P
n
B

D
N
a

b

c

B
d

2

a

A
R
R
A
E
A

K
C
I
L
S

1

i
fl
e
1
B
i
s
s
s

h
0

Flora 223 (2016) 19–29

Contents lists available at ScienceDirect

Flora

j o ur na l ho me  page: www.elsev ier .com/ locate / f lora

ollination  biology  and  breeding  system  of  syntopic  Adenocalymma
odosum  and  A.  peregrinum  (Bignonieae,  Bignoniaceae)  in  the
razilian  savanna

iana  S.  Sampaioa,  Clesnan  Mendes-Rodriguesa, Thaíssa  B.J.  Engelb,  Tiago  M.  Rezendec,
elson  S.  Bittencourt-Jrd, Paulo  Eugênio  Oliveiraa,∗

Universidade Federal de Uberlândia, Campus Umuarama, Bloco 2D, Instituto de Biologia, 38405-320, Uberlândia, Minas Gerais, Brazil
Universidade Estadual de Campinas, Instituto de Biologia, Programa de Pós-Graduaç ão em Biologia Vegetal, 13083-970, Campinas, São Paulo, Brazil
Instituto Chico Mendes de Conservaç ão da Biodiversidade − ICMBio, EQSW 103/104, Bloco A, Complexo Administrativo, Setor Sudoeste, 70670-350,
rasília, Distrito Federal, Brazil
Universidade Estadual Paulista “Júlio de Mesquita Filho”, Campus São José do Rio Preto, Departamento de Zoologia e Botânica, Rua Cristóvão Colombo,
265,  Jardim Nazareth, São José do Rio Preto, 15054-000, São Paulo, Brazil

 r  t  i  c  l  e  i  n  f  o

rticle history:
eceived 16 November 2015
eceived in revised form 1 April 2016
ccepted 15 April 2016
dited by Stefan Dötterl
vailable online 20 April 2016

eywords:
errado

nterspecific compatibility
ate-acting self-incompatibility

a  b  s  t  r  a  c  t

The  tropical  Bignoniaceae  is  mostly  late-acting  self-incompatible  (LSI)  and  depends  on a guild of  medium
to  large  sized  bees  for their  pollination.  Adenocalymma  nodosum  and  A.  peregrinum  are  syntopic  shrubs
in  savanna  areas  with  similar  flowers  and  flowering  overlap.  In this  sense,  the aims  of  the  present  study
were  to  analyse  the pollination  biology  and  breeding  systems  of  these  species,  and  to check  for  bilateral
compatibility  which  could  hinder  reproductive  isolation  and  species  boundaries.  Flower  features  such  as
yellow corolla,  sweet  scent  and  diurnal  one-day  anthesis  were  similar  between  species.  However,  they
differed  in  nectar  production  patterns,  which  showed  higher  volume  in A.  nodosum  and  was  irregular  in
A. peregrinum.  The  main  pollinators  for  both  species  were  medium  to large  Centridini  bees.  Many  nectar
and  pollen  robbers  may  disturb  effective  pollination  and  help  to explain  the  low  natural  fruit-set.  The
species  were  self-sterile  but  pollen  tube  growth  down  to the  ovules  and  differential  ovary  development
ynchronopatry indicated  LSI,  as in  most  Bignoniaceae  studied  to  date,  which  reinforce  the  idea  of  family  clustering  for
this  self-incompatibility  system.  Fruit-set  from  interspecific  hand  pollinations  was  similar  to  those  of
intraspecific  cross-pollinations,  with  high  seed  viability  (88%)  and  seedling  development.  Similar  floral
biology  and  guild  of  pollinators,  and  bilateral  inter-compatibility  indicate  that  natural  hybridization  is
possible  among  these  species.

©  2016  Elsevier  GmbH.  All  rights  reserved.
. Introduction

Self-sterility mechanisms are important for co-sexual flower-
ng plants in order to avoid inbreeding and assure adequate pollen
ow (e.g. Neuffer and Paetsch, 2013). The Bignoniaceae is consid-
red predominantly self-sterile (e.g. Gibbs, 1990; Gibbs and Bianchi,
993, 1999; Bittencourt and Semir, 2005, 2006; Gandolphi and
ittencourt, 2010; Bittencourt et al., 2011). Among the 58 species

nvestigated for breeding systems, 49 are self-sterile and nine are

elf-fertile, but seven species described as self-sterile also show
elf-fertile populations. In this sense, self-fertility was  found in 16
pecies (Bittencourt and Moraes, 2010; Firetti-Leggieri et al., 2013;

∗ Corresponding author.
E-mail address: poliveira@ufu.br (P. Eugênio Oliveira).

ttp://dx.doi.org/10.1016/j.flora.2016.04.009
367-2530/© 2016 Elsevier GmbH. All rights reserved.
Sampaio et al., 2013; Alves in press). It seems that self-fertility has
evolved independently in the tribe Tecomeae, Tabebuia Alliance,
and Bignonieae clades (sensu Olmstead et al., 2009), and the pres-
ence of more than one kind of breeding system in a single species
was recorded in the latter two  clades.

Among the self-sterile species of Bignoniaceae, 34 have post-
pollination events studied. These investigations show that the
self-pollen tubes grow to the ovary, penetrate and fertilize ovules,
which are able to develop an initial endosperm and a proembry-
onal tube with two  nuclei (e.g. Gibbs, 1990; Gibbs and Bianchi,
1993, 1999; Bittencourt and Semir, 2005, 2006; Gandolphi and Bit-
tencourt, 2010; Bittencourt et al., 2011). Those aspects correspond

to a late-acting self-incompatibility system, LSI (Seavey and Bawa,
1986; Bittencourt et al., 2003; Gibbs, 2014), which is commonly
neglect due to its similarity to early inbreeding depression, espe-
cially in cases like those observed in Bignoniaceae where rejection

dx.doi.org/10.1016/j.flora.2016.04.009
http://www.sciencedirect.com/science/journal/03672530
http://www.elsevier.com/locate/flora
http://crossmark.crossref.org/dialog/?doi=10.1016/j.flora.2016.04.009&domain=pdf
mailto:poliveira@ufu.br
dx.doi.org/10.1016/j.flora.2016.04.009


2 / Flora

o
f
u
t
m
a
t
B
a

u
(
E
M
S
c
p
a
t
s
1
t
Y

f
(
(
s
(
a
c
T
w
p
2
b
A
A
u
A
t
o

o
i
o
d
o
s
t
t

2

2

b
a
l
l
a
0
b
D

0 D.S. Sampaio et al. 

f selfed pistils is predominantly post-zygotic (Gibbs, 2014). The
acts that LSI shows family clustering, that in Bignoniaceae it is the
nique self-incompatibility system (SI) described until now, and
hat the endosperm and proembryonal tube do not present any

alfunction in self-fertilized ovules, are strong arguments for the
cceptance of LSI (Bittencourt et al., 2003; Gibbs, 2014). The inves-
igation of the breeding system of a higher number of species of
ignoniaceae will help to understand its evolution in the group,
nd to reinforce LSI as a genetic controlled SI.

Bignoniaceae shows a high diversity within communities, with
p to 20 species per area, and 75% of species are pollinated by bees
Gentry, 1976, 1980, 1990). Medium and large sized Centridini and
uglossini bees are the most important pollinators (Ynagizawa and
aimoni-Rodella, 2007; Guimarães et al., 2008; Milet-Pinheiro and

chlindwein, 2009; Almeida-Soares et al., 2010). These bees are
apable of flying long distances, enabling the occurrence of cross-
ollinations (Gottsberger and Silberbauer-Gottsberber, 2006), but
lso the occurrence of interspecific pollination. On the other hand,
he great diversity of floral morphology, flowering phenology and
easonality of Bignoniaceae species in a given community (Gentry,
974b) avoid competition for pollinators and maintain reproduc-
ive isolation among congeners (Gentry, 1976, 1980; Amaral, 1992;
nagizawa and Maimoni-Rodella, 2007; Zjhra, 2008).

There are only two Adenocalymma species already studied
or their pollination biology and breeding system, A. marginatum
Cham.) DC. and A. bracteatum (Cham.) DC., both described with LSI
Amaral, 1992), though the later species was recently reported as
elf-fertile (Almeida-Soares et al., 2010). Adenocalymma nodosum
Silva Manso) L. Lohmann and A. peregrinum (Miers) L. Lohmann
re closely related species of shrubs from Cerrado, the Neotropi-
al savanna region in Central Brazil (Lohmann and Taylor, 2014).
hese species were previously placed in the genus Memora, which
as recently found to be nested inside Adenocalymma as a mono-
hyletic group of species (Lohmann, 2006; Lohmann and Taylor,
014). There are no data available about pollination biology and
reeding system for species of the “Memora clade” of the genus
denocalymma. In Uberlândia, Minas Gerais State, A. nodosum and
. peregrinum are syntopic, and although their flowering seasons
sually differ (A. peregrinum flowers from December to April and
. nodosum from March to June) and may  function as a reproduc-
ive barrier among these species, some flowering overlap has been
bserved (D. Sampaio, personal observations).

