Flora 207 (2012) 662–672
Contents lists available at SciVerse ScienceDirect
Flora
journal homepage: www.elsevier.de/flora
Floristics and environmental factors determining the geographic distribution
of epiphytic bromeliads in the Brazilian Atlantic Rain Forest
Talita Fontoura a,∗ , Veridiana Vizzoni Scudeller b , Andrea Ferreira da Costa c
a
Universidade Estadual de Santa Cruz, Departamento de Ciências Biológicas, Rodovia Ilhéus-Itabuna km16, 45662-900 Ilhéus, BA, Brazil
Universidade Federal do Amazonas, Instituto de Biociências – ICB, Departamento de Biologia – Laboratório de Botânica, Av. General Rodrigo Octávio Jordão Ramos, 3000, Campus
Universitário, Coroado I, Manaus, AM, Brazil
c
Museu Nacional/UFRJ, Departamento de Botânica, Quinta da Boa Vista, São Cristóvão, 20940-040 Rio de Janeiro, RJ, Brazil
b
a r t i c l e
i n f o
Article history:
Received 25 September 2011
Accepted 1 May 2012
Keywords:
Bromeliaceae
Endemism
Diversity
Biogeography
a b s t r a c t
We investigated how epiphytic species and subfamilies of Bromeliaceae change along the extent of
the Atlantic Rain Forest, to answer the questions: (i) How do the epiphytic genera and subfamilies
of Bromeliaceae change along the domain? (ii) How similar are the different regions of the Atlantic
Rain Forest in relation to the epiphytic species of bromeliads? (iii) Which environmental variables
are the most important factors in determining species composition along the domain? We found 114
species of Bromelioideae and 73 of Tillandsioideae. The predominance of Bromelioideae was unexpected,
because they are not wind-dispersed as would be expected for most epiphytes. The smaller number of
species of Tillandsioideae, and the high frequency of species of Vriesea with limited geographic distributions indicated that epiphytes with rather limited geographic distributions predominate in this domain.
Species similarity was divided into one block of south–southeastern localities, and a second block of
northeastern–southeastern localities. These results suggest that the distribution of epiphytic bromeliad
species resembles that of the phorophyte trees, more than a previous pattern suggested for all epiphytes
in the domain. Latitude, temperature and altitude were important factors affecting the species composition along the domain. In general, our results differ from those of other studies in Latin America, and we
suggest that historical and evolutionary events generated these differences.
© 2012 Elsevier GmbH. All rights reserved.
Introduction
General distributional patterns of epiphytes were described in
the seminal work of Gentry and Dodson (1987). Based on the
number of epiphytic species and individuals found at different
sites in Ecuador, these authors emphasized that one of the most
striking distributional patterns shown by these plants is a tremendous decrease in the numbers of both species and individuals in
drier habitats (Gentry and Dodson, 1987). However, an increase in
humidity does not translate into a linear response in the number
of epiphytic species. Rather, epiphytes are best developed at midelevations (Ibisch et al., 2001; Kessler, 2000, 2001, 2002a,b; Krömer
and Kessler, 2005; Wolf and Flamenco-S, 2003), where moisture
supply is usually continuous (Benzing, 1990).
Several other investigations have explored large-scale or
regional aspects of the distribution of these plants in northern and
∗ Corresponding author. Tel.: +55 73 3680 5105; fax: +55 73 3680 5226.
E-mail addresses: fontoura.talita@gmail.com (T. Fontoura),
scudellerveridiana@hotmail.com (V.V. Scudeller), afcosta@acd.ufrj.br
(A.F.d. Costa).
0367-2530/$ – see front matter © 2012 Elsevier GmbH. All rights reserved.
http://dx.doi.org/10.1016/j.flora.2012.05.003
western regions of South America (Ibisch et al., 2001; Kessler, 2000,
2001, 2002a,b; Krömer and Kessler, 2005; Küpper et al., 2004).
For the eastern coast of Brazil the floristics and biogeography of
epiphytes have been investigated on a large scale only recently
(Kersten, 2010; Menini-Neto et al., 2009a). According to Kersten
(2010) the Atlantic Rain Forest harbors approximately 2250 epiphytic species.
The analysis of Menini-Neto et al. (2009a) focused on the floristic
similarities of epiphytes sensu lato (see Benzing, 1990) in southern
and south-eastern Brazil. According to these authors, the greatest similarity amongst the analyzed species distributions occurs
in two groups. The first includes species of south-eastern Brazil,
and a second group is composed by species of southern Brazil.
However, investigations on the arboreal flora indicated a different biogeographic pattern than that described for epiphytes when
areas from the north-eastern region of the Atlantic Rain Forest were
taken into account. For arboreal plants, there is a floristic continuum between forests in southern Bahia (north-eastern Brazil)
and the state of Rio de Janeiro in south-eastern Brazil (OliveiraFilho et al., 2005). Thus, an analysis of the epiphytic flora taking
into account also north-eastern localities could indicate whether
the floristic situation of epiphytes resembles the patterns that are
T. Fontoura et al. / Flora 207 (2012) 662–672
seen in arboreal plants. In addition, further analysis focusing on
species with a mainly epiphytic habit could clarify the extent to
which the proposed biogeographic pattern of Menini-Neto et al.
(2009a) is valid for species that depend solely on accumulated
debris and throughfall, for instance the epiphytes of the family
Bromeliaceae.
The Bromeliaceae is the second largest family in terms of epiphytic species (Gentry and Dodson, 1987), and contains some
2500 species with this habit (Benzing, 1990). Although epiphytic
bromeliads are characteristic of many humid areas of the Brazilian Atlantic Rain Forest (see Rizzini, 1992), much of the knowledge
about these plants in this domain remains embedded in general
studies of the family (Costa and Wendt, 2007; Fontoura et al., 1991;
Leme and Siqueira-Filho, 2006a,b; Pontes and Agra, 2006; Sousa
and Wanderley, 2000; Versieux and Wendt, 2006, 2007; Wanderley
and Martins, 2007), or is based on local studies of epiphytes (see
Alves et al., 2008; Menini-Neto et al., 2009a for references). These
investigations are heavily biased toward the southern region of
the Atlantic Rain Forest, where studies began long ago (Hertel,
1949).
In the Bromeliaceae, epiphytic species are concentrated in the
subfamilies Tillandsioideae and Bromelioideae (Smith and Downs,
1974, but see also Kessler, 2002a) which have distinct morphologies and dispersal modes. Tillandsioideae produce dehiscent
capsules and the seeds are wind-dispersed, whereas the Bromelioideae produce berry-type fruits with animal-dispersed seeds
(Smith and Downs, 1974). Local surveys in the Atlantic Rain Forest domain (e.g., Amorim et al., 2008, 2009; Fischer and Araújo,
1995; Kersten and Silva, 2001; Waechter, 1998) indicate that
species of Tillandsioideae predominate in the southern region
of Brazil, and that members of Bromelioideae predominate in
the northern region. However, based on the different environmental factors affecting the floristics of the Atlantic Rain Forest
(Marques et al., 2010; Oliveira-Filho and Fontes, 2000; Rizzini,
1992; Stehman et al., 2009), it seems unreasonable to expect
that the subfamilies and species of epiphytes change based on
only one environmental gradient. For instance, a floristic analysis of 125 localities in the domain indicated that the distance
from the ocean is correlated with seasonality, clearly separating
the effect of rainfall from that of semideciduous forest environments (Oliveira-Filho and Fontes, 2000). In addition, the floras are
influenced by altitude, temperature range, annual mean temperature, and rainfall distribution (Oliveira-Filho and Fontes, 2000).
Thus, the analysis of the epiphytes in southern, south-eastern
and north-eastern regions of this domain in relation to certain
environmental variables may help us to understand the flora
and how it is affected by environmental variables in different areas of this domain. Additionally, we compare our results
with other Latin American areas for which biogeographic studies
exist.
We characterized epiphytic bromeliad assemblages along the
domain, in order to answer the following questions: (i) How do
the epiphytic genera and subfamilies of Bromeliaceae change along
the domain? (ii) How similar are the different areas of the Atlantic
Rain Forest in relation to the epiphytic species of bromeliads? (iii)
Which environmental variables are the most important factors in
determining species composition along the domain?
