd, Ecuador, Loja, B. B. Klitgaard 551
2009] PIRIE ET AL.: CENTROLOBIUM 359
(AAU), E3, EU401422, EU401411, EU401434; Centrolobium ochroxylum Rose
ex Rudd, Ecuador, El Oro, B. B. Klitgaard 421 (AAU, K), E5, -, -, EU401435;
Centrolobium paraense Tul., Brazil, Rio de Janeiro, Botanic Garden (cul-
tivated), H. C. de Lima 5688 (RB), 770, EU401423, EU401412, EU401436;
Centrolobium paraense Tul., Guyana, Rapununi, Jansen-Jacobs 4192 (U),
1278, EU401424, EU401413, EU401437; Centrolobium robustum (Vell.) Mart.
ex Benth., Brazil, Rio de Janeiro, Botanic Garden (cultivated), H. C. de Lima
2506 (RB), 771, EU401425, EU401414, EU401438; Centrolobium sclerophyl-
lum Lima, Brazil, Bahia, Coradin 6358 (K), E24, -, EU401415, EU401439;
Centrolobium sclerophyllum Lima, Brazil, Rio de Janeiro, Botanic Garden
(cultivated), H. C. de Lima 1635 (K), 772, EU401426, EU401416, EU401440;
Centrolobium tomentosum Guill. ex Benth., Brazil, Rio de Janeiro, Botanic
Garden (cultivated), H. C. de Lima 2518 (RB), 773, -, EU401417, -; Centrolobium
tomentosum Guill. ex Benth., Brazil, Gioás, B. M. T. Walter 3343 (NY), E33,
EU401427, EU401418, EU401441; Centrolobium yavizanum Pittier, Panama,
Darien, Stern 731 (NY, MO, GH), 775, EU401428, -, EU401442; Geoffroea
decorticans (Gillies ex Hook. & Arn.) Burkhart, USA, Arizona (seed source),
M. Lavin 750 (MONT), -, AF270880, AF189057; Geoffroea decorticans (Gillies
ex Hook. & Arn.) Burkhart, Chile, Gardner & Nees 5823 (E), AF208962, -, -;
Grazielodendron riodocense Lima, Brazil, Rio de Janeiro, Botanic Garden
(cultivated), H. C. de Lima s.n. (E), AF208952, AF270862, -; Paramachaerium
schomburgkii (Benth.) Ducke, Guyana, Kanuku Mts., Jansen-Jacobs 97 (K),
AF208959, AF272062, AF204237; Platymiscium dimorphandrum Donn. Sm.,
Mexico, J. I. Calzada 14786 (MEXU), -, -, EU401430; Platymiscium floribun-
dum Vogel, Brazil, Rio de Janeiro, B. B. Klitgaard 84 (K), -, -, EU401429;
Platymiscium sp., Colombia, Antioquia, R. T. Pennington 692 (E), AF208955,
AF270871, -; Platymiscium stipulare Benth., Ecuador, Napo, R. T. Pennington
649 (E), -, AF270872, -; Platypodium elegans Vogel, Colombia, Antioquia,
R. T. Pennington 688 (E), -, AF270877, -;
Platypodium elegans Vogel, Brazil,
Gioás, R. T. Pennington 488 (E), AF208961, -, -; Pterocarpus indicus Willd.,
Phillipines, Luzon, R. T. Pennington 718 (E), AF208953, -, -; Pterocarpus indi-
cus Willd., unknown locality, Henderson s.n. (NY), -, AF142691, -; Pterocarpus
macrocarpus Kurz, Puerto Rico, Fajardo (cultivated), M. Lavin 721 (MONT), -,
AF203588, AF269176; Ramorina girolae Speg., Argentina, Royal Botanic
Gardens, Kew (cultivated), (no voucher), AF208957, AF270881, AF204236;
Tipuana tipu (Benth.) Kuntze, Argentina, Salta, M. Lavin 5796 (MONT), -,
AF270882, AF189056; Tipuana tipu (Benth.) Kuntze, Spain, Barcelona (cul-
tivated), R. T. Pennington s.n. (E), AF208956, -, -.