As the “Memora clade” has not been evaluated and more than
ne kind of breeding system was recorded for A. bracteatum, we
nvestigated the flower biology, pollination, and breeding system
f A. nodosum and A. peregrinum in order to understand the repro-
uctive ecology of these species and to contribute to the knowledge
f breeding system evolution in Bignoniaceae. Finally, since these
pecies occur in syntopic populations with flowering overlap, we
ested for inter-compatibility, which can affect reproductive isola-
ion and species boundaries.

. Materials and methods

.1. Studied species and study sites

Adenocalymma nodosum is a caespitose shrub with flexible
ranches up to 2 m high. Its leaves are petiolate, bi or tripinnate
nd imparipinnate (Fig. 1A, B). Leaflets are narrow and lanceo-
ate, sessile when simple, and petiolate when they are compound
eaflets (Fig. 1A, B). Flowers arise in the axils of bracts. Bracteoles are

lso present and the flower pedicel is short, with a mean length of
.67 cm (SD ± 0.15; n = 15). The calyx tube is green with five lobes. It
ears nectaries and glandular trichomes that make it sticky (Fig. 1C,
). The corolla (Fig. 1A, E) is golden yellow and glabrous. The pollen
 223 (2016) 19–29

grains are yellow when observed by the naked eye. The staminode
is inconspicuous (Fig. 1E), but in some individuals it shows a devel-
oped anther with pollen grains. Usually, the style rises the stigma
above the stamens (Fig. 1E), but occasionally it lies below the sta-
men  level, favouring spontaneous self-pollination. The flower has
a sweet scent and produces nectar on a disk in the ovary base. The
fruit is a septicide capsule dorsi-ventrally flattened (Fig. 1F) with
active nectaries and glandular trichomes when immature, similar
to those found in the calyx tube. Fruits reach maturity in three to
four months after pollination. The species occurs in Brazilian States
of Mato Grosso, Goiás, Minas Gerais, Rio de Janeiro, Tocantins and
Piauí (Lohmann, 2010a).

Adenocalymma peregrinum is also a caespitose shrub up to 1 m
high. The leaves are petiolate, have two or three leaflets or can also
be pinnate or bipinnate (Fig. 1G–J). When bipinnate, only the basal
leaflets are compound (Fig. 1I). They can be paripinnate or imparip-
innate. Leaflets are ovate, sessile when simple and petiolate when
compound (Fig. 1G–J). The mean length of the flower pedicel is
1.64 cm (SD ± 0.26; n = 15). The calyx is light green and during the
corolla emission it breaks, forming one or two longitudinal slits,
which make the calyx spathaceous or bilabiate with irregular edges
(Fig. 1G, K, L). Calyx and fruit show sparse lepidote trichomes, being
almost glabrous and without nectaries (Fig. 1L, M).  The pollen grains
are white when observed by the naked eye. Other features are simi-
lar of those described for A. nodosum. The species occurs in Brazilian
States of Mato Grosso, Goiás, Minas Gerais, Rio de Janeiro, Mato
Grosso do Sul, São Paulo and Paraná (Lohmann, 2010b). Adenoca-
lymma peregrinum is considered an important invasive species of
pastures (Nunes, 1999, 2001; Grassi et al., 2005).

Both species present hemixyle habit, re-sprouting from below-
ground structures each year, after fire or pruning (Gottsberger and
Silberbauer-Gottsberger, 2006). Thus, it was  often difficult to define
individuals since a cluster of plants may  contain sprouts (ramets)
from different genetic origins.

Fieldwork was carried out from 2006 to 2009 in campo sujo
areas, a plant formation of scattered shrubs on a grassland matrix
common in the Cerrado region (Oliveira-Filho and Ratter, 2002).
The main area was at the edge of the Clube de Caç a e Pesca
Itororó de Uberlândia preservation area (CCPIU) (18◦58′48.5”S,
48◦17′45.8”W), where the two  species are syntopic. Complemen-
tary studies were done also at Campo Alegre farm (18◦58′43.83”S,
48◦14′59.73”W), where only A. peregrinum occurs in another campo
sujo used as pastureland. The Campo Alegre farm is 5 km away
from the CCPIU area, and both are in the outskirts of Uberlândia,
Minas Gerais State, Brazil. The vouchers were deposited in Herbar-
ium Uberlandense-HUFU, Uberlândia, Minas Gerais State, Brazil,
under registration numbers: HUFU 45287 for A. nodosum, and HUFU
44908, 48925 − CCPIU, and HUFU 47293 − Campo Alegre farm, for
A. peregrinum.

2.2. Floral biology

In A. nodosum, 47 floral buds in pre-anthesis were bagged in
10 clusters of plants at CCPIU and, in A. peregrinum, 10 buds in
pre-anthesis were bagged in five clusters at Campo Alegre farm.
Duration of anthesis, beginning of nectar and scent production,
stigmatic sensitivity, pollen presentation and anther color were
observed in the resulting flowers. Stigmatic sensitivity was tested
by touching the stigma lobes with tweezers, and it was  considered
sensitive if the two  lobes closed after being touched.

To estimate pollen viability (based on stainability and form)
ten and nine floral buds at the day before anthesis were collected,

respectively, from five clusters of plants of A. nodosum and five clus-
ters of A. peregrinum at CCPIU, and fixed in ethanol 70%. Slides were
mounted with the pollen grains of one anther of each flower, a drop
of aceto-carmine, and a drop of glycerol 50% (Dafni et al., 2005). The



D.S. Sampaio et al. / Flora 223 (2016) 19–29 21

Fig. 1. Leaf, flower and fruit of Adenocalymma nodosum and A. peregrinum (drawings by Natanael Nascimento dos Santos). A–F, A. nodosum. A, Branch with leaf and flowers.
B,  Leaf. C, Calyx tube. D, Detail of glandular trichomes and extrafloral nectaries on the calyx tube. E, Corolla tube opened longitudinally to make the stamens and the pistil
v Branc
l omes
I

a
3
s
t

g
w

isible.  The staminode is indicated (arrow). F, Mature fruit. G–M, A. peregrinum. G, 

eaf.  K, Flower. L, Spathaceous calyx. M,  Detail of the calyx with sparse lepidote trich
).

nalysis was done using an optical microscope Olympus BX51 and
00 pollen grains were analysed per slide. Pollen grains reduced in
ize, without cytoplasm, or microspores that were still enclosed in

etrads were considered non-viable.

In order to establish the pollen/ovule ratio, the number of pollen
rains per flower was estimated and the number of ovules per ovary
as counted (Cruden, 1977). To estimate the number of pollen
h with trifoliolate leaves and flowers. H, Pinnate leaf. I, Bipinnate leaf. J, Bifoliolate
. Scale bars = 5 cm (A, B, G, J), 1 cm (C, L), 1 mm (D, M),  2 cm (E), 3 cm (F, K),  4 cm (H,

grains per flower, 16 pre-anthesis buds (i.e., collected a day before
flower opening) from eight clusters of plants of A. nodosum and
10 pre-anthesis buds from six clusters of A. peregrinum were col-

lected at CCPIU, and stored in ethanol 70%. The pollen grains of two
anthers were diluted in 1 ml  of glycerol 50% and the number of
pollen grains was estimated in a Neubauer Chamber (Dafni et al.,
2005) under optical microscope Olympus BX51. The counting of



2 / Flora

o
p
n
l
b
(

s
c
o
1
m
a
c
e
a
c
N
f
m
o
w
w
p
f
c
t
W
b
i
S

2

o
o
2
J
v
o

a
h
i
n
t
n
c
t
e
o
a
c

l
t
v
A
p
o

2

s

2 D.S. Sampaio et al. 

vules per ovary was performed under a stereomicroscope Olym-
us SZX12 in 116 dissected ovaries from five clusters of plants of A.
odosum and in 113 ovaries from nine clusters of A. peregrinum col-

ected at CCPIU. Whenever necessary, parameters were compared
etween species using Student t-test for heterogeneous variances
Zar, 2010).

To analyse nectar features, 36 buds in pre-anthesis for each
pecies from five clusters of plants of A. nodosum and from nine
lusters of A. peregrinum were bagged at CCPIU. During the first day
f anthesis, nectar was measured in three time intervals (10:00 h,
3:30 h, 16:30 h) using different flowers per interval. We  used 10 �l
icropipettes to estimate the amount of accumulated nectar, and

 handheld refractometer (Eclipse − scale of 1%) to estimate nectar
oncentration of soluble sugars (Kearns and Inouye, 1993). Flow-
rs without nectar and/or damaged were not included in statistical
nalysis because they presented some damage in the base of the
alyx and in the nectary that was probably done before bagging.
ectar data for A. nodosum were analysed using a one-way ANOVA

or volume and another one for concentration, after confirming nor-
ality for the residuals using Shapiro-Wilk test and homogeneity

f variance with Levene test at significance level of 0.05. The means
ere compared pairwise with Tukey test, and the data of volume
ere log10 transformed before statistical analyses. In order to com-
are nectar features in A. peregrinum, we used a Mann-Whitney test
or volume due to absence of normality for each interval. Since con-
entration data showed normality in each group, we used a Student
-test to compare intervals. The normality was tested with Shapiro-

ilk test. We performed the analyses separately for each species
ecause some of the A. peregrinum flowers did not present nectar

n the third interval of analysis. The analyses were carried out using
PSS 17.0 (SPSS, Inc., Chicago, IL, USA).