Methods
Data matrix preparation
We used 25 surveys performed in the Atlantic Rain Forest to analyze epiphytic bromeliads along the domain (Table 1). Nine surveys
represented the southern region, 11 the south-eastern, and 5 the
663
north-eastern region of the domain (Table 1). For the Una region,
two surveys (Amorim et al., 2008; Fontoura and Santos, 2010) were
combined into one in the data matrix because both inventories
were carried out in the same region but in different periods of time,
using different field methodologies. Because of the large number of
synonymies and new species published for the state of Bahia, all
species names and identifications were verified by examination of
voucher specimens in the following herbaria: Centro de Estudos e
Pesquisa do Cacau (CEPEC), Herbarium Bradeanum (HB), Instituto
de Botânica (IB), Jardim Botânico do Rio de Janeiro (RB), Museu
Nacional do Rio de Janeiro (R), and Universidade Estadual de Santa
Cruz (HUESC).
Because Aechmea lingulata (L.) Baker has been split into 16
species (Leme and Siqueira-Filho, 2006a), all records of this species
in the surveys analyzed were excluded, in order to avoid imprecision in the results due to voucher misidentification. This decision
was taken in order to avoid false similarities in the multivariate
analysis.
Even though specificity of relationships between epiphyte
species and epiphyte phorophytes is rarely recorded (Benzing,
1990; Laube and Zotz, 2006), we avoided epiphyte surveys based
only on a small number of phorophytic species (Gonçalves and
Waechter, 2002, 2003; Reis and Fontoura, 2009), in order to maximize species richness in the sampled areas.
General aspects
We characterized epiphytic bromeliads along the domain based
on the percentage of subfamilies in each locality and in each geopolitical region. We also used the richest genus in each locality to
evaluate general patterns. Because the inventories used different
areas, methodologies and field efforts, it was expected that only the
most major changes in subfamilies and genera could be detected
along the domain.
Data analysis
Species similarities were analyzed using cluster and ordination
analyses based on a binary matrix of 184 species × 25 localities
(Appendix 1). Because 102 species were restricted to only one locality, a second matrix of 82 species × 25 localities was used in the
multivariate analysis, to assess possible modifications in the results
because of single species. The program FITOPAC 1 (Shepherd, 1995)
was used to calculate the Jaccard coefficient amongst sample units
(localities) and to generate a dendrogram using the unweighted
pair group method using arithmetic averages (UPGMA; Sneath and
Sokal, 1973).
The following environmental variables were used as possible
predictors of the epiphyte community along the domain: (i) mean
annual rainfall, (ii) distance from the ocean, (iii) altitude, (iv) latitude, (v) mean temperature, (vi) minimum mean temperature,
and (vii) relative humidity. Rainfall, altitude and mean temperature
were based on the information available in the epiphyte surveys.
The variable “distance from the ocean” was obtained by plotting
all localities on a map using the Geographic Information System
ArcGis ver. 9.2. Minimum mean temperature and relative humidity
were obtained from www.leb.esalq.usp.br/angelocci/.
The main matrix (184 species × 25 localities) and the secondary
matrix (25 localities × 7 environmental variables) were analyzed
using Canonical Correspondence Analysis (CCA), running the PCOrd
program ver. 3.11 (McCune and Mefford, 1997). The Monte Carlo
test was performed using 1000 iterations, to assess the significance
values of correlations. Biplot graphs were interpreted following Ter
Braak (1997).
Surveys (codes of
survey areas; for
locations see Fig. 1)
Geopolitical
region
Area (ha)
Sampling method
Vegetation
type and
reference
BA-1
BA-2
BA-3
BA-4
ES
MG-1
MG-2
NE
NE
NE
NE
SE
SE
SE
13(57.1)
29(59.2)
19(55.9)
14(60.9)
21(67.7)
6(46.2)
7(31.8)
9(42.9)
20(40.8)
15(44.1)
9(39.1)
10(32.3)
7(53.8)
15(68.2)
21
49
34
23
31
13
21
300
21 500
6000
7500
0.34
n.i.
n.i.
Random
Random
Random
Random
34 samples of 10 m × 10 m
Random
Random
8(80)
22(56.4)
5(45.5)
10
39
11
0.9
7500
1250
0
2(50)
11(40.7)
7(31.8)
8(47.1)
10(34.4)
1(6.3)
2(22.2)
3(33.3)
3(100)
2(50)
16(59.3)
15(68.2)
9(52.9)
21(65.6)
15(93.8)
7(77.8)
6(66.6)
3
4
27
22
17
32
16
9
9
22 500
22 500
22 500
Random
Random
119 phorophytes sampled in point
center quarter method
320 m × 320 m sample plot
320 m × 320 m sample plot
320 m × 320 m sample plot
Seven transects: 600–2400 m2
Transect 3 m × 2 km
320 m × 320 m sample plot
30 samples of 100 m2
110 random phorophytes
Random and 90 phorophytes
Highland ombrophilous dense foresta
Lowland ombrophilous dense forestb , c
Highland ombrophilous dense foresta
Highland ombrophilous dense foresta
Highland ombrophilous dense forestd
Seasonal semideciduous montane and gallery forestse
Gallery forest, evergreen upper montane forest, and
seasonal montane forest grovese
Alluvial ombrophilous dense forestf
Highland ombrophilous dense forestg
Restinga foresth
MG-3
RJ-1
RJ-2
SE
SE
SE
2(20)
17(43.6)
6(54.5)
SP-1
SP-2
SP-3
SP-4
SP-5
SP-6
PR-1
PR-2
PR-3
SE
SE
SE
SE
SE
SE
S
S
S
SC-1
S
4(33.3)
8(66.7)
12
n.i.
RS-1
RS-2
RS-3
RS-4
S
S
S
S
4(40)
2(33.3)
2(16.7)
2(16.7)
6(60)
4(66.7)
10(83.3)
10(83.3)
10
6
12
12
n.i.
n.i.
20 180
1
RS-5
S
2(22.2)
7(77.8)
9
Species
number of
subfamilies (%)
Bro
c
d
e
f
g
h
i
j
k
l
m
n
o
p
q
r
s
t
Amorim et al. (2009).
Amorim et al. (2008).
Fontoura and Santos (2010).
Varassin (2002).
Menini-Neto et al. (2009a).
Menini-Neto et al. (2009b).
Fontoura et al. (1997).
Fontoura and Reinert (2009).
Breier (2005).
Santos (2000).
Fischer and Araújo (1995).
Kersten and Silva (2001).
Kersten and Silva (2002).
Kersten et al. (2009).
Bonnett and Queiroz (2006).
Rogalski and Zanin (2003).
Buzatto et al. (2008).
Oliveira et al. (2005).
Waechter (1998).
Giongo and Waechter (2004).
Til
0.6
22 500
0.3
8.6
n.i.
n.i.
240 phorophytes sampled in point
center quarter method
Random
Random
Random
60 phorophytes in point center quarter
method
60 phorophytes sampled in point
center quarter method
Seasonal semideciduous foresti
Savannah vegetationi
Ombrophilous dense foresti
Dense forestj
Riparian, dense and restinga forestk
Restinga foresti
Ombrophilous dense forest and restingal
Mixed ombrophilous alluvial forestm
Mixed ombrophilous alluvial and high-altitude montane
forestn
Ombrophilous dense foresto
Semideciduous forestp
Mixed ombrophilous forestq
Semideciduous seasonal forestr
Restinga forests
Riparian forestt
T. Fontoura et al. / Flora 207 (2012) 662–672
a
b
Total number of
Bromeliaceae
664
Table 1
General aspects of field surveys along the Brazilian Atlantic Rain Forest. Geopolitical regions: NE – northeastern, SE – southeastern, S – south. State codes: BA – Bahia, ES – Espírito Santo, MG – Minas Gerais, RJ – Rio de Janeiro, SP
– São Paulo, PR – Paraná, SC – Santa Catarina, RS – Rio Grande do Sul. Subfamilies of Bromeliaceae: Bro – Bromelioideae, Til – Tillandsioideae. n.i., not indicated in reference.
T. Fontoura et al. / Flora 207 (2012) 662–672
665
Table 2
Surveys (codes), total amount of epiphytic Bromeliaceae species (richness), and
richness of the richest genus occuring in the Brazilian Atlantic Rain Forest.
Survey
1. BA-1
2. BA-2
3. BA-3
4. BA-4
5. MG-1
6. MG-2
7. MG-3
8. ES
9. RJ-1
10. RJ-2
11. SP-1
12. SP-2
13. SP-3
14. SP-4
15. SP-5
16. SP-6
17. PR-1
18. PR-2
19. PR-3
20. SC
21. RS-1
22. RS-2
23. RS-3
24. RS-4
25. RS-5
Richness
21
49
34
23
13
21
10
31
39
11
3
4
27
22
17
32
16
9
9
12
10
6
12
12
9
Genera (number of species)
Vriesea (6)
Vriesea (11)
Vriesea (10)
Vriesea (7)
Vriesea (4)
Vriesea (7)
Tillandsia (4), Vriesea (4)
Aechmea (9)
Vriesea (17)
Aechmea (3), Tillandsia (3)
Tillandsia (3)
Tillandsia (2)
Vriesea (12)
Vriesea (10)
Vriesea (8)
Vriesea (14)
Vriesea (11)
Tillandsia (6)
Tillandsia (4)
Tillandsia (4), Vriesea (4)
Tillandsia (5)
Tillandsia (4)
Tillandsia (6)
Tillandsia (5), Vriesea (5)
Tillandsia (6)
Floristic similarities amongst localities
Fig. 1. Location of the surveys along the Brazilian Atlantic Rain Forest; dominant
subfamilies and richest genus: asterisks – higher percentage of Tillandsioideae
species, dots – higher percentage of Bromelioideae; A – Aechmea, T – Tillandsia,
V – Vriesea.