Appendix 2. List of exsiccatae Abbott & Jardim 16395 (1); Abbott
16742 (1), 17279 (1); Aguilar et al. 680 (1); Aristeguieta 4710 (3), 7045 (3), 7155
(3); Asplund 15387 (2); Barreto 1509 (1); Beck 17203 (1); Belém et al. 954 (1);
Belém & Pinheiro 2219 (1), 3392 (6); Bernardi 7386 (3); Blanco 381 (3); Brant
1266 (7); Briceño & Rosales 142 (3); Bunting 17 (3), 230 (3), 5196 (3), 5211 (3),
5772 (3), 5773 (3) 7643 (3); Camara et al. s.n. (6); Camp 585 (2); Carabot et al.
4383 (3); Claussen s.n. (6); Coradin et al. 6342 (5), 6358 (5), 8587 (5); Corcovado
15921 (4); Cornejo & Bonifaz 2341 (2); Curran 483 (7); Daly 124 (2); Davis 786
(3); Delgado 105 (2); Diaz et al. 4695 (2), 4878 (2), 5054 (2), 5410 (2), 5787 (2),
6204 (2), 6289 (2), 6432 (2), 6467 (2), 6606 (2), 6714 (2); Dodson & Clendenin
11030 (2); Dodson & Thien 1293 (2); Dodson et al. 6921 (2), 8646 (2), 9651 (2),
14551 (2); Ducke 332 (3), 516 (3), 35506 (3); Dugand 1126 (3); Duke 15424 (7);
Elias 1580 (3); Fairchild 2814 (3); Forestry Department of British Guyana 138
(3), 2160 (3), 2181 (3); Foster 9940 (3); França & Melo 20331 (6); Froes 19962
(1), 20171 (5), 23055 (3); Ganev 3462 (6); Gentry & Berry 15032 (3); Gentry
& Dodson 54849 (2); Gentry & Tyson 1731 (7); Gentry 8646 (2), 10306 (3);
Glaziou 2539 (6), 5715 (6), 15922 (6); Gottsberger & Gottsberger 9471 (6); Hage
1646 (6); Harris et al. 1088 (3); Hartshorn et al. 2763 (2); Hatschbach 15719 (6),
16355 (1), 16427 (6), 46038 (1), 46089 (1), 46250 (1), 52799 (6); Hatschbach
et al. 64144 (6);
Hatschbach & Zelma 49536 (6); Haught 1688 (7); Helmreishen 89
(6); Herbarium Guillem 592 (6); Hoffman 1048 (3); Hohenkerk 849 (3); Hopkins
et al. 521 (3); Irwin et al. 12022 (6); 14033 (6); 15685 (6); 19160 (6); 31238 (5);
Ivker s.n. (6); Jansen-Jacobs et al. 4012 (3), 4192 (3); Jardim 1370 (1), 1878 (1);
Johansen 12 (2); Kinnup et al. 1004 (6); Kirkbride et al. 3581 (6); Klitgaard et al.
17 (1), 27 (6), 319 (2), 421 (2), 551 (2), 658 (2), 67008 (2), 99461 (2), 99487 (2);
Koortfor s.n. (4); Koscinski & Pickel 4995 (6); Krukoff 10625 (1); Kuhlmann 3225
(3); 31626 (6); Kuniyoshi & Pizani 4644 (1); Lachman et al. 62 (7); Léon 550 (7);
Lewis 1420 (3), 1697 (3), (2); Liesner & Gonzales 5419 (3), 5748 (3), 6000 (3),
11077 (3); Liesner et al. 12654 (3); H. C. Lima 1635 (5); H. C. Lima et al. 2622
(4), 3986 (1); J. Lima 790 (3); Little 6496 (2), 6557 (2), 6615 (2); Little & Dixon
21175 (2); Louman & Oliviera 8701 (1); Madison et al. 3271 (2), 5232 (2) 5271
(2); Mamani & Rodriguez 515 (1); Mamani & Saucedo 616 (1), 819 (1), 875 (1);
Martius 160 (4); Matthew 72812 (2), 71882 (2); Mattos et al. 333 (5); Mecenas &
Cardoso 87 (6); Medri et al. 578 (6); Menacho & Jiminez 782 (1); Mendonca et al.