.3. Floral visitors

The floral visitors’ study was performed on 10 clusters of plants
f A. nodosum at CCPIU during nine days in April of 2006, a total
f 29 h and 15 min  of observation. In A. peregrinum, we studied
2 clusters of plants at Campo Alegre farm during eight days in
anuary of 2007, a total of 20 h and 19 min  of observation. The obser-
ations were done from 8:00 h to 17:00 h, alternating the time of
bservation among the clusters of plants.

Bee’s behaviour and number of visits per flower were recorded
nd the visitation frequency was calculated (number of visits per
our). Bees that did not enter the corolla tube, but did a perforation

n its base (or even use perforations that already existed) to get
ectar was considered nectar robbers. Small sized bees that entered
he corolla tube to get pollen or nectar were considered pollen or
ectar robbers, respectively. Those small sized bees can damage the
orolla tube in pre-anthesis buds to get pollen or nectar, or enter
he corolla during anthesis causing no damage. Even when they
ntered the corolla tube they seldom touched the stigma. Medium
r large sized bees that entered the corolla tube searching for nectar
nd contacted the stigma and the anthers during these visits were
onsidered effective pollinators.

Only the first individual of each bee species observed was  col-
ected and pinned. After identified, those individuals were taken to
he field and used as material for comparison with the following
isitors, which were captured, identified, and released in the field.
fter some training, most species could be identified in the field. The
inned visitors were incorporated to the entomological collection
f Instituto de Biologia of Universidade Federal de Uberlândia.
.4. Breeding system and interspecific pollination

As it was often difficult to define individuals for both species, we
tudied more or less isolated clusters of plants, whereas the number
 223 (2016) 19–29

of ramets per cluster varied from three to 21 for A. nodosum, and
from eight to 21 for A. peregrinum at CCPIU. At least 28 clusters
of plants of the latter species were also studied at Campo Ale-
gre farm in 2007. Due to the possiblity that a cluster is made of
different individuals, we used flowers as sample units and tried
to keep the number of flowers used for the experiments similar
among the clusters. We  also brought pollen for cross-pollination
from a safe distance to avoid the use of ramets as different indi-
viduals. Numbers of flowers used in each treatment are presented
in the results (see Table 3). Flower buds were bagged with nylon
mesh to exclude pollinators and other floral visitors. Self- and cross-
hand pollinations were carried out in first-day flowers emascu-
lated early the day, before pollen was  released. In self-pollinations,
we used the pollen grains of the same flower. In cross-pollinations
and interspecific pollinations we used pollen from clearly distinct
clusters of plants at least ten meters away and from clusters of
plants of the other species, respectively. Bagged buds were marked
to check for spontaneous self-pollination and others were emas-
culated before isolation to check for autonomous apomixis (for
numbers, see Table 3). Natural fruit set was estimated by moni-
toring marked, non-bagged flowers. The self-incompatibility index
(ISI) was determined as the ratio of hand self-pollination/cross-
pollination fruit-set (Bullock, 1985). The ratio of natural/hand
cross-pollination fruit-set was  also determined as an estimate of
reproductive efficacy (RE) (Ruiz and Arroyo, 1978).

In situ pollen tube growth was analysed to verify if the self-
pollen tubes reach the ovary and penetrate ovules as an indication
of the presence of LSI in the case of self-sterile species. Pistils from
hand self- and cross-pollinations were collected 24 h after polli-
nation and stored in ethanol 70%. Six pistils per treatment were
collected in A. nodosum, and five pistils per treatment were col-
lected in A. peregrinum in five clusters of plants of each species at
CCPIU. The ovary wall was  removed to expose the ovules. Pistils
were softened in 9 M NaOH solution at 55 ◦C for 15 min, washed in
flowing water, and stained with aniline blue (adapted from Martin,
1959). The pistils were analysed under optical microscope Zeiss
Axioscop equipped with epi-fluorescence.

In order to verify if self-pollinated pistils showed ovary expan-
sion before being aborted, and if this expansion differed from the
expansion observed in ovaries of cross-pollinated pistils, we took
measurements of ovary length and width. Unpollinated pistils were
used as a control group. Measurements of ovary length and width
from hand self- and cross-pollinated pistils collected 24, 48, 72,
96 and 120 h after pollination and of ovaries of unpollinated pis-
tils from emasculated flowers collected 24, 48, 72, 96 and 120 h
after the onset of anthesis were carried out with digital calliper
Digimess (300 mm/12′ - 0.01). Five to ten pistils were measured
per treatment and per time interval in A. nodosum and four to
eight pistils in A. peregrinum collected from five clusters of plants
of each species at CCPIU. All pistils were fixed in a 1% glutaralde-
hyde plus 4% formaldehyde solution (McDowell and Trump, 1976)
in sodium phosphate buffer 0.1 M,  pH 7.2 before measurements.
In order to check if there was any growth in ovaries along the
time intervals, and if there were differences in growth rates among
pollination treatments, we calculated the area of the ovaries (esti-
mated by: area = length × width). Although robust, data for ovary
area did not fit the assumptions for ANOVA analyses (normality
from residual and homogeneity of variances) and we therefore used
Generalized Linear Models with Gaussian distribution. For main
level post hoc comparisons of the pollination treatments we  used
the Least Significant Difference test and the time intervals were
adjusted to polynomial models of different orders whenever nec-

essary. The GLM was  performed separately to each species. As for
both species the interactions were statistically significant, we  per-
formed a GLM for time interval comparisons for each pollination
treatment independently and separately for each species, and for



D.S. Sampaio et al. / Flora 223 (2016) 19–29 23

Table  1
Features of the floral biology of Adenocalymma nodosum (Silva Manso) L. Lohmann and A. peregrinum (Miers) L. Lohmann. Quantitative parameters followed by different
letters differed significantly between species (Student-t test for heterogeneous variance, P < 0.01).

Features Adenocalymma nodosum Adenocalymma peregrinum

Corolla opening 7:00–10:00 h 8:00–10:00 h
Scent  production From 9:00 h to the end of anthesis From 10:00 h to the end of anthesis
Nectar production Before 7:00 to 17:00 h Before 8:00 to 17:00 h
Stigmatic sensitivity Since the pre-anthesis Since the pre-anthesis
Pollen  presentation From 9:00 h From 9:00 h
Darkening of the anthers From 14:00 h From 13:30 h
Duration of anthesis About 24 h About 24 h
Pollen viability% (±DP) 95.5 (±2.9) a 77.8 (±6.7) b
Mean  number of pollen grains per flower (±DP) 41750 (±29896.5) b 64666 (±13185.9) a
Mean number of ovules per ovary (±DP) 10.0 (±3.4) b 14.1 (±2.5) a
Pollen/ovule ratio 4175.0 4599.3

Table 2
Frequency of visitation by bees to the flowers of Adenocalymma nodosum (Silva Manso) L. Lohmann at Clube de Caç a e Pesca Itororó de Uberlândia, and of A. peregrinum
(Miers)  L. Lohmann at Campo Alegre farm, both in Uberlândia City, Minas Gerais State, Brazil.

Species Behaviour Number of visits per hour (Total number of visits)

Adenocalymma
nodosum (total
29 h 15 min)

Adenocalymma
peregrinum (total
20 h 19 min)

Andrenidae
Oxaea flavescens Klug, 1807 NR 3.56 (104) 55.31 (1124)

Apidae
Apini  Apis mellifera Linnaeus, 1758 PR 1.40 (41) 0.05 (1)
Centridini Centris (Centris) aenea Lepeletier, 1841 EP 0.07 (2) –

Centris  (Trachina) fuscata Lepeletier,1841 EP 1.91 (56) –
Centris  (Hemisiella) tarsata Smith, 1874 EP 0.38 (11) –
Centris  (Heterocentris) analis Fabricius, 1804 EP 0.17 (5) –
Epicharis (Epicharis) bicolor Smith, 1874 EP 0.17 (5) 0.39 (8)
Epicharis (Epicharana) flava Friese, 1900 EP – 0.05 (1)
Centridini sp. 1 EP – 0.10 (2)

Emphorini Alepidoscelis imitatrix Schrottky, 1909 PR 0.14 (4) 0.15 (3)
Emphorini sp. 1 EP 0.15 (3)

Ericrocidini Mesoplia (Mesoplia) rufipes Perty, 1833 EP 0.21 (6) –
Mesoplia (Eumelissa) frisei Perty, 1833 EP 0.07 (2) –

Euglossini Euglossa (Euglossa) cordata Linnaeus, 1758 EP 1.16 (34) –
Euglossa (Euglossa) melanotricha Moure, 1967 EP 0.10 (3) 13.24 (269)
Eulaema (Apeulaema) nigrita Lepeletier, 1841 EP 0.41 (12) 0.05 (1)
Exarete sp. EP – 0.05 (1)

Exomalopsini Exomalopsis (Exomalopsis) fulvofasciata Smith, 1879 PR 2.02 (59) 0.49 (10)
Meliponini Friseomelitta sp. NR 0.03 (1) –

Paratrigona lineata Lepeletier, 1836 PR 0.10 (3) –
Trigona  spinipes Fabricius, 1793 NR/PR 13.91 (407) 4.68 (95)
Tetragonisca angustula Latreille, 1811 PR 0.03 (1) –

Tetrapedini Paratetrapedia sp. PR 0.34 (10) –
Xylocopini Ceratina (Crewella) sp. PR 1.09 (32) 2.02 (41)

Xylocopa sp. EP/NR 0.07 (2) 0.05 (1)
Rhathymini Rhatymus sp. EP/NR – 1.23 (25)

Halictidae
sp.  1 PR 0.03 (1) 0.05 (1)
sp.  2 PR – 0.05 (1)
sp.  3 PR – 0.44 (9)

Total  27.38 (801) 78.54 (1596)

Note: EP = effective pollinator; NR = nectar robber; PR = pollen robber (enter the corolla tube and may  also act as occasional pollinators). Xylocopa sp. behaved as an effective
pollinator in the flowers of A. nodosum and as a nectar robber in A. peregrinum.

