Floral traits driving reproductive isolation of two co-flowering taxa that share vertebrate pollinators
AoB PLANTS www.aobplants.oxfordjournals.org
Downloaded from https://academic.oup.com/aobpla/article
Floral traits driving reproductive isolation of two co-flowering taxa that share vertebrate pollinators
Joel A. Queiroz 2
Zelma G. M. Quirino 1
Isabel C. Machado 0 2
Associate Editor: Anna Traveset
0 Departamento de Botaˆ nica, CCB, Universidade Federal de Pernambuco , 50372-970 Recife, PE , Brazil
1 Departamento de Engenharia e Meio Ambiente, CCAE, Universidade Federal da Para ́ıba , Rio Tinto, PB , Brazil
2 Programa de Po ́s-Graduac ̧ a ̃o em Biologia Vegetal, Departamento de Botaˆ nica, CCB, Universidade Federal de Pernambuco , 50372-970 Recife, PE , Brazil
Floral attributes evolve in response to frequent and efficient pollinators, which are potentially important drivers of floral diversification and reproductive isolation. In this context, we asked, how do flowers evolve in a bat - hummingbird pollination system? Hence, we investigated the pollination ecology of two co-flowering Ipomoea taxa (I. marcellia and I. aff. marcellia) pollinated by bats and hummingbirds, and factors favouring reproductive isolation and pollinator sharing in these plants. To identify the most important drivers of reproductive isolation, we compared the flowers of the two Ipomoea taxa in terms of morphometry, anthesis and nectar production. Pollinator services were assessed using frequency of visits, fruit set and the number of seeds per fruit after visits. The studied Ipomoea taxa differed in corolla size and width, beginning and duration of anthesis, and nectar attributes. However, they shared the same diurnal and nocturnal visitors. The hummingbird Heliomaster squamosus was more frequent in I. marcellia (1.90 visits h21) than in I. aff. marcellia (0.57 visits h21), whereas glossophagine bats showed similar visit rates in both taxa (I. marcellia: 0.57 visits h21 and I. aff. marcellia: 0.64 visits h21). Bat pollination was more efficient in I. aff. marcellia, whereas pollination by hummingbirds was more efficient in I. marcellia. Differences in floral attributes between Ipomoea taxa, especially related to the anthesis period, length of floral parts and floral arrangement in the inflorescence, favour reproductive isolation from congeners through differential pollen placement on pollinators. This bat - hummingbird pollination system seems to be advantageous in the study area, where the availability of pollinators and floral resources changes considerably throughout the year, mainly as a result of rainfall seasonality. This interaction is beneficial for both sides, as it maximizes the number of potential pollen vectors for plants and resource availability for pollinators.
Bats; chiropterophily; generalized and specialized systems; hummingbirds; Ipomoea; pollinator sharing; Trochilidae
Historically, pollination biology has considered the
diversification of floral traits of angiosperms as an adaptive
response to selective pressures mediated by their
(Darwin 1862; Stebbins 1970)
. It has also been a
general idea that the set of floral attributes of a species
should be related to its more frequent and effective
pollinator (Stebbins 1970). However, determination of the
most important pollinator does not seem so easy in
some systems, especially those in which the number of
floral visitors of different taxa is high
Olesen et al. 2007; Maldonado et al. 2013; Avila et al.
. On the other hand, since floral features may in
fact reflect a rather diverse pollination history (Waser
et al. 1996), classifying the flowers in a given pollination
syndrome can mask the importance of ‘secondary’ or
‘tertiary’ pollinators as drivers of particular floral traits.
The pollination efficiency of different floral visitors has
been measured by quantitative (e.g. visits ratio) and
qualitative (e.g. per-visit pollen removal and deposition,
pollen germination ratio or fruit and seed set)
(Muchhala et al. 2009; Maldonado et al. 2013;
Rocca and Sazima 2013; Zych et al. 2013; Aslan et al.
. Such efficiency components exhibit a high
variation in pollination systems whose pollinators are
taxonomically very close
(Rocca and Sazima 2013)
, and even
more in those systems whose pollinators belong to
different functional groups or to distant taxa
Horvitz 1984; Sazima et al. 1994; Marte´ n-Rodr´ıguez
et al. 2009; Muchhala et al. 2009; Amorim et al. 2013;
Maldonado et al. 2013)
Bat – hummingbird pollination systems have been
interpreted as transitions from ornithophily to
(Buzato et al. 1994; Sazima et al. 1994)
evolutionary stable generalist systems (Muchhala et al.
2009). Variations in pollinator effectiveness have been
demonstrated, for example, in pollination systems
involving bats and hummingbirds and different plant species,
such as Siphocampylus sulfureus (Campanulaceae)
et al. 1994)
, Abutilon (Malvaceae)
(Buzato et al. 1994)
, Aphelandra acanthus
(Muchhala et al. 2009)
(Christianini et al. 2013; Queiroz et al.