Results
General aspects
The 184 species analyzed comprised 112 Bromelioideae and 72
Tillandsioideae, i.e., 1.6 times more Bromelioideae than Tillandsioideae. In general, the number of species decreased toward the
southern region of the domain (34 species) where Tillandsioideae
reached the highest percentage (76.5%). The predominance of this
subfamily remains high from the southern region to the inland
(MG-1, 2, 3) and the mountainous areas (RJ-1) of the south-eastern
region of the domain (Fig. 1, Table 1). Bromelioideae predominated from the restinga forest of the south-eastern region (RJ-2)
northwards to all other localities (Fig. 1). The north-eastern (83
species) and south-eastern (117 species) regions harbored the
highest percentages of Bromelioideae (63.9% and 53.8%, respectively).
Vriesea was the most frequent genus, occurring in 15 (60%)
localities, followed by Tillandsia, which occurred in nine (36%,
Table 2). Although species of Tillandsia occur in some localities of the south-eastern region, the genus is mainly represented
in the southern-most region and at inland localities of the
Atlantic Rain Forest. Vriesea predominated from the northeast until
parts of the southern region, and in localities near the ocean
(Fig. 1).
The cluster analysis resulted in two large groups and five subgroups with similarity values around 0.1 (Fig. 2a). The first group
was composed by three subgroups with localities in the southern
and south-eastern regions of the Atlantic Rain Forest (Fig. 2a). The
localities of the first subgroup were characterized by the presence
of Tillandsia recurvata, T. stricta, and T. usneoides in at least nine of 12
localities. The second subgroup comprised seven localities, where
Vriesea carinata, V. flammea, and V. philippocoburgii were shared
species in six or all localities of this subgroup. The third subgroup
consisted only of SP-2, where B. zebrina and T. recurvata were shared
with seven and nine other localities of this first, large group.
The second group was composed by two subgroups, with one
locality from the south-eastern and all the surveys from the northeastern region (Fig. 2a). All the surveys in southern Bahia formed
the fourth subgroup. Twelve species occurred in at least three localities of this subgroup: B. saundersii, Neoregelia kerryi, Nidularium
bicolor, Ni. innocentii, Racinaea spiculosa, T. meridionalis, V. vagans,
V. ensiformis, V. psittacina, V. flammea, V. minuta, and V. simplex. The
fifth subgroup consisted of the Espírito Santo survey, with seven
species in common with three or four other localities of this group:
Aechmea nudicaulis, Portea petropolitana, R. spiculosa, T. tenuifolia,
V. drepanocarpa, V. psittacina, and V. simplex.
The cluster analysis with 82 species, excluding singletons,
formed the same two large groups and subgroups as shown in the
full analysis (Fig. 2b). The only differences occurred in the linkage
level of the subgroups.
Environmental variables
The high eigenvalues (>0.5) on CCA axes 1, 2, and 3 indicate
that much of the variation in the species data is explained by
these axes (Table 3), and considerable turnover of species occurs
along these axes. The cumulative variance of 21.6% on the third
axis indicates that considerable noise remained unexplained by
the CCA. Correlation values between species and environmental
variables were consistently high (all >0.96) and significant values
were obtained on the first and second axis (Table 3), indicating
666
T. Fontoura et al. / Flora 207 (2012) 662–672
Fig. 2. Cluster analysis of epiphytic bromeliad species along the Brazilian Atlantic Rain Forest domain. Localities divided in first group (G1) and second group (G2). South
surveys denoted in gray, southeastern in black, and northeastern surveys in italics. (A) Dendrogram based on 184 species and 25 localities along the domain. (B) Dendrogram
based on 82 species and 25 localities.
Table 3
Summary statistics of CCA and P values of Monte Carlo tests. Data matrix of 187 bromeliad epiphytic species × 25 localities along the Brazilian Atlantic Rain Forest.
Eigenvalue
Cumulative variance explained in species data
Pearson correlation between species and environmental variables
P values for eigenvalues
P values for correlations between species and environmental variables
Axis 1
Axis 2
Axis 3
0.711
9.1
0.992
0.001
0.002
0.524
15.8
0.977
0.006
0.01
0.459
21.6
0.970
0.014
0.09
T. Fontoura et al. / Flora 207 (2012) 662–672
Table 4
Intraset correlations between environmental and geographical variables and CCA
axis. Data matrix of 187 bromeliad epiphytic species × 25 localities along the Brazilian Atlantic Rain Forest. Highest values denoted in bold.
Variable
Axis 1
Axis 2
Axis 3
Latitude
Altitude
Distance from ocean
Rainfall
Mean temperature
Minimum temperature
Relative humidity
0.945
0.078
0.385
0.061
−0.792
−0.875
−0.413
0.063
−0.543
−0.483
0.438
0.199
−0.176
0.341
−0.138
0.685
−0.150
0.349
−0.266
−0.228
0.069
that results can be interpreted based on the variation of environmental variables of both axes of the biplot. The high values of
correlations (r > 0.6) between environmental variables and the CCA
axes (Table 4) indicate that latitude and mean and minimum temperatures are important gradients on axis 1. The most important
gradient on axis 3 of the biplot is represented by altitude.
667
All localities of Bahia, Espírito Santo, and RJ-2 were segregated in
the left region of the biplot, and all other localities were positioned
in the right region (Fig. 3). The projection of localities onto the vectors indicated the degree to which these variables influenced the
species composition of epiphytes in the different localities of the
Atlantic Rain Forest. Latitude moderately influenced the species
composition of MG-1, RJ-1, SP-3, 5, 6, PR-1, and influenced species
composition of SP-4, PR-2, SC-1 and RS-4. The distance from the
ocean and, in lesser degree, the altitude influenced epiphytic community of localities MG-2, 3, SP-1, 2, PR-3 and RS-1, 2, 3, 4, 5.
Temperature was an important factor for epiphytes in all localities
of Bahia and Espírito Santo, and moderately influenced epiphytes
of RJ-2 (Fig. 3).
Discussion
General aspects
Our analysis added further details on how members of the
Bromelioideae predominate in the Atlantic Rain Forest, and how
Axis 2
SP-6
1,0
SP-3
SP-5
SP-4
PR-1
BA-2
PR-2
RH
RJ-1
Tme
MG-1
Lat
BA-4
Tmi
-1,5
BA-1
-0,5
BA-3
0,5
Axis 1
RS-4
SC-1
1,5
RJ-2
Alt
RS-5
DistOce
RS-3
PR-3
RS-2
MG-2
-1,0
MG-3
RS-1
SP-1
ES
SP-2
-2,0
Fig. 3. Biplot of Canonical Correspondence Analysis based on 184 species of epiphytic bromeliads in 25 localities along the Brazilian Atlantic Rain Forest. Species names were
omitted to preserve clarity. Rainfall eigenvector is clear only to the third axis (not shown). Environmental variables: Lat – latitude, DistOce – distance from the ocean, Alt –
altitude, Tmi – mean of minimum temperature, Tme – mean temperature, RH – relative humidity.
668
T. Fontoura et al. / Flora 207 (2012) 662–672
the main subfamilies of epiphytic species shift along the domain.
In the surveys, Bromelioideae comprised 1.6 times more epiphytic species than Tillandsioideae. For the family as a whole (see
Martinelli et al., 2008), Bromelioideae comprises 2.3 times more
species than Tillandsioideae. The preponderance of this subfamily
in both studies indicates that the Bromelioideae generally surpasses the Tillandsioideae in number of species, considering either
all life forms or only the epiphytic species. If our findings represent the general pattern of subfamilies of epiphytic bromeliads in
the domain, this implies that most members of Bromeliaceae in
the Atlantic Forest should be animal-dispersed epiphytes, according to the fruit type. This pattern contrasts with bromeliads of the
Bolivian Andes, which are mainly wind-dispersed species (Kessler,
2002a).