Table 3
Experimental pollinations and fruit-set in Adenocalymma nodosum (Silva Manso) L. Lohmann and A. peregrinum (Miers) L. Lohmann at Clube de Caç a e Pesca Itororó de
Uberlândia (CCPIU) and at Campo Alegre farm, Uberlândia City, Minas Gerais State, Brazil.

Species/Population Adenocalymma nodosum
(CCPIU)

Adenocalymma peregrinum

Campo Alegre farm CCPIU Total

Experimental pollinations Number of
flowers

Number of
fruits (%)

Number of
flowers

Number of
fruits (%)

Number of
flowers

Number of
fruits (%)

Number of
flowers

Number of
fruits (%)

Hand self-pollination 95 0 (0) 61 0 (0) 142 2 (1.41) 203 2 (0.98)
Intraspecific hand cross-pollination 89 35 (39) 59 8 (13.56) 205 62 (30.24) 264 70 (26.51)
Spontaneous self-pollination 92 1 (1.09) 64 0 (0) 41 0 (0) 105 0 (0)
Emasculated flowers 84 0 (0) 73 0 (0) 40 0 (0) 113 0 (0)
Natural  pollination 81 5 (6) 62 0 (0) 77 2 (2.60) 139 2 (1.44)
Interspecific hand cross-pollination 114 40 (35.09) – – 63 24 (38.09) 63 24 (38.09)



2 / Flora

p
h
i
n
p
t
W
1
f
m

t
m
o
n
l
p
c
s
a
a
d
b
D
S
U

l
w
m
(
w
l
v
s
(
m

3

3

A
e
1
s
a
o
i
t
p
p
T
d
i
a
c
w
s
t

A
t
(

4 D.S. Sampaio et al. 

ost-hoc comparisons the data were adjusted to either linear or
igher order polynomial models. We  used this approach since a full

nteraction analysis would not be biologically meaningful and did
ot show any clear pattern. We  adjusted the data for ovary area to
olynomial models whenever previous analyses with others tradi-
ional post-hoc tests did not show biologically meaningful patterns.

e compared the ovary area between pollination treatments only
20 h after pollination, using in this case the Least Significant Dif-
erence test. The analyses were carried out using SPSS 17.0, GLzM

odule with test Type III (SPSS, Inc., Chicago, IL, USA).
Fruit maturation was monitored for three to four months after

he last experimental pollination. Developed fruits from experi-
ental and natural pollinations were collected and the number

f seeds per fruit was evaluated. In order to compare the mean
umber of seeds per fruit among treatments, we used generalized

inear models with Poisson loglinear distribution and post-hoc com-
arisons were made with Least Significant Difference. Seed-set was
alculated based on the mean ovule number per ovary and the mean
eed number per fruit. The presence or absence of an embryo was
lso recorded for seeds of each treatment. Comparisons of seed-set
nd the percentage of seeds with embryo among treatments were
one with generalized linear models using a binary logistic distri-
ution, and post-hoc comparisons were made with Least Significant
ifference. Those analyses were done with the statistical package
PSS 17.0, GLzM module with test Type III (SPSS, Inc., Chicago, IL,
SA).

Since our interspecific hand pollinations did set fruits, we ana-
ysed seed viability to verify if viable hybrids were formed. Seeds

ere sown in Gerbox® plastic boxes on cotton and filter paper
oistened with distilled water, and kept at room temperature

about 25 ◦C) and natural light. Seeds were considered germinated
hen the radicle protruded. Seedlings were transferred to cel-

ularized polystyrene trays, in a 1:1 mixture of Plantmax® and
ermiculite substrate (expansion volume of 0.1 m3) to observe the
ubsequent seedling development. Trays were kept in greenhouse
mean temperature about 25 ◦C) under natural light conditions and

oistened when necessary.

. Results

.1. Floral biology

Adenocalymma peregrinum flowered from December to April and
. nodosum from March to June, and both showed a few flow-
rs open per day (“steady state” flowering pattern sensu Gentry,
974a). The flowering periods of the species were similar in the
tudied areas. The overlap between flowering seasons of the species
t CCPIU was observed mostly in March of 2007 and 2008. This
verlap was observed during a month or even a bit more, when the
nterspecific hand pollinations were done. The floral events of the
wo species were similar, showing the corolla completely opened,
rofuse scent and nectar production, sensitive stigma, and pollen
resentation between 9:00 and 10:00 h in the morning (Table 1).
he anthers became dark and devoid of pollen grains, which fell
own in the corolla tube, about 13:00 and 14:00 h in the afternoon,

ndicating the end of the pollen presentation period (Table 1). The
nthesis lasted about 24 h, when the corolla withered and lost its
olor and scent (Table 1), although many corollas abscised before
ithering. Pollen viability estimates were high (78–95%) for both

pecies, but A. peregrinum presented about 20% less viable pollen
han A. nodosum at CCPIU (Table 1).
Nectar production patterns differed between species (Fig. 2).
denocalymma nodosum showed ca. 33 �l of accumulated nec-
ar with ca. 22% sugar concentration at the end of the afternoon
17:00 h) with an increase in nectar volume along the time inter-
 223 (2016) 19–29

vals, but no marked variation in nectar concentration (Fig. 2A, B).
A. peregrinum produced ca. 25 �l with a sugar concentration of ca.
19%. However, among the 36 flowers analysed for nectar features
in A. peregrinum, 13 (36%) did not produce nectar, and in the third
interval of analysis, we did not find any flowers presenting nectar.
Nevertheless, for the flowers that did have nectar, A. peregrinum
showed no significant variation in nectar volume and a significant
increase in nectar concentration from the first to the second time
interval (Fig. 2C, D). The coefficients of variation for A. nodosum
were 48.25%, 34.87% and 53.21% for nectar volume and 13.25%,
12.69% and 14.97% for nectar concentration in the 1st, 2nd and
3rd time intervals, respectively. While the coefficients of variation
for A. peregrinum were 75.43% and 90.01% for nectar volume and
65.03% and 41.87% for nectar concentration in the 1st and 2nd time
intervals, respectively.

3.2. Floral visitors

Adenocalymma peregrinum showed more than the double of
visits by bees at Campo Alegre farm than A. nodosum at CCPIU,
mostly due to the abundant visits of the nectar thieving bee Oxaea
flavescens to the first species (Table 2). The visitation frequency
was constant during the morning, declining in the end of the after-
noon for both species (data not shown). Nectar robbers showed the
highest frequency of visits for the two species, with 63.9% of vis-
its in A. nodosum and 77.8% in A. peregrinum (Table 2). In the first
species, the prevalent nectar robber was Trigona spinipes, and in
the second species, O. flavescens (Table 2, Fig. 3A). Pollen robbers,
which may  also act as occasional pollinators, were more frequent
in A. nodosum, with 18.8% of visits, than in A. peregrinum, with
4.3% of the visits, being represented by Ceratina (Crewella) sp., T.
spinipes, Exomalopsis fulvofasciata and Apis mellifera (Table 2). The
pollen robbers frequently opened the corolla in pre-anthesis buds
(Fig. 3A, B). They made long visits to a same flower, usually walking
around the anthers collecting pollen. They may  occasionally touch
the stigma. The pollen robbers frequently exhaust the pollen grains
during their visits, making the pollen scarce during the rest of the
day.

Effective pollinators for both species were medium to large bees.
The frequency of visits by effective pollinators was similar between
the two  species, with 17.2% of the visits in A. nodosum and 17.7%
in A. peregrinum (Table 2). In A. nodosum, the most frequent bees
showing the behaviour of effective pollinators were Centris fuscata
and Euglossa cordata (Fig. 3C), while in A. peregrinum were Euglossa
melanotricha and Epicharis bicolor.  Although E. cordata and E. melan-
otricha (that are about 0.4 cm dorsi-ventrally wide) performed a
great amount of visits showing the behaviour of effective pollina-
tors, the wide dorsiventral opening of corolla tubes, 0.57 ± 0.11 cm
in A. nodosum (n = 7) and 0.82 ± 0.17 cm in A. peregrinum (n = 28),
sometimes hindered the contact of these bees with the stigma
(Fig. 3C). In this sense, E. cordata and E. melanotricha were less
effective pollinators than the larger Centridini bees. Due to their
size, Centridini bees were probably the most effective pollinators
of both Adenocalymma species.