136 (1); Meneces & Hartshorn 2365 (1), 2375 (1); Mori et al. 9393 (6); Neill 6192
(2); Neill & Nuñez 10472 (2); Neill et al. 10527 (2), 11644 (2); Occhioni 16363
(5); Oliveira 43 (6); Palacios et al. 818 (2); Peckolt s.n. or 260 (~Martius 160) (1);
Pedra do Cavalo 653 (1); Pendry 674 (1); R. T. Pennington & Carvalho 293 (6); R.
T. Pennington 898 (1), 916 (1); T. D. Pennington et al. 15012 (2); Pereira 11718
(6); Peter 49720 (2); Pirie & Chatrou 1 (1); Pittier 6572 (7), 6611 (3), 9736 (3),
10514 (3), 11358 (3), 12240 (3); Prance et al. 9562 (3); Quevedo & Centurión 360
(1), 504 (1), 609 (1); Raimundo 1120 (1); Ramirez 2472 (3); Ratter et al. 5489 (3);
Reidel 464 (4); Reidel & Luschnath 448 (4); Reitz 2931 (1); Reitz & Klein 1877
(1); Ribiero 283 (3); Rosa 3119 (3); Rosales et al. 658 (3); Rose & Rose 23370 (2);
Ruiz 365 (3); Saldías 377 (1), 688 (1); Samaniego et al. 35 (2); F. S. Santos 3 (6),
4 (5); T. S. Santos, 806 (1), 1899 (1); Schomburgk 314 (3), 590 (3); Seidel & Beck
240 (1); Sello s.n. (6); Silva 58331 (4); Silva et al. 2827 (1); Simpson & Schunke
522 (2); A. C. Smith 3207 (3); D. N. Smith et al. 14429 (1); F. D. Smith 280 (3);
Souza 2 (6); Stern et al. 731 (7), 761 (7); Steyermark 54016 (2), 61487 (3), 86772
(3), 122979 (3), 123053 (3); Steyermark & Fernández 99547 (3); Steyermark
et al. 107725 (3); Tamayo 2106 (3), 3524 (3); Thomas et al. 11906 (6); Toledo
et al. 510 (1); Valverde 436 (2); I. G. Vargas C. 2450 (1), 5096 (1); I. G. Vargas C. &
Contreras 5096 (1); I. G. Vargas C. et al. 2450 (1); J. A. Vargas 16 (2); Walter
et al. 3343 (6); Whitford 10 (3); Whitford & Pinzon 16A (7); Williams 11646 (3),
11832 (3); Wilson-Browne 138 (3); Zarucchi et al. 4957 (7).
... Centrolobium (Papilionoideae) se caracteriza pela presença de glândulas peltadas alaranjadas nos ramos, folhas, flores e frutos, além das folhas imparipinadas, flores papilionáceas, corola amarela, 10 estames unidos em bainha e fruto sâmara. É um gênero Neotropical e possui sete espécies distribuídas na América do Sul (Pirie et al., 2009). No Brasil são registradas cinco espécies distribuídas principalmente na Mata Atlântica (BFG, 2018). ...
... Centrolobium sclerophyllum foi originalmente descrito com ocorrência nas florestas de tabuleiros do Sul da Bahia e Norte do Espírito Santo e disjuntamente em formações semi áridas da Caatinga, no Oeste da Bahia e Norte de Minas Gerais (Lima, 1985;Pirie et al., 2009). Posteriormente foi registrado no estado do Rio de Janeiro principalmente em Floresta Estacional Semidecidual de tabuleiros (BFG, 2018;Silva & Lima, 2014b). ...
... É facilmente reconhecido por possuir glândulas peltadas, curto-estipitadas, de coloração alaranjada ao longo de toda a superfície da planta e pelo fruto do tipo sâmara com núcleo seminífero basal equinado. Distingue-se facilmente das demais espécies do gênero pelas flores com cálice urceolado e também é a única espécie que apresenta os folíolos bulados e rígido-coriáceos (Lima 1985;Pirie et al., 2009). ...