In the Brazilian Atlantic Rain Forest, the highest percentage
of Tillandsioideae is located in the southern region, where the
overall species richness is lower. The variety of sampling methods and areas amongst the different surveys prevents detailed
comparisons of raw richness numbers. However, it is noteworthy that four field surveys in the north-eastern region collected
2.4 times as many species as did nine surveys in southern Brazil,
and 0.7 times fewer species than 12 surveys in south-eastern
Brazil. Even though our data suggest that differences in field efforts
did not affect the general pattern of low richness in the southern region, field surveys using the same methodology along the
domain would be useful to measure more precisely modifications
in alpha, beta and gamma diversity along different gradients (see
Cody, 1993).
The relatively few species in southern Brazil were mainly
members of Tillandsioideae, which suggests that the low winter
temperatures in this region may limit species richness. In addition, if we consider that Tillandsioideae evolved in northern South
America (Barfuss et al., 2005; Givnish et al., 2007), then the geographic distance between the center of diversity of this subfamily
and southern Brazil may be a second factor contributing to the
rather low richness of Tillandsioideae in this region.
The prominence of Tillandsioideae in the south is very clear until
the mountainous region of Rio de Janeiro state. However, from the
restinga forest, which at its closest point is located approximately
53 km from the mountains, to the more northern regions of the
domain, there is a shift in the predominant subfamily. This turnover
of species suggests that the region of Rio de Janeiro may have an
important biogeographic role, where both subfamilies co-occur in
the different habitats of this state.
The occurrence of higher percentages of Tillandsioideae in
most localities of the Atlantic Rain Forest indicates that the mesic
Vriesea and the Tillandsia species which depend on atmospheric
humidity, exploit different habitats establishing and increasing
their species number along the domain. Regardless of the different methodologies and field efforts used in the inventories, it is
clear that mountainous areas of Serra do Mar (represented by
RJ-1, SP-3, 4, 5, 6) and several other regions (e.g., MG-1, 2) are
dominated by Vriesea species. Thus, if the bromelioid epiphytic
clade in and around the Serra do Mar originated roughly 5.5 Ma
(Givnish et al., 2011), it would be interesting to investigate which
group of Bromeliaceae first arrived in this mountainous region.
Nowadays, members of Tillandsioideae are overwhelmingly represented.
Although Gentry and Dodson (1987) and Wolf and Flamenco-S
(2003) stated that wind-dispersed epiphytes are better represented
in relatively dry forests, this pattern does not apply well to the
Atlantic Rain Forest. Wind-dispersed epiphytic bromeliads dominate in different parts of the south-eastern and southern regions of
the domain, either in humid (e.g., SP-5, 6) or in drier locations of
the domain (SP-2, RS-1). Actually, our data indicate that Tillandsia is
the richest genus in the localities farthest inland, and Vriesea is the
richest genus in localities near the Atlantic Ocean. Although several
investigations have explored how Bromelioideae have evolved and
gained their rosulate and epiphytic habit (Sass and Specht, 2010;
Schulte et al., 2005, 2009), it is virtually unknown how members of
Tillandsia and Vriesea succeeded in occupying such a large area of
humid forest in the Brazilian Atlantic Rain Forest. For the other parts
of the domain, our data indicate that animals have played a prominent role in dispersing epiphytic bromeliads. This is an unexpected
result in terms of epiphytism, because usually wind is the most
common dispersal mode amongst epiphytes (Gentry and Dodson,
1987).
The dominance of Bromelioideae begins in the restinga forest of
Rio de Janeiro (RJ-2), increases toward the drier (see also Amorim
et al., 2005; Reis and Fontoura, 2009) and humid areas of the northeastern region (e.g., BA-2), and decreases toward the south-eastern
and southern regions of Brazil. The various ecosystems, rainfall
amounts, and altitudes indicate that the dominance of this subfamily is not related to one or another habitat, but rather to evolutionary
trends and to the means by which the domain was occupied by this
life form. Apparently, the southern-most regions of the domain are
unfavorable for the establishment of many bromelioid epiphytes.
As the members of this subfamily are exclusively animal-dispersed,
possibly the available vectors were or still are unable to disperse
epiphytic bromelioids along the domain. Perhaps the refuge theory (Whitmore and Prance, 1987) may explain all or part of the
observed pattern. However, further studies based on palaeopalynological and phylogenetic investigations may reveal to what degree
this theory is valid for this region with high species richness and
endemism.
Our data for the most species-rich genera in the domain, added
to the data contained in Martinelli et al. (2008), Smith and Downs
(1974), and Stehmann et al. (2009), suggest intriguing questions
about the geographic range of epiphytism. Vriesea is the largest
genus in the Atlantic Rain Forest (Martinelli et al., 2008), and about
87% of the Vriesea species in this domain are epiphytic (see Smith
and Downs, 1977; Stehmann et al., 2009). Added to these facts,
our data show that Vriesea is also the most frequent genus occurring in almost all Brazilian states. Of the 166 Vriesea species in the
domain, only 3.5% occur throughout its extent, and 70% occur in
only one or two Brazilian states (see Martinelli et al., 2008); i.e., the
great majority of species have a rather restricted geographic distribution. Therefore, even if surveys were regularly distributed in
forested areas to allow the collection of this highly epiphytic genus,
the geographic distribution of species would remain biased toward
a narrow geographic distribution, reflecting the distributions of
Vriesea species. Although quantitative investigations are highly
desirable, the evaluation of the geographic distributions of our 187
analyzed species, added to the data provided by Martinelli et al.
(2008) and Versieux and Wendt (2007), reinforces the hypothesis
that epiphytic bromeliads of this domain probably have generally smaller ranges than those occurring in the Bolivian Andes
(Kessler, 2002a). Seventy-six species (40.6%) occur in only one or
two Brazilian states (see Martinelli et al., 2008), mainly represented
by localities in southern Bahia and Espírito Santo, representing
a geographic distribution of approximately 14 squared degrees.
Twenty-four species (12.8%) occur in Minas Gerais, Espírito Santo
and Rio de Janeiro. If these 24 species would occupy this geographic range completely, they would cover in total an area of 74
squared degrees (see Versieux and Wendt, 2007). These figures are
different from those shown by Kessler (2002a), which indicated a
mean range size of 147 squared degrees for the epiphytic bromeliads of the Bolivian Andes. Further quantitative investigations may
confirm whether bromeliads of the Atlantic Rain Forest have as
extensive a geographic distributions as those reported by Kessler
(2000, 2001, 2002a,b), and by Givnish et al. (2007) for epiphytes in
general.
T. Fontoura et al. / Flora 207 (2012) 662–672
Floristic similarities
Both multivariate analyses resulted in very similar clusters,
indicating that the exclusive species had no effect on the cluster
analysis. The analyses also revealed that the similarities of groups
and subgroups of epiphytic bromeliads along the domain were not
based on the dichotomy between the south-eastern and southern
regions, as observed for the epiphytes as a whole (Menini-Neto
et al., 2009a). With the available data for epiphytic plants, it also
cannot be said that the north-eastern region is strongly dissimilar to
other regions of this domain, as occurs for arboreal plants (Scudeller
et al., 2001). The number of shared species amongst localities was
apparently the determining factor in the division of groups and subgroups. This characteristic is especially clear when the two main
groups are compared. In the first group, all subgroups have three to
four shared species amongst many localities in the southern and
part of the south-eastern region of the domain. In contrast, the
second group has seven to 12 shared species amongst localities
in Espírito Santo and southern Bahia. Actually, the resulting dendrogram and main divisions of localities indicate that the general
pattern of epiphytic bromeliads along the domain conforms to the
pattern known as the “Campos dos Goytacazes gap” as described
by Oliveira-Filho and Fontes (2000) for arboreal plants. According
to this pattern, the low rainfall in northern Rio de Janeiro state has
led to the development of a semi-deciduous forest. Apparently, in
their history the epiphytic bromeliads were sensitively affected by
the low rainfall of this region, and the net result of such effects is
reflected in our dendrograms that clearly separate the species in
southern Bahia and Espírito Santo from all other localities of the
domain. Although additional field surveys are necessary to better
compare the floras of Rio de Janeiro and Espírito Santo, it is interesting to note that the epiphyte flora of mountainous areas of Espírito
Santo is dissimilar to the flora of the Rio de Janeiro mountains –
which is different from the situation of arboreal plants (OliveiraFilho et al., 2005). Most probably, it was indeed the low rainfall in
northern Rio de Janeiro that contributed to separate the epiphyte
flora into two blocks.
Environmental variables
The CCA indicated that latitude is an important factor determining the species composition in almost all localities near the Atlantic
Ocean, where rainfall occurs in all months of the year. In most of
these localities, Vriesea is the most species-rich genus. The first
exception is in extreme southern Brazil (RS-4), where Tillandsia and
Vriesea share the highest richness in a restinga forest near the ocean.