3.3. Breeding system and interspecific pollination

Breeding system results indicated that both species are self-
sterile (Table 3), with a very small number of fruits formed after
either spontaneous or hand self-pollination. The pollen-ovule ratio
of 4.2 × 103 and 4.6 × 103 (Table 1) included the species and pop-
ulations in the obligate xenogamy category (sensu Cruden, 1977),

supporting the results found in experimental pollinations (Table 3).
In the first analysis of fruit development after pollination experi-
ments, all pistils of hand-self pollinations had been aborted 45 days
after pollination in A. nodosum, and there were only two developing



D.S. Sampaio et al. / Flora 223 (2016) 19–29 25

Fig. 2. Nectar features of Adenocalymma nodosum and A. peregrinum. A and B, Adenocalymma nodosum. A, nectar volume. B, nectar concentration of soluble sugars. C and D, A.
peregrinum.  C, nectar volume. D, nectar concentration of soluble sugars. F = F statistic value of ANOVA; U = Mann-Whitney statistic, t = Student t-test statistic; P = probability.
In  each chart, means followed by different letters are statistically different (P < 0.05).

Table 4
Mean seed number per fruit, seed-set (percentage of ovules converted into seeds), and percentage of seeds with embryo in fruits from experimental and natural pollinations
of  Adenocalymma nodosum (Silva Manso) L. Lohmann and A. peregrinum (Miers) L. Lohmann.

Species Adenocalymma nodosum Adenocalymma peregrinum

Experimental pollinations Mean seed
number per
fruit ± SD (n)

Seed-set% % of seeds with
embryo (n)

Mean seed
number per
fruit ± SD (n)

Seed-set% % of seeds with
embryo (n)

Hand self-pollination – – 6.00 (1) 42.67 0.00 (1)
Intraspecific hand cross-pollination 8.00 ± 2.12 (26)a 80.00a 83.57 (35)a 6.70 ± 3.67 (53)b 47.52b 84.45 (328)a
Interspecific hand pollination 7.68 ± 2.94 (40)a 76.75a 75.88 (257)b 7.20 ± 3.78 (20)ab 51.06ab 55.91 (93)c
Natural pollination 6.27 ± 2.96 (49)b 62.65b 66.15 (65)b 7.95 ± 2.72 (63)a 56.42a 74.49 (196)b
Wald  Chi-square (P) 9.50 (P < 0.01) 32.77 (P < 0.01) 5.96 (P = 0.05) 3010.46 (P < 0.01) 12.99 (P < 0.01) 31.69 (P < 0.01)

N nifican
t

f
A
A
g
f
t

f
i
A
i
a

p
o
t
(
a

b
d
i

ote: means followed by different letters in column are distinct based on Least Sig
he  analyses.

ruits from this treatment in A. peregrinum 20 days after pollination.
lthough one fruit from spontaneous self-pollination developed in
. nodosum and two of hand self-pollination developed in A. pere-
rinum (Table 3), the ISI was 0.028 for the first species and 0.037
or the second. In addition, the only fruit from hand self-pollination
hat reached maturity showed all seeds without embryos (Table 4).

Natural fruit-set was low (ranging between 1.4 and 6.0%), while
ruit-set from cross-pollinations was higher in both species (rang-
ng between 26 and 39%), leading to low reproductive efficacy in
. nodosum (0.15) and A. peregrinum (0.05) (Table 3). Fruit-set from

nterspecific hand pollinations was also high (ranging between 35
nd 38%).

Fifty per cent of the pistils from self- and 67% from cross-
ollinations of A. nodosum and 40% of the pistils of both treatments
f A. peregrinum presented pollen tubes in the style, reaching
he ovary and initiating ovule penetration 24 h after pollination
Fig. 3D). No differences in pollen tube growth or evidences of tube
rresting mechanisms were observed in either species.
In A. nodosum, ovary area did not differ among treatment,
ut increased along time intervals. This increase was  depen-
ent on treatment (treatment: �2 = 2.44, d.f. = 2, P = 0.294; time

nterval: �2 = 23.90, d.f. = 4, P = 0.002; treatment × time interval:
t Difference (P < 0.05) and the hand self-pollination treatment was  excluded from

�2 = 62.85, d.f. = 8, P < 0.001) (Fig. 4A-C). The ovary area var-
ied along time intervals for all treatments in a linear manner
(cross-pollination: �2 = 60.04, d.f. = 4, P < 0.001; self pollination:
�2 = 25.89, d.f. = 4, P < 0.001; unpollination treatment: �2 = 18.79,
d.f. = 4, P = 0.001). However, the ovary growth rates (mm2 per hour)
varied among treatments, with a rate of 0.044 for cross-, 0.028 for
self-, and 0.016 for unpollinated pistils (Fig. 4A–C). Ovary areas
from cross-pollinated pistils were significantly larger than those
of unpollinated ones 120 h after pollination/onset of anthesis, but
no difference was  detected among ovary area for the other com-
parisons.

In A. peregrinum, an increase in ovary area along time intervals
was detected, ovary area differed among treatments, and there
was also a significant interaction effect (treatment: �2 = 28.48,
d.f. = 2, P < 0.001; time interval: �2 = 54.45, d.f. = 4, P < 0.001; treat-
ment × time interval: �2 = 62.85, d.f. = 8, P < 0.001) (Fig. 4D–F). The
ovary area grew along time intervals and was well adjusted
to a linear regression model in cross-pollination (�2 = 69.91,

d.f. = 4, P < 0.001) and self-pollination (�2 = 21.33, d.f. = 4, P < 0.001)
treatments. However, growth was  irregular in the unpollination
treatment and adjusted only to a cubic model (�2 = 15.98, d.f. = 4,
P = 0.003). The ovary growth rates (mm2 per hour) varied among



26 D.S. Sampaio et al. / Flora

Fig. 3. Flower visitors and fertilization in Adenocalymma. A, Oxaea flavescens obtain-
ing nectar from Adenocalymma peregrinum flower. Note that the flower had been
already damaged by Trigona spinipes (arrow). B, Floral bud of A. peregrinum being
damaged by the pollen robber T. spinipes. C, Euglossa cordata inside the flower of A.
nodosum.  The arrow points to the open stigma. D, Ovules from self-pollinated pistils
of  A. nodosum being penetrated by pollen tubes. Scale bar = 10 �m (D).
 223 (2016) 19–29

treatments, with a rate of 0.045 for cross-, 0.026 for self-pollination
(Fig. 4D, E). Ovary areas differed among all pairwise pollination
treatment comparisons 120 h after pollination/onset of anthesis.

Fruits from cross-pollinations and interspecific hand pollina-
tions showed more seeds and a higher seed-set than those from
natural pollinations in A. nodosum, whereas in A. peregrinum fruits
from natural pollinations showed more seeds and a higher seed-set
than those from intraspecific hand pollinations (Table 4). Seed-
set values were higher in A. nodosum than A. peregrinum in all
treatments (Table 4). For both species the percentage of seeds
with embryo was  higher in intraspecific hand cross-pollinations
than in the other treatments, and in A. peregrinum it was higher
in natural pollinations than in interspecific hand pollinations
(Table 4). Seed germination from interspecific pollinations ceased
40 days after sowing and all germinated seeds presented only one
seedling. Among the seeds developed in A. nodosum with A. pere-
grinum as pollen donor, 88.5% (n = 87) germinated, and among
seeds developed in A. peregrinum with A. nodosum as pollen donor
88.3% (n = 48) germinated. Sixty-six days after being transferred
to the greenhouse, seedlings already showed one or two pairs of
leaves.

4. Discussion

Adenocalymma nodosum and A. peregrinum were self-sterile
(sensu Bullock, 1985) but presented pollen tube growth to the
ovary and ovule penetration irrespective of treatment. This
characterizes a late-acting self-incompatibility system, LSI, the
only self-incompatibility mechanism reported so far for Bignon-
iaceae (e.g. Gibbs, 1990, 2014; Gibbs and Bianchi, 1993, 1999;
Bittencourt and Semir, 2005, 2006; Gandolphi and Bittencourt,
2010; Bittencourt et al., 2011). The low natural fruit-set and repro-
ductive success in both species may be related to pollen limitation
or ovule discounting (Barrett, 2002). This low natural fruit-set is not
caused by interspecific pollination interference, since fruit-set after
interspecific pollinations was high and the ensuing high percent-
ages of seed germination and viable seedlings indicated a bilateral
compatibility between A. nodosum and A. peregrinum.  This general
discussion is detailed below.