Article
Full-text available
This study presents the taxonomic treatment of tree Leguminosae species recorded in a Seasonal Semideciduous tabuleiros forest at Estação Ecológica Estadual de Guaxindiba, São Francisco de Itabapoana, Rio de Janeiro. In the study area 49 species belonging to 36 genera were inventoried, of which Machaerium Pers (five species) and Inga Mill. (four species) are the most representative. The species Ormosia nitida Vogel is a new record for the state of Rio de Janeiro. Identification key, morphological descriptions, field photos, geographical distribution and habitat data and comments on all treated species are provided.
... The diversification of Pterocarpus appears to have involved multiple biome shifts, largely from dry forest into lowland habitats (namely, savanna and tropical moist forest). A similar pattern has been shown in Centrolobiumand Platymiscium, which are close relatives of Pterocarpus (Klitgård, 2005;Pirie et al., 2009;Saslis-Lagoudakis et al., 2008). ...
Article
Phenotypes promoting dispersal over ecological timescales may have macroevolutionary consequences, such as long‐distance dispersal and diversification. However, whether dispersal traits explain the distribution of pantropical plant groups remains unclear. Here we reconstruct the biogeographical history of a tree clade to assess whether seed dispersal traits and biome‐switching explain the clade’s pantropical distribution. Pantropical. The Pterocarpus clade (Leguminosae/Fabaceae). We sequenced 303 nuclear loci using target capture and generated a time‐calibrated phylogenomic tree. We also generated a corroborative time‐calibrated phylogenetic tree from data‐mined Sanger‐sequencing data. We then collated distribution data and seed dispersal morphology traits to compare trait‐dependent and trait‐independent biogeographical models, allowing us to infer whether dispersal traits influenced Pterocarpus’ spatio‐temporal evolution. Finally, using the results of these model tests, we estimated the ancestral ranges and biomes of Pterocarpus species to better understand their biogeographical history, and assessed the degree and direction of biome‐switching over the course of their diversification. We recovered well‐supported phylogenetic relationships within Pterocarpus, within which there were two subclades – one neotropical and the other palaeotropical. Our divergence date estimates suggested that Pterocarpus diversified from around 12 Ma, during the Miocene. Trait‐dependent biogeographical models were rejected for both range and biome evolution within Pterocarpus, but models including dispersal were supported. Pterocarpus’ ancestral node shared a range across the new‐world and old‐world tropics, followed by divergence into palaeotropical and neotropical clades. Biome‐switching occurred most frequently into rainforest and grassland. Our analyses suggest that Pterocarpus underwent infrequent cross‐continental dispersal and establishment into novel biomes. While this was minimally impacted by seed dispersal traits, biome‐switching following long‐distance dispersal and climate change have played an important role in diversification within Pterocarpus since the Miocene. Indeed, rare events of long‐distance dispersal likely explain the wide distributions of many pantropical plant species.
... Centrolobium is a neotropical genus in which there are seven species found in parts of Brazil, Bolivia, Ecuador, Peru, Colombia, Venezuela, Panama and Guyana, with disseminated species and endemic species more geographically restricted (Pirie et al. 2009). Vasconcelos et al. (2015), purified and characterized a new mannose-binding lectin present in plant seeds of the genus Centrolobium microchaete (CML). ...
Article
This study aims to evaluate the wound healing potential of lectin isolated from the seeds of Centrolobium microchaete (Mart. ex Benth) (CML) on cutaneous wounds in mice. CML did not show cytotoxicity on murine dermal fibroblasts (L929 cell line). The wounds treated with CML (200 μg/mL) showed a decrease in area within 12 days post-operative (P.O.) when compared to control. On 3rd and 7th day P.O., the CML-treated group exhibited fibroblast proliferation and neovascularization. On 12th day P.O., complete restructuring of the epithelial layer and connective tissue was observed in the CML-treated group, whereas control groups exhibited incomplete reepithelialization. CML treatment enhanced the wound closure via the wound contraction process, resulting in the restructuring of the skin layers on 12th day P.O. In conclusion, CML induced a fast and efficient wound healing, suggesting that it can be used as a promising therapeutic tool to heal acute wounds.