The second exception is represented by the mixed ombrophilous
alluvial forest of PR-2 where Tillandsia is the most species-rich
genus. Because the presence of Tillandsia is usually associated with
high light intensities (Benzing, 1990, 2000), finer studies may reveal
whether natural forest formation or human disturbances may have
allowed the presence of this genus in both localities. However, it
would be reasonable to state that latitude is an important factor
where Vriesea is the richest genus – in other words, in most of
the humid localities in south-eastern and southern Brazil. In these
localities, other environmental variables have relatively little effect
on the species composition. Apparently, species of this genus take
advantage of the high levels of rainfall in these humid locations
to accumulate water in their tanks, as a means to compensate for
physiological dryness imposed by their occupation of tree crowns
and trunks.
Distance from the ocean and, in lesser degree, altitude were
important variables affecting species composition in most localities
where Tillandsia is the most species-rich genus. The first exception
is represented by the surveyed areas in MG-2 (gallery forest, evergreen upper montane forest, and seasonal montane forest groves)
669
and the second, by the alluvial ombrophilous dense forest in MG3. It is probable that in spite of the large distance between these
localities and the Atlantic Ocean, there is enough humidity in the
gallery forest, evergreen upper forest, and in the alluvial forest to
allow the establishment of several Vriesea species. In relation to
the low contribution of altitude, it is probable that our large scale
study encompassing almost 15◦ in latitude and 25◦ in longitude is
too large to detect altitudinal patterns – that may be better analyzed
at a smaller scale. The Brazilian Atlantic Rain Forest does not form
a massif barrier of, e.g., mountain ridges (see Fiaschi and Pirani,
2009; Oliveira-Filho and Fontes, 2000) along the entire domain,
and do not range from 0 to more than 3000 m.a.s.l., as is the altitudinal span of Bromeliaceae occurrences in the Andes (see Haffer,
1987; Küpper et al., 2004). In eastern Brazil the mountainous areas
are scattered along the whole domain (Safford, 2007) and probably
other local factors may be responsible for the species distribution, so that the altitude parameter has only low weight in our
results. As noted by Oliveira-Filho and Fontes (2000), the influence
of altitude is far more complex than simply creating temperature
gradients and frost events. Rising elevation also decreases atmospheric pressure, increases solar radiation, accelerates air masses,
promotes greater cloudiness, and increases rainfall (Jones, 1992).
Although evaluation of other variables is necessary to better investigate the effects of environmental variables in drier locations, it is
interesting to note that the occupation of these regions probably
became possible because Tillandsia species became able in their
evolution to use atmospheric humidity overcoming the dry periods characteristic of these more-inland regions of the Atlantic Rain
Forest.
Finally, in the correlations, minimum temperature turned out
to be important affecting the species composition of localities at
altitudes from 640 to 775 m.a.s.l. (BA-1, 3 and 4), and mean annual
temperature resulted as an important factor for the lowland area
of Una (BA-2), located about 100 m.a.s.l. For arboreal vegetation
temperature can be one of the most important factors, shaping the
world’s vegetation formations as well as the plant cover in southern Brazil (Oliveira-Filho and Fontes, 2000) where arboreal species
richness is low. However, to epiphytic bromeliads both temperature variables are important especially in localities where, at low
latitude, the majority of species are animal dispersed – probably in one of the richest areas of the domain with concern of the
Bromeliaceae flora. Apparently, the association between epiphytic
bromeliads and animals was evolutively profitable, promoting both
species richness and endemism. Clearly, other variables ought to be
taken into account, and for some parameters our scale of analysis,
encompassing a large part of the Atlantic Rain Forest, perhaps may
obscure processes that are better understood on smaller scales (e.g.,
pollination and dispersal processes). Moreover phylogeographic
studies must complement now the statistical treatments of floristic
relations in order to support or correct the biogeographic patterns
presented here.
Acknowledgments
This paper was part of post-doctoral work of TF in the Departamento de Botânica do Museu Nacional/Universidade Federal do
Rio de Janeiro. We are grateful to the curators of the Centro de
Estudos e Pesquisa do Cacau (CEPEC), Herbarium Bradeanum (HB),
Instituto de Botânica (IB), Jardim Botânico do Rio de Janeiro (RB),
Museu Nacional do Rio de Janeiro (R), and Universidade Estadual de
Santa Cruz (HUESC), and to the curators of the M and Gray Herbaria,
who kindly provided images of Bromeliaceae from Bahia. We are
also grateful for many suggestions from reviewers, which led to
substantial improvements in the manuscript.
T. Fontoura et al. / Flora 207 (2012) 662–672
670
Appendix 1. Species of epiphytic bromeliads of the
Brazilian Atlantic Rain Forest used in the data matrix (184
species × 25 localities) as the basis of multivariate analyses
Species
Surveys (codes of survey areas; for
locations see Fig. 1)
Acanthostachys strobilacea (Schult.
f.) Klotzsch
Aechmea araneosa L.B. Sm.
A. blanchetiana (Baker) L.B. Sm.
A. bromeliifolia (Rudge) Baker
A. bruggeri Leme
A. caesia E. Morren ex Baker
A. calyculata Baker
A. capitata Baker
A. cariocae L.B. Sm.
A. castanea L.B. Sm.
A. coelestis (K. Koch) E. Morren
A. conifera L.B. Sm.
A. depressa L.B. Sm.
A. distichantha Lem.
A. fasciata (Lindl.) Baker
A. floribunda Mart. ex Schult. f.
A. gamosepala Wittm.
A. hostilis E. Pereira
A. leonard-kentiana H.E. Luther &
Leme
A. leucolepis L.B. Sm.
A. marauensis Leme
A. miniata Baker
A. mollis L.B. Sm.
A. nudicaulis (L.) Griseb.
SP-2, ES
A. organensis Wawra
A. ornata Baker
A. pectinata Baker
A. perforata L.B. Sm.
A. pineliana (Brongn. ex Planch.)
Baker
A. ramosa Mart. ex Schult. f.
A. recurvata (Klotzsch) L.B. Sm.
A. sphaerocephala Baker
A. tentaculifera Leme, Amorim &
J.A. Siqueira
A. triangularis L.B. Sm.
A. turbinocalyx Mez
A. victoriana L.B. Sm.
A. viridostigma Leme & Luther
Billbergia alfonsi-joannis Reitz
B. amoena (Lodd.) Lindl.
B. distachia (Vell.) Mez
B. euphemiae E. Morren
B. horrida Regel
B. morelii Brongn.
B. nutans H. Wendl. ex Regel
B. pyramidalis (Sims) Lindl.
B. sanderiana E. Morren
B. saundersii Read
B. tweedieana Baker
B. vittata Brongn. ex Morel
B. zebrina (Herb.) Lindl.
Canistropsis billbergioides (Schult.
f.) Leme
Canistrum camacaense Martinelli &
Leme
C. montanum Leme
C. seidelianum W. Weber
C. triangulare L.B. Sm. & Reitz
Catopsis berteroniana (Schult. &
Schult. f.) Mez
C. sessiliflora (Ruiz & Pav.) Mez
Edmundoa lindenii (Regel) Leme
Guzmania lingulata (L.) Mez
Hohenbergia augusta (Vell.) E.
Morren
H. belemii L.B. Sm. & Read
ES
BA-2
ES
MG-3
RJ-1
RS-1
BA-4, BA-3
BA-2
ES
SP-3, RJ-1
BA-2
BA-2
PR-3, PR-2
RJ-2
RJ-2
SP-6, SP-5, SP-3
ES
BA-2
BA-2
BA-2
BA-1
BA-2
SC-1, PR-1, SP-6, SP-5, SP-4, SP-3,
ES, MG-2, BA-4, BA-3
SP-5, SP-4
SP-4, SP-3, RJ-1
SP-6, SP-5, SP-3
BA-2
RJ-1, ES
MG-1, BA-1
RS-5, RS-4, RS-2, RS-1, PR-3, PR-2
RJ-2
BA-4, BA-3
ES
BA-2, BA-1
ES
BA-4, BA-3, BA-1
MG-2
SP-6, SP-4, SP-3, RJ-2, RJ-1, ES
MG-3
MG-1, BA-3
BA-4
BA-2
RS-3, RS-2, RS-1, PR-3
RJ-2, RJ-1
RJ-1
BA-4, BA-3, BA-2, BA-1
MG-1
ES
RS-5, RS-4, RS-3, RS-1, SC-1, SP-6,
SP-2, MG-1
SP-3
BA-4, BA-3
BA-3, BA-1
BA-3, BA-1
ES
PR-1
SP-6, BA-2
SC-1, SP-6, SP-4, SP-3, RJ-1
BA-3, BA-2
ES
BA-4, BA-2
Species
Surveys (codes of survey areas; for
locations see Fig. 1)