Although these species presented a very similar floral and polli-
nation biology, they showed distinct patterns of nectar production.
In A. nodosum flowers, a constant increase of nectar volume with
no variation in concentration was  observed, allowing visits along
the anthesis, as reported for other Bignoniaceae shrub species (e.g.
Milet-Pinheiro and Schlindwein, 2009). On the other hand, A. pere-
grinum showed a different pattern, with no variation in nectar
volume and an increase in concentration, many nectarless flowers
and coefficients of variation much higher than those observed for A.
nodosum. Nectarless flowers or low nectar production may  reduce
attraction and would explain the lower fruit-set in A. peregrinum.
However, visitation rate by effective pollinators was not different
between studied species. Actually variation in floral resources may
induce bees’ movement among flowers and even favour outcross-
ing (Feinsinger, 1978; Ishii et al., 2008), though it did not reflect in
a higher natural fruit-set in A. peregrinum.

The low natural fruit-set and reproductive success may  be asso-
ciated to the large number of nectar and pollen robbers in both
species. The prevalence of nectar robbers in both species decreases
the amount of nectar available for pollinators and may  have an
impact on fruit-set, as observed in other Bignoniaceae (Milet-
Pinheiro and Schlindwein, 2009). Pollen robbers may  have an even

more direct impact, reducing the amount of pollen early in the
morning and making it scarce for effective cross-pollinations dur-
ing the rest of the day (Guimarães et al., 2008). In addition, even
when they occasionally act as pollinators, they seem to perform



D.S. Sampaio et al. / Flora 223 (2016) 19–29 27

F bmitte
t a (mm
p l in F) 

m
S

o
e
s
a
a
D
a
(
a
c
(
2
c

p
(
i
2
2
o
t
s
p
h

ig. 4. Ovary development in Adenocalymma nodosum and A. peregrinum pistils su
he  onset of anthesis. A–C, A. nodosum. D–F, A. peregrinum. A, D, Ovary estimated are
istils. C, F, Ovary estimated area (mm2) for unpollinated pistils. y = linear (or cubica

ostly self-pollinations (Guimarães et al., 2008; Milet-Pinheiro and
chlindwein, 2009).

Although quantitative studies of flower visitors were carried
ut in different areas and timing, which may  explain the differ-
nt species of effective pollinators, A. nodosum and A. peregrinum
hare the same pollinator guild of medium to large sized Centridini
nd Euglossini bees. Centridini bees seem to be the main pollinators
lso for other Adenocalymma species such as A. bracteatum (Cham.)
C. (Amaral, 1992). Xylocopa species, which are often reported
s nectar robbers in Bignoniaceae flowers of Anemopaegma type
Correia et al., 2005; Dutra and Machado, 2001; Millet-Pinheiro
nd Schlindwein, 2009; Ynagizawa and Maimoni-Rodella, 2007),
an also act as effective pollinators of Adenocalymma marginatum
Cham.) DC. (Amaral, 1992), A. bracteatum (Almeida-Soares et al.,
010) and A. nodosum, probably due to the wider corolla tube when
ompared to flowers of the other genera.

The pollen tube growth and ovule penetration in self-pollinated
istils with their posterior abscission indicate the action of LSI
Seavey and Bawa, 1986), as described for most Bignoniaceae stud-
ed so far (e.g. Gibbs and Bianchi, 1993, 1999; Bittencourt and Semir,
005, 2006; Gandolphi and Bittencourt, 2010; Bittencourt et al.,
011). But pollen tube growth alone does not allow us to rule
ut early inbreeding depression processes (Gibbs, 2014). Consis-

ently higher growth rates of ovaries after cross-pollinations in both
pecies indicate that this may  be the differential threshold avoiding
istil abortion. Similar differences in ovules and pistil development
ave been described for other LSI species (e.g. Gibbs et al., 1999),
d to distinct pollination treatments in five time intervals after pollination or after
2) for self-pollinated pistils. B, E, Ovary estimated area (mm2) for cross-pollinated

regression model; R2 = coefficient of determination; P = probability of model fitting.

even with the unexpected residual growth of unpollinated pistils
also observed in A. nodosum. The ovary growth rates were much
higher in cross- than in self-pollinated pistils, pointing out to a
delay in ovary growth in self-pollinated ones, which evidence the
action of LSI in the first hours after pollination (Bittencourt and
Semir, 2005, 2006; Gandolphi and Bittencourt, 2010; Bittencourt
et al., 2011). The fact that 99% of selfed pistils were aborted up to
20 or 45 days after pollination in A. peregrinum and A. nodosum,
respectively, also corroborates the action of a LSI, which is related
to a synchronized abortion of these pistils (Bittencourt and Semir,
2005, 2006; Gandolphi and Bittencourt, 2010; Bittencourt et al.,
2011). However, the development of embryoless seeds in the single
developed self-pollinated fruit, as also reported to Handroathus ser-
ratifolius (Vahl) S. Grose (Alves et al., 2013), was somewhat distinct
from the mostly synchronous abortion postulated for LSI species
(Bittencourt et al., 2003; Bittencourt and Semir, 2005, 2006; Gan-
dolphi and Bittencourt, 2010). However, since the fruit set from
self-pollination was rare (1% of self-pollinated flowers), it may  rep-
resent more a case of failure of LSI mechanism than an evidence of
inbreeding depression.

Low natural fruit-set and reproductive success in the studied
species seem to be related to pollen limitation, due to pollen rob-
bers, or ovule discounting (sensu Barrett, 2002). Ovule discounting

is the loss of ovules and pistils due to self- or mixed pollinations (i.e.
mixed loads of cross and self-pollen deposited on the same stigma)
that are frequent in plants with bisexual flowers that opened more
than one flower per day (Kalisz et al., 2004; Bianchi et al., 2005;



2 / Flora

G
d
h
i
t
c
o
a
a
2
s
t
w
f
t

s
b
u
o
s
n
A
a
r
s
a
a
p
S
r
w
t
fi
2
h

A

i
N
U
A
w
D
4
2
e
f
G
t
S
t
w
C
N

R

A

A

8 D.S. Sampaio et al. 

oodwillie et al., 2005). Although we did not directly observe pollen
eposition and pollen tube growth after natural pollination, the
igher seed-set and seed number in fruits from cross-pollinations

n A. nodosum evidence a higher pollen quality or quantity in those
han in natural pollinations. It can be explained because in natural
onditions mixed pollinations often occur, and the self-fertilized
vules may  be unable to develop into seeds due to LSI action,
lthough pistils may  develop into fruits when there is a greater
mount of cross-fertilized ovules (Bertin et al., 1989; Gibbs et al.,
004). However, in A. peregrinum the seed number and the seed-
et were higher in natural pollinations, which is more difficult
o explain. The general lower seed-set recorded for A. peregrinum
hen compared with A. nodosum can be caused by embryo sac mal-

ormations or other reproductive problems, but this speculation has
o be further investigated by histological analysis.

The fact that interspecific hand pollinations presented more
eeds without embryos than intraspecific hand pollinations for
oth species, evidence that some interspecific fertilizations were
nable to form viable embryos, and this could be another source
f ovule discount in natural pollinations. Nevertheless, our data
howed a bilateral compatibility between A. peregrinum and A.
odosum with seeds and seedlings developed without problems.
s the speciation in Bignoniaceae seems to be predominantly
llopatric (Gentry, 1980, 1990), we would expect stronger bar-
iers maintaining the reproductive isolation among congeneric
pecies in a community (Gentry, 1976; Amaral, 1992; Ynagizawa
nd Maimoni-Rodella, 2007; Zjhra, 2008). However, since in the
bsence of fire there is usually a temporal separation between A.
eregrinum and A. nodosum flowering seasons in the region (D.
ampaio personal observation) functioning as a reproductive bar-
ier among these species, interspecific pollen flow may  be rare and
ould explain the absence of incompatibility mechanisms main-

aining species boundaries. We  could speculate that increase in
re frequency due to human disturbance (Hoffmann and Moreira,
002) may  have increased flowering overlap and favours natural
ybridization.

cknowledgements

This work was part of the Doctoral Thesis of the first author
n the Post-Graduate Program of Ecology and Conservation of
atural Resources at Universidade Federal de Uberlândia −
FU, supported by a research fellowship from Coordenaç ão de
perfeiç oamento de Pessoal de Nível Superior (CAPES). This work
as also supported by research grants by Conselho Nacional de
esenvolvimento Científico e Tecnológico (CNPq) (486091/2007-
) and by CAPES (Programa Nacional de Pós-Doutorado − PNPD
3038.008068/2010-95). We  thank administration of Clube de Caç a

 Pesca Itororó de Uberlândia, as administration of Campo Alegre
arm for permissions to carry out the study in these areas; Dra. Lúcia
. Lohmann (Universidade de São Paulo − USP) for plants iden-

ification; Dr. Léo Correia da Rocha Filho (USP), Dra. Cláudia Inês
ilva (USP), and Dra. Solange Cristina Augusto (UFU) for bees iden-
ification; Msc. Thatiana Martins dos Santos Mesquita for the help
ith nectar measurements of A. peregrinum; Dra. Júlia Yamagishi-
osta (UFU) for critical reading of the manuscript; and Natanael
ascimento dos Santos for illustrations.

eferences

lmeida-Soares, S., Polatto, L.P., Dutra, J.C.S., Torezan-Silingardi, H.M., 2010.