... Pods vary greatly in morphology among legume taxa [1], from winged types as in Centrolobium spp. [2], to those buried in the ground in Arachis [3]. Within a species, pod morphology is generally conserved, but as Darwin [4,5] noted: "Any part or organ developed to an extraordinary size or in an extraordinary manner, in comparison with the same part or organ in the allied species, must have gone through an extraordinary amount of modification since the genus arose; and thus we can understand why it should often still be variable in a much higher degree than other parts; for variation is a long-continued and slow process, and natural selection will in such cases not as yet have had time to overcome the tendency to further variability and to reversion to a less modified state". ...
Article
Full-text available
The seed-containing pod is the defining structure of plants in the legume family, yet pods exhibit a wide range of morphological variation. Within a species pod characters are likely to be correlated with reproductive strategy, and within cultivated forms will correspond to aspects of yield determination and/or end use. Here variation in pod size, described as pod length: pod width ratio, has been analyzed in pea germplasm represented by 597 accessions. This pod size variation is discussed with respect to population structure and to known classical pod morphology mutants. Variability of the pod length: width ratio can be explained by allelic variation at two genetic loci that may correspond to organ-specific negative regulators of growth.
... Elle est composée d'une graine et d'une ou deux ailes : ce dernier cas n'est valable que pour la graine d'érable, la plus connue, qui est aussi appelée "disamare" et qui possède donc une forme d'aile asymétrique. La longueur totale des samares est très variable, de 1 cm à 18 cm pour les cas "classiques" [76], même si des spécimens plus grands encore existent notamment en Amérique du sud, jusqu'à 30 cm [77]. Les ailes des samares ont pour fonction de favoriser l'anémochorie : en effet, après leur chute depuis une branche, les graines utilisent les vents latéraux pour se disséminer, et ainsi toucher un périmètre le plus large possible autour de l'arbre, de quelques mètres jusqu'à plusieurs kilomètres [78,79]. ...
Thesis
L'Internet des objets est en pleine croissance, et nécessite le développement de capteurs autonomes communicants. L'autonomie de tels capteurs peut passer par la récupération d'énergie ambiante afin de limiter l'utilisation de batteries. L'objectif de la thèse est d'étendre les possibilités de conversion en développant une micro éolienne aussi petite que possible pour récupérer l'énergie cinétique des vents à faibles vitesses. Une rupture au niveau de la géométrie est nécessaire et se fonde dans ces travaux sur le biomimétisme.Dans cet objectif, une étude du comportement des samares par stéréocorrélation d'images est menée afin de reconstruire leur mouvement en trois dimensions. Dans un premier cas, les samares chutent verticalement ; dans une deuxième configuration elles sont percées à l'emplacement de leur centre de rotation et sont placées sur un axe horizontal dans une soufflerie. Les samares tournent alors jusqu'à 89% plus rapidement qu'en chute libre avec presque 2000 trs/min : elles sont donc un bon modèle pour des pales de micro éolienne efficace.Puis, plusieurs hélices bioinspirées sont conçues et fabriquées par stéréolithographie. Les hélices sont ensuite assemblées avec une génératrice électromagnétique miniature réalisée au CEA de Grenoble. La meilleure d'entre elles est identifiée après une batterie de tests. Par la suite, l'efficacité de la micro éolienne optimale est vérifiée en mesurant les puissances développées pour une large gamme de vents (jusqu'à 8 m/s). La puissance maximale s'étale de 50 µW pour 1,2 m/s à plus de 80 mW à 8 m/s ; le coefficient de puissance et le rendement total optimaux apparaissent à 4 m/s et sont respectivement de 29% et 17,5%. Cela nous place au-dessus des meilleurs prototypes de la littérature. Par conséquent, la validité du choix initial du biomimétisme et de son apport pour les petites dimensions est confirmée
... In C. microchaete, which is widely distributed across Bolivian tropical lowlands (Pirie et al. 2009), the spatial correlations indicated that its growth variability is a reliable proxy of growing-season (November) precipitation across the eastward Andean-Amazonian piedmont at longitudes from 60° to 72° W and latitudes from 12° to 20° S. This agrees with what López et al. (2017López et al. ( , 2019 reported. ...