H. blanchetii (Baker) E. Morren ex
Mez
H. brachycephala L.B. Sm.
H. edmundoi L.B. Sm. & Read
H. hatschbachii Leme
H. minor L.B. Sm.
H. ramageana Mez
Lymania alvimii (L.B. Sm. & Read)
Read
L. azurea Leme
L. corallina (Beer) Read
L. marantoides (L.B. Sm.) Read
L. smithii Read
Neoregelia azevedoi Leme
N. carolinae (Beer) L.B. Sm.
N. crispata Leme
N. eltoniana W. Weber
N. farinosa (Ule) L.B. Sm.
N. fluminensis L.B. Sm.
N. ibitipocensis (Leme) Leme
N. kerryi Leme
N. laevis (Mez) L.B. Sm.
N. longisepala E. Pereira & Leme
N. lymaniana R. Braga & Sucre
N. magdalenae L.B. Sm. & Reitz
N. oligantha L.B. Sm.
N. pauciflora L.B. Sm.
N. wilsoniana M.B. Foster
Nidularium amorimii Leme
N. antoineanum Wawra
N. bicolor (E. Pereira) Leme
N. cariacicaense (W. Weber) Leme
N. ferdinandocoburgii Wawra
N. innocentii Lem.
BA-2
N. krisgreeniae Leme
N. logiflorum Ule
N. marigoi Leme
N. microps E. Morren ex Mez
N. procerum Lindm.
N. rubens Mez
N. rutilans E. Morren
N. scheremetiewii Regel
Portea filifera L.B. Sm.
P. nana Leme & H.E. Luther
P. petropolitana (Wawra) Mez
Quesnelia arvensis (Vell.) Mez
Q. clavata Amorim & Leme
Q. humilis Mez
Q. koltesii Amorim & Leme
Q. lateralis Wawra
Q. liboniana (De Jonghe) Mez
Q. quesneliana (Brongn.) L.B. Sm.
Q. strobilispica Wawra
Q. testudo Lindm.
Racinaea aeris-incola (Mez) M.A.
Spencer & L.B. Sm.
R. spiculosa (Griseb.) M.A. Spencer
& L.B. Sm.
Ronnbergia brasiliensis E. Pereira &
L.A. Penna
R. silvana Leme
Tillandsia aeranthos (Loisel.) L.B.
Sm.
T. bulbosa Hook.
T. crocata (E. Morren) Baker
T. dura Baker
T. gardneri Lindl.
T. geminiflora Brongn.
T. globosa Wawra
T. kautskyi E. Pereira
T. loliacea Mart. ex Schult. f.
T. mallemontii Glaziou ex Mez
T. meridionalis Baker
T. pohliana Mez
BA-2
BA-3
BA-2
BA-4
BA-2
BA-2
BA-2
BA-2
BA-1
BA-2
BA-3
RJ-1
BA-3, BA-2
RJ-2
MG-1
RJ-1
MG-2
BA-4, BA-3, BA-1
SP-6, SP-3
BA-2
MG-2
ES
MG-2
ES, BA-3
BA-2
BA-2
SP-5
BA-4, BA-3, BA-1
ES
MG-2
SC-1, SP-6, SP-5, SP-4, RJ-1, BA-4,
BA-3, BA-2, BA-1
SP-3
RJ-1, ES, MG-1
MG-2
RJ-1
SP-6, RJ-1, ES, BA-2
SP-4
SP-3
RJ-1
BA-1
BA-3
ES, BA-4, BA-2
SP-6
BA-3
SP-5
BA-4
RJ-1
RJ-1
ES
ES
SP-5
MG-2, RJ-1
SP-6, SP-3, RJ-1, ES, BA-3, BA-2,
BA-1
BA-3, BA-2
BA-1
RS-5, RS-4, RS-3, RS-2
BA-2
PR-2
SP-6
RS-4, RS-1, PR-1, SP-6, RJ-2, MG-3,
MG-2, BA-2
RS-5, RS-4, RS-3, RS-1, SC-1, PR-1,
SP-6, SP-4, RJ-1, ES, MG-2, MG-1
SP-6, SP-3
ES
SP-1
SC-1, PR-2
BA-4, BA-3, BA-1
SP-2, BA-3
T. Fontoura et al. / Flora 207 (2012) 662–672
Species
Surveys (codes of survey areas; for
locations see Fig. 1)
T. recurvata (L.) L.
RS-5, RS-3, RS-2, RS-1, PR-3, PR-2,
SP-2, SP-1, MG-3, MG-2
BA-2
RS-5, RS-4, RS-3, RS-2, RS-1, SC-1,
PR-3, PR-2, SP-6, SP-5, SP-4, SP-3,
SP-1, RJ-2, RJ-1, MG-2, MG-1, BA-4,
BA-3, BA-2
RS-5, RS-3, RS-1, PR-3, PR-2, PR-1,
SP-6, SP-4, SP-3, RJ-1, ES, MG-2,
BA-2
MG-3
RS-5, RS-4, RS-3, RS-2, SC-1, PR-3,
PR-2, SP-4, RJ-2, MG-3, MG-2,
MG-1, BA-2
PR-1, SP-6, SP-4
RJ-1
RJ-1
PR-1, SP-6, SP-5, RJ-1
RJ-1, ES, MG-2
BA-1
PR-1, SP-6, SP-5, SP-4, SP-3, RJ-1,
MG-2
SP-3, BA-2, ES
BA-4, BA-3, BA-2
SP-6, SP-5, SP-3, BA-4, BA-3, BA-2,
BA-1
SP-6, SP-3
T. sprengeliana Klotzsch ex Mez
T. stricta Sol. ex Sims
T. tenuifolia L.
T. tricholepis Baker
T. usneoides (L.) L.
Vriesea altodaserrae L.B. Sm.
V. altomacaensis A. Costa
V. arachnoidea A. Costa
V. atra Mez
V. bituminosa Wawra
V. blackburniana Leme
V. carinata Wawra
V. drepanocarpa (Baker) Mez
V. duvaliana E. Morren
V. ensiformis (Vell.) Beer
V. erythrodactylon (E. Morren) E.
Morren ex Mez
V. flammea L.B. Sm.
V. friburgensis Mez
V. gamba F. Mull.
V. gigantea Mart. ex Schult. f.
V. gracilior (L.B. Sm.) Leme
V. guttata Linden & André
V. heterostachys (Baker) L.B. Sm.
V. hieroglyphica (Carrière) E.
Morren
V. hydrophora Ule
V. incurvata Gaudich.
V. inflata (Wawra) Wawra
V. longicaulis (Baker) Mez
V. longiscapa Ule
V. minuta Leme
V. modesta Mez
V. paludosa L.B. Sm.
V. paratiensis E. Pereira
V. pauperrima E. Pereira
V. penduliflora L.B. Sm.
V. philippocoburgii Wawra
V. platynema Gaudich.
V. platzmannii E. Morren
V. procera (Mart. ex Schult. f.)
Wittm.
V. psittacina (Hook.) Lindl.
V. regina (Vell.) Beer
V. reitzii Leme & A. Costa
V. rhodostachys L.B. Sm.
V. rodigasiana E. Morren
V. ruschii L.B. Sm.
V. scalaris E. Morren
V. simplex (Vell.) Beer
V. sparsiflora L.B. Sm.
V. sucrei L.B. Sm. & Read
V. tijucana E. Pereira
V. triligulata Mez
V. tucumanensis Mez
V. vagans (L.B. Sm.) L.B. Sm.
Witrrockia gigantea (Baker) Leme
SC-1, PR-1, SP-6, SP-5, SP-4, RJ-1,
BA-4, BA-3, BA-2
RS-4, RS-3, RS-1, PR-3, PR-1, SP-4,
SP-3, MG-2
BA-3
RS-4, RS-3, PR-1, SP-6, SP-5, SP-4,
MG-1
ES, MG-3, MG-1, BA-2
MG-2
SP-6, RJ-1, MG-2
SP-4, RJ-1
SP-4, RJ-1
SC-1, SP-6, SP-5, SP-4, SP-3
SP-6, SP-3, RJ-1
RJ-1, ES, MG-2, BA-3
SP-3, RJ-1, BA-3
BA-4, BA-2, BA-1
MG-3
SP-6
RJ-1, MG-3, BA-4, BA-3
MG-1
MG-2
RS-4, SC-1, PR-1, SP-5, SP-4, SP-3,
RJ-1
PR-3, SP-4, BA-2
RS-4, PR-1, SP-6
RS-3, PR-1, RJ-2, BA-2
RS-3, ES, BA-2, BA-1
BA-3
PR-2
BA-1
PR-1, SP-6, SP-3, BA-3
BA-3, BA-4
SP-3, MG-1
ES, BA-4, BA-2, BA-1
RJ-1
RJ-2
BA-2
RJ-1
RS-5
RS-4, SC-1, PR-1, SP-6, SP-5, SP-4,
SP-3, RJ-1, MG-3
MG-2
671
References
Alves, R.J.V., Kolbek, J., Becker, J., 2008. Vascular epiphyte vegetation in rocky savannas of southeastern Brazil. Nord. J. Bot. 26, 101–117.