Pollination of Adenocalymma bracteatum (Bignoniaceae): floral biology and
visitors. Neotrop. Entomol. 39, 941–948.

lves, M.F., Duarte, M.O., Oliveira, P.E., Sampaio, D.S., 2013. Self-sterility in the
hexaploid Handroanthus serratifolius (Bignoniaceae), the national flower of
Brazil. Acta Bot. Bras. 27, 714–722.
 223 (2016) 19–29

Alves, M.F., Duarte, M.O., Bittencourt, N.S., Oliveira, P.E., Sampaio, D.S., 2016.
Sporophytic apomixis in polyembryonic Handroanthus serratifolius (Vahl) S.
Grose (Bignoniaceae) characterizes the species as an agamic polyploid
complex. Plant Syst. Evol., http://dx.doi.org/10.1007/s00606-016-1291-9 (in
press).

Amaral, M.E.C., 1992. Ecologia Floral De Dez Espécies Da Tribo Bignonieae
(Bignoniaceae), Em Uma  Floresta Semi-decídua No Município De Campinas, SP.
Doctoral Thesis. Universidade Estadual de Campinas, Campinas, Brazil.

Barrett, S.C.H., 2002. Sexual interference of the floral kind. Heredity 88, 154–159.
Bertin, R.I., Barnes, C., Guttman, S.I., 1989. Self-sterility and cryptic self-fertility in

Campsis radicans (Bignoniaceae). Bot. Gaz. 150, 397–403.
Bianchi, M.B., Harris, S.A., Gibbs, P.E., Prado, D.E., 2005. A study of the mating

system in Dolichandra cynanchoides (Bignoniaceae): an Argentinian Chaco
woodlands liana with a late-acting self-incompatibility. Plant Syst. Evol. 251,
173–181.

Bittencourt, N.S., Moraes, C.I.G., 2010. Self-compatibility and polyembryony in
South American yellow trumpet trees (Handroanthus chrysotrichus and H.
ochraceus:  Bignoniaceae): a histological study of post-pollination events. Plant
Syst.  Evol. 288, 59–76.

Bittencourt, N.S., Semir, J., 2005. Late-acting self-incompatibility and other
breeding systems in Tabebuia (Bignoniaceae). Int. J. Plant Sci. 166, 493–506.

Bittencourt, N.S., Semir, J., 2006. Floral biology and late-acting self-incompatibility
in  Jacaranda racemosa (Bignoniaceae). Aust. J. Bot. 54, 315–324.

Bittencourt, N.S., Gibbs, P.E., Semir, J., 2003. Histological study of post-pollination
events in Spathodea campanulata beauv. (Bignoniaceae): a species with
late-acting self-incompatibility. Ann. Bot. 91, 827–834.

Bittencourt, N.S., Pereira, E.J., São-Thiago, P.S., Semir, J., 2011. The reproductive
biology of Cybistax antisyphilitica (Bignoniaceae), a characteristic tree of the
South American savannah-like Cerrado vegetation. Flora 206, 872–886.

Bullock, S.H., 1985. Breeding systems in the flora of a tropical deciduous forest in
Mexico. Biotropica 17, 287–301.

Correia, M.C.R., Pinheiro, M.C., Lima, H.A., 2005. Biologia floral e polinizaç ão de
Arrabidaea conjugata (Vell.) Mart. (Bignoniaceae). Acta Bot. Bras. 19, 501–510.

Cruden, R.W., 1977. Pollen-ovule ratios: a conservative indicator of breeding
systems in flowering plants. Evolution 31, 32–46.

Dafni, A., Kevan, P.G., Husband, B.C., 2005. Practical Pollination Ecology.
Enviroquest, Cambridge.

Dutra, J.C.S., Machado, V.L.L., 2001. Entomofauna visitante de Stenolobium stans
(Juss.) Seem (Bignoniaceae): durante seu período de floraç ão. Neotrop.
Entomol. 30, 43–53.

Feinsinger, P., 1978. Ecological interactions between plants and hummingbirds in a
successional tropical community. Ecol. Monogr. 48, 269–287.

Firetti-Leggieri, F., Lohmann, L.G., Alcantara, S., Costa, I.R., Semir, J., 2013.
Polyploidy and polyembryony in Anemopaegma (Bignonieae, Bignoniaceae).
Plant Reprod. 26, 43–53.

Gandolphi, G., Bittencourt, N.S., 2010. Sistema reprodutivo do Ipê-Branco −
Tabebuia roseo-alba (Ridley) Sandwith. Acta Bot. Bras. 24, 840–853.

Gentry, A.H., 1974a. Flowering phenology and diversity in tropical Bignoniaceae.
Biotropica 6, 64–68.

Gentry, A.H., 1974b. Coevolutionary patterns in Central American Bignoniaceae.
Ann. Missouri Bot. Gard. 61, 728–759.

Gentry, A.H., 1976. Bignoniaceae of southern Central America: distribution and
ecological specificity. Biotropica 8, 117–131.

Gentry, A.H., 1980. Bignoniaceae − Part I (Crescentieae and Tourretieae). Flora
Neotrop. Monogr. 25, 1–131.

Gentry, A.H., 1990. Evolutionary patterns in neotropical Bignoniaceae. Mem.  N. Y.
Bot. Gard 55, 118–129.

Gibbs, P.E., 1990. Self-incompatibility in flowering plants: a neotropical
perspective. Rev. Bras. Bot. 13, 125–136.

Gibbs, P.E., 2014. Late-acting self-incompatibility −the pariah breeding system in
flowering plants. New Phytol. 203, 717–734.

Gibbs, P.E., Bianchi, M.B., 1993. Post-pollination events in species of Chorisia
(Bombacaceae) and Tabebuia (Bignoniaceae) with late-acting
self-incompatibility. Bot. Acta 106, 64–71.

Gibbs, P.E., Bianchi, M.B., 1999. Does late-acting self-incompatibility (LSI) show
family clustering? Two  more species of Bignoniaceae with LST: Dolichandra
cynanchoides and Tabebuia nodosa.  Ann. Bot. 84, 449–457.

Gibbs, P.E., Oliveira, P.E., Bianchi, M.B., 1999. Postzygotic control of selfing in
Hymenaea stigonocarpa (Leguminosae-Caesalpinioideae), a bat-pollinated tree
of  the Brazilian cerrados. Int. J. Plant Sci. 160, 72–78.

Gibbs, P.E., Bianchi, M.B., Ranga, N.T., 2004. Effects of self-, chase and mixed
self/cross-pollinations on pistil longevity and fruit set in Ceiba species
(Bombacaceae) with late-acting self-incompatibility. Ann. Bot. 94, 305–310.

Goodwillie, C., Kalisz, S., Eckert, C.G., 2005. The evolutionary enigma of mixed
mating systems in plants occurrence, theoretical, explanations, and empirical
evidence. Ann Rev. Ecol.Evol. Syst. 36, 47–79.

Gottsberger, G., Silberbauer-Gottsberber, I., 2006. Life in the Cerrado: a South
American Tropical Seasonal Ecosystem. Reta Verlag, Ulm.

Grassi, R.F., Resende, U.M., Silva, W.,  Macedo, M.L.R., Butera, A.P., Tulli, E.O., Saffran,
F.P., Siqueira, J.M., 2005. Estudo fitoquímico e avaliaç ão alelopática de Memora
peregrina −ciganinha −Bignoniaceae, uma espécie invasora de pastagens em

Mato Grosso do Sul. Quím. Nova 28, 199–203.

Guimarães, E., Di Stasi, L.C., Maimoni-Rodella, R.C.S., 2008. Pollination biology of
Jacaranda oxyphylla with an emphasis on staminode function. Ann. Bot. 102,
699–711.