Chapter
Full-text available
Seasonally, dry tropical forests (SDTFs) are among the most diverse and threatened biomes in America. Several tree species coexist in these forests but their long-term growth responses to climate are unknown, and this is needed to make growth forecasts as a function of climate scenarios. We investigated the responses to climate, drought and ocean-atmosphere patterns of three tree species (Acosmium cardenasii H.S. Irwin & Arroyo, Centrolobium microchaete (Mart. ex Benth.) H.C. de Lima ex G.P. Lewis and Zeyheria tuberculosa (Vell.) Bureau coexisting in a Bolivian dry tropical forest. Species chronologies of ring-width indices were related to temperature, precipitation, drought indices and sea temperatures. A growth model was also used to forecast growth variability. C. microchaete and A. cardenasii presented similar year-to-year growth variability. Cool and wet conditions enhanced growth. Shorter droughts constrained more growth of C. microchaete and A. cardenasii, whilst longer droughts negatively impacted Z. tuberculosa. These different growth responses to climate and drought contribute to explain the coexistence of tree species in SDTFs. The growth patterns of the study species are valuable climate proxies for Bolivia. Forecasted warmer conditions after the 2050s will differently affect the growth variability of these species depending on their responses to climate and drought.
... Yet, it is interesting to note that mean ages for nodes associated to migration and subsequent allopatric speciation are dated around two especially dry and severe glacial periods: the Anglian Kansas glaciation, c. 455,000-300,000 years (nodes 4, 11-13, 20, 22-24) and the LGM (21,000-7,000 years, nodes 5, 32) ( Figure 4). This pattern stands in contrast to that found in other SDTF-centred genera, where species divergence largely predates the Pleistocene, and is therefore inconsistent with a Pleistocene diversification scenario (De-Nova et al., 2012;Pennington et al., 2009;Pirie, Klitgaard, & Pennington, 2009). This does not discard that intraspecies divergence within widespread SDTF taxa could be driven by Quaternary climatic changes (Pennington et al., 2004). ...
Article
Among small rotorcraft, the use of multiple compact rotors in a mechanically simple design leads to impressive agility and maneuverability but inevitably results in high energetic demand and acutely restricted endurance. Small spinning propellers used in these vehicles contrast with large lifting surfaces of winged seeds, which spontaneously gyrate into stable autorotation upon falling. The pronounced aerodynamic surfaces and delayed stalls are believed key to efficient unpowered flight. Here, the bioinspired principles are adopted to notably reduce the power consumption of small aerial vehicles by means of a samara-inspired robot. We report a dual-wing 35.1-gram aircraft capable of hovering flight via powered gyration. Equipped with two rotors, the underactuated robot with oversized revolving wings, designed to leverage unsteady aerodynamics, was optimized for boosted flight efficiency. Through the analysis of flight dynamics and stability, the vehicle was designed for passive attitude stability, eliminating the need for fast feedback to stay upright. To this end, the drone demonstrates flight with a twofold decrease in power consumption when compared with benchmark multirotor robots. Exhibiting the power loading of 8.0 grams per watt, the vehicle recorded a flight time of 14.9 minutes and up to 24.5 minutes when equipped with a larger battery. Taking advantage of the fast revolving motion to overcome the severe underactuation, we also realized position-controlled flight and illustrated examples of mapping and surveillance applications with a 21.5-gram payload.
Article
Full-text available
Resumo Dalbergieae s.l. possui 49 gêneros e 1.324 espécies distribuídas principalmente na região neotropical. No Brasil está representada por 380 espécies e 18 gêneros. Estudos florísticos/taxonômicos sobre as Dalbergieae no Brasil são pontuais, o conhecimento sobre a sua diversidade ainda é fragmentado ou encontra-se desatualizado. Com relação a Caetité, esses estudos ainda são inexistentes. Este trabalho teve como objetivo realizar um estudo taxonômico dos representantes de Dalbergieae ocorrentes em Caetité, fornecendo chave de identificação, descrições, ilustrações para a maioria dos táxons e comentários sobre a distribuição geográfica e afinidades taxonômicas das espécies. Foram registradas 30 espécies distribuídas em 12 gêneros, sendo Stylosanthes (6 spp.), Dalbergia (5) e Machaerium (5) os gêneros mais representativos. Nissolia tomentosa está sendo registrada pela primeira vez para Bahia neste estudo. Pterocarpus zehntneri, considerada uma espécie de distribuição incerta no estado da Bahia, foi recoletada após 105 anos. Os táxons foram observados em todas as fitofisionomias registradas em Caetité, tendo um predomínio maior nas áreas de Caatinga arbustiva e de Cerrado s.s. Características das folhas, inflorescência, coloração das pétalas e tipo de fruto foram os caracteres mais relevantes para identificação dos táxons.