Amorim, A.M., Fiaschi, P., Jardim, J.G., Thomas, W.W., Clifton, B.C., Carvalho, A.M.V.,
2005. The vascular plants of a forest fragment in southeastern Bahia, Brazil. Sida
21, 1727–1752.
Amorim, A.M.A., Thomas, W.W., Carvalho, A.M.V., Jardim, J.G., 2008. Floristics of
the Una Biological Reserve, Bahia, Brasil. Mem. New York Bot. Gard. 100,
67–146.
Amorim, A.M., Jardim, J.G., Lopes, M.M.M., Fiaschi, P., Borges, R.A., Perdiz, R.O.,
Thomas, W.W., 2009. Angiospermas em remanescentes de floresta montana no
sul da Bahia, Brasil. Biota Neotrop. 9, 313–348.
Barfuss, M.H.J., Samuel, R., Till, W., Stuessy, T.F., 2005. Phylogenetic relationships
in subfamily Tillandsioideae (Bromeliaceae) based on DNA sequence data from
seven plastid regions. Am. J. Bot. 92, 337–351.
Benzing, D.H., 1990. Vascular Epiphytes. General Biology and Related Biota. Cambridge University Press, Cambridge.
Benzing, D.H., 2000. Bromeliaceae. Profile of an adaptative radiation. Cambridge
University Press, Cambridge.
Bonnett, A., Queiroz, M.H., 2006. Estratificação vertical de bromélias epifíticas em
diferentes estádios sucessionais da Floresta Ombrófila Densa, Ilha de Santa Catarina, Santa Catarina, Brasil. Rev. Bras. Bot. 29, 217–228.
Breier, T.B., 2005. O epifitismo vascular em florestas do sudeste do Brasil. PhD Thesis.
Universidade Estadual de Campinas, Campinas.
Buzatto, C.R., Severo, B.M.A., Waechter, J.L., 2008. Composição florística e
distribuição ecológica de epífitos vasculares na Floresta Nacional de Passo Fundo,
Rio Grande do Sul. Iheringia Ser. Bot. 63, 231–239.
Cody, M.L., 1993. Bird diversity components within and between habitats in
Australia. In: Ricklefs, R.E., Schluter, D. (Eds.), Species Diversity in Ecological
Communities. Historical and Geographical Perspectives. Chicago Press, Chicago,
pp. 147–158.
Costa, A.F., Wendt, T., 2007. Bromeliaceae na região de Macaé de Cima, Nova
Friburgo, Rio de Janeiro, Brasil. Rodriguésia 58, 905–939.
Fiaschi, P., Pirani, J.R., 2009. Review of plant biogeographic studies in Brazil. J. Syst.
Evol. 47, 477–496.
Fischer, A.E., Araújo, A.C., 1995. Spatial organization of a bromeliad community in
the Atlantic rainforest, south-eastern Brazil. J. Trop. Ecol. 11, 559–567.
Fontoura, T., Reinert, F., 2009. Habitat utilization and CAM occurrence among epiphytic bromeliads in a dry forest from Southeastern Brazil. Rev. Bras. Bot. 32,
519–528.
Fontoura, T., Santos, F.A.M., 2010. Geographic distribution of epiphytic bromeliads
of the Una region, Northeastern Brazil. Biota Neotrop. 10, 127–132.
Fontoura, T., Costa, A., Wendt, T., 1991. Preliminary checklist of the Bromeliaceae of
Rio de Janeiro State, Brazil. Selbyana 12, 5–45.
Fontoura, T., Sylvestre, L.S., Vaz, A.M.F., Vieira, C.M., 1997. Epífitas vasculares,
hemiepífitas e hemiparasitas da Reserva Ecológica de Macaé de Cima. In: Lima,
H.C., Guedes-Bruni, R.R. (Eds.), Serra de Macaé de Cima: diversidade florística e
conservação em Mata Atlântica. Jardim Botânico do Rio de Janeiro, Rio de Janeiro,
pp. 89–102.
Gentry, A., Dodson, C.H., 1987. Diversity and biogeography of Neotropical vascular
epiphytes. Ann. Missouri Bot. Gard. 74, 205–233.
Giongo, C., Waechter, J.L., 2004. Composição florística e estrutura comunitária de epífitos vasculares em uma floresta de galeria na Depressão Central do Rio Grande
do Sul. Rev. Bras. Bot. 27, 563–572.
Givnish, T.J., Millam, K.C., Berry, P.E., Sytsma, K.J., 2007. Phylogeny, adaptive radiation, and historical biogeography of Bromeliaceae inferred from ndhF sequence
data. In: Columbus, J.T., Friar, E.A., Porter, J.M., Prince, L.M., Simpson, M.G.
(Eds.), Monocots: Comparative Biology and Evolution – Poales. Rancho Santa
Ana Botanic Garden, Claremont, pp. 3–26.
Givnish, T.J., et al., 2011. Phylogeny, adaptive radiation, and historical biogeography
in Bromeliaceae: insights from an eight-locus plastid phylogeny. Am. J. Bot. 98,
872–895.
Gonçalves, C.N., Waechter, J.L., 2002. Epífitos vasculares sobre espécimes de Ficus
organensis isolados no norte da planície costeira do Rio Grande do Sul: padrões
de abundância e distribuição. Acta Bot. Bras. 16, 429–441.
Gonçalves, C.N., Waechter, J.L., 2003. Aspectos florísticos e ecológicos de epífitos
vasculares sobre figueiras isoladas no norte da planície costeira do Rio Grande
do Sul. Acta Bot. Bras. 17, 89–100.
Haffer, J., 1987. Quaternary history of Tropical America. In: Whitmore, T.C., Prance,
G.T. (Eds.), Biogeography and Quaternary History in Tropical America. Clarendon
Press, Oxford, pp. 1–16.
Hertel, R.J.G., 1949. Contribuição à ecologia da Serra do Mar (vertente oeste) do
Paraná. Tese de Livre Docência. Universidade do Paraná, Curitiba.
Ibisch, P.L., Nowicki, C., Vasquez, R., 2001. Towards an understanding of diversity patterns and conservation requirements of the Bolivian Bromeliaceae. J. Bromeliad
Soc. 51, 99–112.
Jones, H.G., 1992. Plants and Microclimate, 2nd ed. Cambridge University Press,
Cambridge.
Kersten, R.A., 2010. Epífitas vasculares – Histórico, participação taxonômica e aspectos relevantes, com ênfase na Mata Atlântica. Hoehnea 37, 9–38.
Kersten, R.A., Silva, M.S., 2001. Composição florística e estrutura do componente
epifítico vascular em floresta de planície litorânea na Ilha do Mel, Paraná, Brasil.
Rev. Bras. Bot. 24, 213–226.
672
T. Fontoura et al. / Flora 207 (2012) 662–672
Kersten, R.A., Silva, S.M., 2002. Florística e estrutura do componente epifítico vascular em Floresta Ombrófila Mista Aluvial do rio Barigüi, Paraná, Brasil. Rev. Bras.
Bot. 25, 259–267.
Kersten, R.A., Kuniyoshi, Y.S., Roderjan, C.V., 2009. Epífitas vasculares em duas
formações ribeirinhas adjacentes na bacia do rio Iguaçu – Terceiro Planalto
Paranaense. Iheringia Ser. Bot. 64, 33–43.
Kessler, M., 2000. Elevational gradients in species richness and endemism of selected
plant groups in the central Bolivian Andes. Plant Ecol. 149, 181–193.
Kessler, M., 2001. Patterns of diversity and range size of selected plant groups along
an elevational transect in the Bolivian Andes. Biodivers. Conserv. 10, 1897–1920.
Kessler, M., 2002a. Environmental patterns and ecological correlates of rangesize among bromeliad communities of Andean forests in Bolívia. Bot. Rev. 68,
100–127.
Kessler, M., 2002b. Species richness and ecophysiological type among Bolivian
bromeliad communities. Biodivers. Conserv. 11, 987–1010.