http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0005
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0005
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0005
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0005
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0005
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0005
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0005
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0005
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0005
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0005
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0005
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0005
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0005
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0005
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0005
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0010
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0010
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0010
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0010
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0010
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0010
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0010
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0010
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0010
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0010
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0010
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0010
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0010
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0010
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0010
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0010
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0010
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0010
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0010
dx.doi.org/10.1007/s00606-016-1291-9
dx.doi.org/10.1007/s00606-016-1291-9
dx.doi.org/10.1007/s00606-016-1291-9
dx.doi.org/10.1007/s00606-016-1291-9
dx.doi.org/10.1007/s00606-016-1291-9
dx.doi.org/10.1007/s00606-016-1291-9
dx.doi.org/10.1007/s00606-016-1291-9
dx.doi.org/10.1007/s00606-016-1291-9
dx.doi.org/10.1007/s00606-016-1291-9
dx.doi.org/10.1007/s00606-016-1291-9
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0020
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0025
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0025
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0025
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0025
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0025
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0025
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0025
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0025
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0025
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0025
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0025
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0030
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0030
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0030
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0030
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0030
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0030
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0030
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0030
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0030
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0030
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0030
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0030
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0030
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0030
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0035
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0040
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0045
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0045
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0045
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0045
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0045
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0045
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0045
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0045
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0045
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0045
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0045
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0045
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0045
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0045
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0045
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0045
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0045
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0050
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0050
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0050
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0050
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0050
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0050
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0050
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0050
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0050
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0050
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0050
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0050
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0050
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0050
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0050
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0050
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0055
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0055
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0055
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0055
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0055
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0055
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0055
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0055
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0055
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0055
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0055
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0055
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0055
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0055
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0055
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0055
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0055
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0055
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0055
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0055
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0055
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0060
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0060
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0060
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0060
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0060
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0060
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0060
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0060
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0060
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0060
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0060
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0060
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0060
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0060
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0060
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0060
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0060
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0060
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0060
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0060
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0060
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0060
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0065
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0065
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0065
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0065
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0065
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0065
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0065
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0065
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0065
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0065
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0065
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0065
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0065
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0065
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0065
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0065
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0065
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0070
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0070
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0070
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0070
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0070
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0070
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0070
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0070
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0070
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0070
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0070
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0070
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0070
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0070
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0070
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0070
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0070
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0070
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0070
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0070
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0075
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0075
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0075
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0075
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0075
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0075
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0075
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0075
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0075
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0075
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0075
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0075
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0075
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0075
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0075
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0075
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0080
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0080
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0080
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0080
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0080
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0085
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0085
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0085
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0085
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0085
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0085
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0085
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0085
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0085
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0085
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0085
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0085
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0085
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0085
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0085
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0085
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0085
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0085
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0085
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0085
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0085
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0085
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0090
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0090
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0090
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0090
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0090
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0090
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0090
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0090
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0090
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0090
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0090
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0090
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0090
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0090
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0090
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0090
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0090
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0095
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0095
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0095
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0095
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0095
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0095
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0095
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0095
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0095
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0095
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0095
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0095
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0095
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0100
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0100
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0100
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0100
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0100
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0100
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0100
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0100
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0100
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0100
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0100
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0100
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0100
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0100
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0100
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0100
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0105
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0105
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0105
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0105
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0105
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0105
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0105
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0105
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0105
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0105
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0105
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0105
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0110
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0110
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0110
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0110
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0110
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0110
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0110
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0110
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0110
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0110
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0110
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0110
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0110
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0110
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0115
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0115
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0115
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0115
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0115
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0115
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0115
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0115
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0115
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0115
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0115
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0115
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0115
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0115
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0120
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0120
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0120
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0120
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0120
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0120
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0120
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0120
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0120
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0120
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0120
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0120
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0120
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0120
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0125
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0125
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0125
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0125
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0125
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0125
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0125
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0125
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0125
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0125
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0125
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0125
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0125
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0125
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0130
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0130
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0130
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0130
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0130
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0130
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0130
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0130
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0130
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0130
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0130
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0130
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0130
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0130
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0135
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0135
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0135
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0135
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0135
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0135
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0135
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0135
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0135
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0135
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0135
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0135
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0135
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0135
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0135
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0135
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0140
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0140
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0140
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0140
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0140
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0140
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0140
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0140
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0140
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0140
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0140
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0140
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0140
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0140
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0140
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0140
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0140
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0140
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0140
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0145
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0150
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0150
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0150
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0150
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0150
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0150
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0150
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0150
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0150
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0150
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0150
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0150
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0150
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0150
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0150
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0150
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0150
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0150
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0150
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0150
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0150
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0150
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0150
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0155
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0160
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0160
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0160
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0160
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0160
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0160
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0160
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0160
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0160
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0160
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0160
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0160
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0160
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0160
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0160
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0160
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0160
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0160
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0160
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0160
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0160
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0160
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0160
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0165
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0165
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0165
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0165
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0165
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0165
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0165
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0165
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0165
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0165
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0165
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0165
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0165
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0170
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0175
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0175
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0175
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0175
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0175
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0175
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0175
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0175
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0175
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0175
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0175
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0175
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0175
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0175
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0175
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0175
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0175


/ Flora

H

I

K

K

L

L

L

L

M

M

M

Braz. Arch. Biol. Tech. 50, 1043–1050.
Zar, J.H., 2010. Biostatistical Analysis. Prentice Hall, Englewood Cliffs-New Jersey.
Zjhra, M.L., 2008. Facilitating sympatric species coexistence via pollinator

partitioning in endemic tropical trees of Madagascar. Plant Syst. Evol. 271,
D.S. Sampaio et al. 

offmann, W.A., Moreira, A.G., 2002. The role of fire in population dynamics of
woody plants. In: Oliveira, P.S., Marquis, R.J. (Eds.), The Cerrados of Brazil:
Ecology and Natural History of a Neotropical Savanna. Columbia University
Press, New York, pp. 159–177.

shii, H.S., Hirabayashi, Y., Gaku, K., 2008. Combined effects of inflorescence
architecture, display size, plant density and empty flowers on bumble bee
behavior: experimental study with artificial inflorescences. Oecologia 156,
341–350.

alisz, S., Vogler, D.W., Hanley, K.M., 2004. Context-dependent autonomous
self-fertilization yields reproductive assurance and mixed mating. Nature 430,
884–886.

earns, C.A., Inouye, D.W., 1993. Techniques for Pollination Biologists. University
Press of Colorado, Niwot.

ohmann, L.G., 2006. Untangling the phylogeny of neotropical lianas (Bignonieae,
Bignoniaceae). Am.  J. Bot. 93, 304–318.

ohmann, L.G., 2010a. Bignoniaceae in Lista De Espécies Da Flora Do Brasil. Jardim
Botânico Do Rio De Janeiro [online]. Jardim Botânico do Rio de Janeiro, Rio de
Janeiro, Available from: http://floradobrasil.jbrj.gov.br/reflora/floradobrasil/
FB112400 (accessed 25.04.2016).

ohmann, L.G., 2010b. Bignoniaceae in Lista De Espécies Da Flora Do Brasil. Jardim
Botânico Do Rio De Janeiro [online]. Jardim Botânico do Rio de Janeiro, Rio de
Janeiro, Available from: http://floradobrasil.jbrj.gov.br/reflora/floradobrasil/
FB112409 (accessed 25.04.2016).

ohmann, L.G., Taylor, C.M., 2014. A new generic classification of Bignonieae
(Bignoniaceae). Ann. Missouri Bot. Gard. 99, 348–489.

artin, F.W., 1959. Staining and observing pollen tubes in the style by means of
fluorescence. Stain Tech. 34, 125–128.
cDowell, E.M., Trump, B., 1976. Histological fixatives for diagnostic light and
electron microscopy. Arch. Pathol. Lab. Med. 100, 405–414.

ilet-Pinheiro, P., Schlindwein, C., 2009. Pollination in Jacaranda rugosa
(Bignoniaceae) Euglossine pollinators, nectar robbers and low fruit set. Plant
Biol. 11, 131–141.
 223 (2016) 19–29 29

Neuffer, B., Paetsch, M.,  2013. Flower morphology and pollen germination in the
genus Capsella (Brassicaceae). Flora 208, 626–640.

Nunes, S.G., 1999. Ciganinha Memora Peregrina (Miers) Sandw Nova Planta
Invasora De Pastagem. Embrapa Gado De Corte Divulga/N◦35 [online]. Jardim
Botânico do Rio de Janeiro, Rio de Janeiro, Available from: http://www.cnpgc.
embrapa.br/publicacoes/divulga/GCD35.html (accessed 31.05.2014.).

Nunes, S.G., 2001. Controle De Plantas Invasoras Em Pastagens Cultivadas Nos
Cerrados. Embrapa Gado de Corte, Campo Grande.

Oliveira-Filho, A.T., Ratter, J.A., 2002. Vegetation physiognomies and woody flora of
the  Cerrado biome. In: Oliveira, P.S., Marquis, R.J. (Eds.), The Cerrados of Brazil:
Ecology and Natural History of a Neotropical Savanna. Columbia University
Press, New York, pp. 91–120.

Olmstead, R.G., Zjhra, M.L., Lohmann, L.G., Grose, S.O., Eckert, A.J., 2009. A molecular
phylogeny and classification of Bignoniaceae. Am.  J. Bot. 96, 1731–1743.

Ruiz, T.Z., Arroyo, M.T.K., 1978. Plant reproductive ecology of a secondary decidual
tropical forest. Biotropica 10, 221–230.

Sampaio, D.S., Bittencourt, N.S., Oliveira, P.E., 2013. Mating in the pseudogamic
apomictic Anemopaegma acutifolium DC: another case of
pseudo-self-compatibility in Bignoniaceae? Plant Biol. 15, 919–924.

Seavey, S.R., Bawa, K.S., 1986. Late-acting self-incompatibility in angiosperms. Bot.
Rev. 52, 195–219.

Ynagizawa, Y.A.N., Maimoni-Rodella, R.C.S., 2007. Floral visitors and reproductive
strategies in five Melittophilous species of Bignoniaceae in Southeastern Brazil.
157–176.

http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0180
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0180
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0180
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0180
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0180
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0180
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0180
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0180
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0180
http://refhub.elsevier.com/S0367-2530(16)30058-5/sbref0180
http://refhub.elsevier.com/S0367-2530(16)30