Book
Full-text available
Los árboles han sido un elemento secundario en el paisaje urbano de miles de ciudades en el mundo. Sin embargo, en Medellín cada vez se valoran más, no sólo por los aportes ambientales que puedan hacer a nuestra ciudad, sino por el valor estético y de identidad que representan para la población. El concepto de Silvicultura Urbana, integrado al manejo de las zonas verdes, es relativamente nuevo y está todavía en evolución. Con la transformación de Medellín en estos últimos años, este concepto ha tomado fuerza, y para la Administración ha implicado mejorar en su planeación y ordenación a gran escala, tanto en el área urbana como rural. La Secretaría del Medio Ambiente, como entidad encargada de determinar la política, ordenamiento, manejo y gestión de la Silvicultura Urbana, adelanta un trabajo responsable en torno al estudio, divulgación y valoración de la flora urbana de la ciudad. Dentro de este marco se inscribe la presente publicación, ARBOLES NATIVOS Y CIUDAD, que fue elaborada con el acompañamiento de personal especializado del Jardín Botánico de Medellín Joaquín Antonio Uribe. Lo anterior ha llevado a que la Administración, a traves del programa de Silvicultura Urbana, con su vivero Ciudad Verde y viveros comunitarios, le apunte a tener una oferta propia de árboles nativos con altos estándares de calidad, para ser establecidos en los proyectos de obra pública. El trabajo que presentamos permitirá a paisajistas, diseñadores, profesionales y población en general, conocer información morfológica, ecológica y paisajística de más de cincuenta especies arbóreas, arbustivas y palmas nativas, poco conocidas, susceptibles de ser incorporadas en espacios urbanos de la Medellín imparable por la que todos trabajamos.
Article
Full-text available
Studies of the distributions of species of seasonal woodland habitats in South America by means of dot-mapping and phytosociological analyses indicate the presence of three nodal areas: the Caatingas nucleus of arid northeastern Brazil; the Misiones nucleus, comprising a roughly right-angled triangular area enclosed by lines connecting Corumba-Puerto Suarez (Brazil/Bolivia) southward to Resistencia-Corrientes in northern Argentina, and eastward to the upper Uruguay River valley system in Argentinian Misiones and Brazilian Santa Catarina, and thus including most of eastern Paraguay and the west bank of the Paraguay River; and the Piedmont nucleus, which extends from Santa Cruz de la Sierra in Boliva to Tucuman and the sierras of east Catamarca in northwestern Argentina. Distribution maps are presented for over 70 taxa that show this kind of distribution pattern or its modifications. Most of the seasonal woodland species of this affinity are notable by their absence from the cerrado vegetation of central Brazil, although some occur on calcareous outcrops in the general cerrados area, and they also avoid the Chaco of northern Argentina. It is proposed that these fragmentary and mostly disjunct distributional patterns are vestiges of a once extensive and largely contiguous seasonal woodland formation, which may have reached its maximum extension during a dry-cool climatic period ca. 18,000-12,000 BP, coinciding with the contraction of the humid forest.
Article
Tipuana ecuatoriana sp. nov. (Leguminosae; Dalbergieae), is described from winged fruits collected in Miocene sediments of the Loja and Nabon Basins of Ecuador. Extensive comparisons with families that include genera with similar samaroid fruits or disperal units led to recognition of legume affinities for the fossil. Based upon the texture of the fruit, the position of the style remnant, and the venation of the fruit wing, the fossil conforms to the extant monotypic genus, Tipuana, native to subtropical Bolivia and northwest Argentina, but widely cultivated in the subtropics. Tipuana, and thus the Dalbergieae as presently recognized, were present in South America prior to the establishment of permanent land connection between North and South America.