Küpper, W., Kreft, H., Nieder, J., Köster, N., Barthlott, W., 2004. Large-scale diversity
patterns of vascular epiphytes in Neotropical montane rain forests. J. Biogeogr.
31, 1477–1487.
Krömer, T., Kessler, M., 2005. Diversity patterns of vascular epiphytes along an elevational gradient in the Andes. J. Biogeogr. 32, 1799–1809.
Laube, S., Zotz, G., 2006. Neither host-specific nor random: vascular epiphytes on
three tree species in a Panamanian Lowland Forest. Ann. Bot. 97, 1103–1114.
Leme, E.M.C., Siqueira-Filho, J.A., 2006a. Taxonomia das bromélias dos fragmentos
de Mata Atlântica de Pernambuco e Alagoas. In: Siqueira-Filho, J.A., Leme, E.M.C.
(Eds.), Fragmentos de Mata Atlântica do nordeste. Biodiversidade, conservação
e suas bromélias. Andrea Jakobsson Estúdio, Rio de Janeiro, pp. 191–382.
Leme, E.M.C., Siqueira-Filho, J.A., 2006b. Considerações taxonômicas correlacionadas
às bromélias de Pernambuco e Alagoas. In: Siqueira-Filho, J.A., Leme, E.M.C.
(Eds.), Fragmentos de Mata Atlântica do nordeste. Biodiversidade, conservação
e suas bromélias. Andrea Jakobsson Estúdio, Rio de Janeiro, pp. 383–407.
Marques, M.C.M., Swaine, M.D., Liebsch, D., 2010. Diversity, distribution, and floristic
differentiation of the coastal lowland vegetation: implications for the conservation of the Brazilian Atlantic Forest. Biodivers. Conserv. 20, 153–168.
Martinelli, G., Vieira, C.M., Gonzalez, M., Leitman, P., Piratininga, A., Costa, A.F.,
Forzza, R.C., 2008. Bromeliaceae da Mata Atlântica brasileira: lista de espécies,
distribuição e conservação. Rodriguésia 59, 209–258.
McCune, B., Mefford, M.J., 1997. Multivariate Analysis of Ecological Data, ver. 3.11.
MjM Software, Oregon.
Menini-Neto, L., Forzza, R.C., Zappi, D., 2009a. Angiosperm epiphytes as conservation
indicators in forest fragments: a case study from southeastern Minas Gerais,
Brazil. Biodivers. Conserv. 18, 3785–3807.
Menini-Neto, L., et al., 2009b. Flora vascular não-arbórea de uma floresta de grota
na Serra da Mantiqueira, Zona da Mata de Minas Gerais, Brasil. Biota Neotrop. 9,
149–161.
Oliveira, M.L.A.A., Balbueno, R.A., Senna, R.M., 2005. Levantamento florístico de fragmentos florestais na bacia hidrográfica do rio Gravataí, Rio Grande do Sul, Brasil.
Iheringia Ser. Bot. 60, 269–284.
Oliveira-Filho, A.T., Fontes, M.A., 2000. Patterns of floristic differentiation among
Atlantic Forests in Southeastern Brazil and the influence of climate. Biotropica
32, 793–810.
Oliveira-Filho, A.T., et al., 2005. Análise florística do compartimento arbóreo de áreas
de Floresta Atlântica sensu lato na região das bacias do leste (Bahia, Minas Gerais,
Espírito Santo e Rio de Janeiro). Rodriguésia 56, 185–235.
Pontes, R.A.S., Agra, M.F., 2006. Flora da Paraíba, Brasil: Tillandsia L. (Bromeliaceae).
Rodriguésia 57, 47–61.
Reis, J.R.M., Fontoura, T., 2009. Diversidade de bromélias epífitas na Reserva Particular do Patrimônio Natural Serra do Teimoso – Jussari, BA. Biota Neotrop. 9,
73–79.
Rizzini, C.T., 1992. Tratado de fitogeografia do Brasil, 2a . ed. Âmbito Cultura Edições
Ltda, Rio de Janeiro.
Rogalski, J.M., Zanin, E.M., 2003. Composição florística de epífitos vasculares no
estreito de Augusto César, Floresta Estacional Decidual do Rio Uruguai, RS, Brasil.
Rev. Bras. Bot. 26, 551–556.
Safford, H.D., 2007. Brazilian Páramos IV. Phytogeography of the campos de altitude.
J. Biogeogr. 34, 1701–1722.
Santos, C.G., 2000. Distribuição espacial, fenologia e polinização de Bromeliaceae
na Mata Atlântica do alto da Serra de Paranapiacaba. PhD Thesis. Universidade
Estadual de Campinas, Campinas.
Sass, C., Specht, C.D., 2010. Phylogenetic estimation of the core Bromelioids with
an emphasis on the genus Aechmea (Bromeliaceae). Mol. Phylogenet. Evol. 55,
559–571.
Schulte, K., Horres, R., Zizka, G., 2005. Molecular phylogeny of Bromeliaceae and
its implications on biogeography and evolution of CAM in the family (Poales,
Bromeliaceae). Senckenbergiana Biol. 85, 113–125.
Schulte, K., Barfuss, M.H.J., Zizka, G., 2009. Phylogeny of Bromelioideae
(Bromeliaceae) inferred from nuclear and plastid DNA loci reveals the evolution of the tank habit within the subfamily. Mol. Phylogenet. Evol. 51,
327–339.
Scudeller, V.V., Martins, F.R., Shepherd, J.G., 2001. Distribution and abundance of
arboreal species in the Atlantic Ombrophilous Dense Forest in Southeastern
Brazil. Plant Ecol. 152, 185–199.
Shepherd, J., 1995. Fitopac. Manual do Usuário. Universidade Estadual de Campinas,
Campinas.
Smith, L.B., Downs, R.J., 1974. Pitcairnioideae (Bromeliaceae). Fl. Neotrop. Monogr.
14, 1–658.
Smith, L.B., Downs, R.J., 1977. Bromeliaceae (Tillandsioideae). Fl. Neotrop. Monogr.
14, 663–1492.
Sneath, P.H.A., Sokal, R.R., 1973. Numerical Taxonomy. The Principles and Practice
of Numerical Classification. W.H. Freeman and Company, San Francisco.
Sousa, G.M., Wanderley, M.G.L., 2000. Aechmea Ruiz & Pav. (Bromeliaceae) do estado
de Pernambuco, Brasil. Acta Bot. Bras. 14, 77–97.
Stehmann, J.R., Forzza, R.C., Sobral, M., Kamino, L.H.Y., 2009. Gimnospermas e
Angiospermas. In: Stehmann, J.R., Forzza, R.C., Salino, A., Sobral, M., Costa, D.P.,
Kamino, L.H.Y. (Eds.), Plantas da Floresta Atlântica. Jardim Botânico do Rio de
Janeiro, Rio de Janeiro, pp. 27–40.
Ter Braak, C.F.J., 1997. Ordination. In: Jongman, R.H.G., Ter Braak, C.J.F., van Tongeren,
O.F.R. (Eds.), Data Analysis in Community and Landscape Ecology. Cambridge
University Press, New York, pp. 91–173.
Varassin, I.G., 2002. Estrutura especial e temporal de uma comunidade de Bromeliaceae e seus polinizadores em Floresta Atlântica no sudeste do Brasil. PhD Thesis.
Universidade Estadual de Campinas, Campinas.
Versieux, L.M., Wendt, T., 2006. Checklist of Bromeliaceae of Minas Gerais, Brazil,
with notes on taxonomy and endemism. Selbyana 27, 107–146.
Versieux, L.M., Wendt, T., 2007. Bromeliaceae diversity and conservation in Minas
Gerais, Brazil. Biodivers. Conserv. 16, 2989–3009.
Waechter, J.L., 1998. Epifitismo vascular em uma floresta de restinga do Brasil subtropical. Ciênc. Nat. 20, 43–66.
Wanderley, M.G.L., Martins, S.E., 2007. Bromeliaceae: coordenação, descrição da
família e chave de gêneros. In: Wanderley, M.G.L., Shepherd, G.J., Melhem, T.S.,
Giulietti, A.M. (Eds.), Flora Fanerogâmica do Estado de São Paulo, vol. 5. FAPESP,
Instituto de Botânica, São Paulo, pp. 39–161.
Whitmore, T.C., Prance, G.T., 1987. Biogeography and Quaternary History of Tropical
America. Clarendon Press, Oxford.
Wolf, J.H.D., Flamenco-S, A., 2003. Patterns in species richness and distribution of
vascular epiphytes in Chiapas, México. J. Biogeogr. 30, 1689–1707.