Article
The genera Pictetia (traditionally with four to eight species) and Belairia (traditionally with six) are combined into a single genus, Pictetia, that now includes eight species confined to the Greater Antilles. Pictetia is diagnosed by pods that have a distinctly two-nerved placental margin, leaves of the long shoots subtended by persistent to caducous spinescent stipules, and coriaceous leaflets that usually have spinescent mucros. These morphological synapomorphic characters, as well as ones derived from the analysis of DNA sequence data from 5.8S and flanking internal transcribed spacers, support the hypothesis of monophyly for the genus. All species except P. sulcata are confined to a single island or island system, either the eastern half of Cuba, Hispaniola (though not including the southwestern peninsula), or Puerto Rico and the Virgin Islands excluding St. Croix. Pictetia sulcata occurs on central Hispaniola and eastern Cuba, and its restricted morphological variation in Cuba is suggestive of a relatively recent occurrence there. The historical biogeographic pattern of Pictetia supports a close area relationship among eastern Cuba, central Hispaniola, and Puerto Rico, which is consistent with Rosen's Caribbean vicariance model. Phylogenetically, Pictetia occupies a pivotal position within the aeschynomenoid legumes. It has been classified in the subtribe Ormocarpinae, which is shown here to be paraphyletic with respect to other aeschynomenoid genera. Among the genera of Ormocarpinae, Pictetia is most closely related to the African-Asian genus Ormocarpum, the Madagascan Ormocarpopsis, and the Meso-American Diphysa. These four genera represent one of the earliest branching clades within the aeschynomenoid radiation, according to analysis of non-molecular data. Combined and separate analyses of molecular and non-molecular data suggest that a clade comprising Ormocarpum and Ormocarpopsis is sister to Pictetia, whereas Diphysa represents the earliest branching lineage within this clade of four genera. This geographical relationship of Meso-America and the Greater Antilles with paleotropical Africa, Asia, and Madagascar is thought to be relictual from a Tertiary boreotropic distribution. Phylogenetic analysis reveals that the eight species of Pictetia are not conveniently grouped into infrageneric taxa, because the traditionally recognized series Racemosa (with P. aculeata and P. obcordata) and Fasciculatae (with P. sulcata and P. marginata) are paraphyletic with respect to the formerly recognized genus Belairia. One new combination, P. sulcata, is proposed.
Book
— We studied sequence variation in 16S rDNA in 204 individuals from 37 populations of the land snail Candidula unifasciata (Poiret 1801) across the core species range in France, Switzerland, and Germany. Phylogeographic, nested clade, and coalescence analyses were used to elucidate the species evolutionary history. The study revealed the presence of two major evolutionary lineages that evolved in separate refuges in southeast France as result of previous fragmentation during the Pleistocene. Applying a recent extension of the nested clade analysis (Templeton 2001), we inferred that range expansions along river valleys in independent corridors to the north led eventually to a secondary contact zone of the major clades around the Geneva Basin. There is evidence supporting the idea that the formation of the secondary contact zone and the colonization of Germany might be postglacial events. The phylogeographic history inferred for C. unifasciata differs from general biogeographic patterns of postglacial colonization previously identified for other taxa, and it might represent a common model for species with restricted dispersal.
Article
Tipuana ecuatoriana sp. nov. (Leguminosae; Dalbergieae), is described from winged fruits collected in Miocene sediments of the Loja and Nabon Basins of Ecuador. Extensive comparisons with families that include genera with similar samaroid fruits or disperal units led to recognition of legume affinities for the fossil. Based upon the texture of the fruit, the position of the style remnant, and the venation of the fruit wing, the fossil conforms to the extant monotypic genus, Tipuana, native to subtropical Bolivia and northwest Argentina, but widely cultivated in the subtropics. Tipuana, and thus the Dalbergieae as presently recognized, were present in South America prior to the establishment of permanent land connection between North and South America.