. Those variations in the role of pollinators can
favour generalization (Waser et al. 1996). For example,
in the case of A. acanthus, primary pollinators (e.g. bats)
transfer a large amount of intra- and interspecific pollen
to the stigma of the plant
(Muchhala et al. 2009)
decrease in the quality component of bat pollination
service makes secondary pollination (e.g. hummingbirds)
beneficial and, hence, the generalized pollination system
(Muchhala et al. 2009)
From the plant’s perspective, a bat– hummingbird
pollination system can positively affect fitness by increasing
the number of pollen vectors and insurance in seed
production, even when one of the pollinator taxa becomes
(Muchhala 2003, J. Queiroz, unpubl. data)
However, it can also negatively affect fitness, by means of
pollen loss and stigma clogging by interspecific pollen
deposition when a pollinator visits other plant species
(Lloyd and Yates 1982; Murcia and Feinsinger 1996;
Caruso and Alfaro 2000; Johnson and Steiner 2000;
Morales and Traveset 2008)
The mixture and loss of pollen are common when
congeneric plant species share their pollinators
Waser 1983; Caruso and Alfaro 2000)
. These negative
effects of pollinator sharing can be reduced if plants
involved show divergences in floral structures that allow
(Grant 1992; Fulton and Hodges
and place their pollen on different body parts
of the pollinator
(Kay 2006; Muchhala and Potts 2007;
. They can also be reduced through floral
specialization in different pollinators
(Grant 1992; Christianini
et al. 2013)
via morphological restriction, which can
guarantee reproductive isolation
through interspecific pollen transfer can drive character
displacement in plant species that coexist and share the
(Muchhala and Potts 2007)
In the present study, we investigated the pollination
ecology of two co-flowering taxa of Ipomoea (Convolvulaceae).
Previous records of bat visits conducted in one of the taxa
studied here, Ipomoea aff. marcellia (Z. G. M. Quirino and
I. C. Machado pers. obs.), indicated that their ‘cup-shaped’
(sensu Fleming et al. 2009)
, whitish green and twilight
flowers were related to chiropterophily. Although there are
suggestions of bat pollination for a few species of Ipomoea
(Butanda-Cervera et al. 1978; Dobat and Peikert-Holle
1985; Sa´ nchez and Medell´ın 2007; Fleming et al. 2009)
the role of bats in the pollination was tested only for
(Caballero-Mart´ınez et al. 2012)
. Here, we
present two new records of bat pollination in the genus
Ipomoea (I. marcellia and I. aff. marcellia). In addition to
bats, floral traits of these two taxa indicate that
hummingbirds are also visitors of both studied Ipomoea. Thus, we
found a suitable model to compare the role of nocturnal
and diurnal vertebrates in pollination and to test
hypotheses about the mechanisms that could drive floral
adaptation to particular pollinators and, consequently,
reproductive isolation in similar and co-flowering species.
Our main questions were: (i) do differences in floral
attributes between both Ipomoea taxa cause
reproductive isolation through differential pollen placement on
pollinators? (ii) What is the breeding system of these
two Ipomoea taxa? and (iii) are bats more efficient
pollinators that hummingbirds in both Ipomoea taxa? We
expected that: (i) floral morphology (e.g. corolla width
and length) to be the main attribute related to
reproductive isolation between the two studied Ipomoea; (ii) higher
fruit and seed per fruit set in outcrossing (intrataxa) than
in self-pollination and outcrossing (intertaxa). And, given
that bats have been considered more efficient than
hummingbirds in some bat–hummingbird pollination system
(Muchhala et al. 2009; Muchhala and Thomson 2010)
also expected: (iii) floral attributes primarily related to bat
pollination and secondarily to hummingbird pollination in
both Ipomoea taxa, and (iv) pollination by bats should
result in a higher production of fruits and seeds than
pollination by hummingbirds.
Ipomoea taxa and study area
Ipomoea is a plant genus with 650 species distributed in
the tropics and subtropics
140 species of Ipomoea are known in Brazil
et al. 2013)
. They can occur in several types of vegetation,
from dry Caatinga shrublands to Amazon wetlands
et al. 2013)
. This genus is predominantly melitophilous
(Bullock et al. 1987; Gottsberger et al. 1988; Galetto and
Bernardello 2004; Pick and Schlindwein 2011)
. There are
only a few cases of pollination by birds
Sazima 1987; Galetto and Bernardello 2004)
cases of pollination by bats reported (Caballero-Mart´ınez
et al. 2012).
The two Ipomoea taxa selected as study models are
climbers with overlapping distribution in the Almas
Farm. They overlap their flowering periods, which occur
at the end of the rainy season and beginning of the dry
season, between July and October. The sampled
individuals were distributed along 2500 m, frequently in open
vegetation along trails and at the edge of rocky outcrops.
There is no information about whether both Ipomoea
studied in this work, have overlapping distribution in
other areas, considering that only one of the two taxa
appears in floristic lists, identified as I. marcellia
et al. 2013)
. The other taxon has not been described
yet, or has been misidentified as I. marcellia, and could
be new to science. Here, we denominated it as Ipomoea
aff. marcellia, due to its similarity to I. marcellia. Several
differences between these two plants, both in floral
morphology and pollination ecology, which will be assessed
in the present study, justified considering them as two
separate taxa. The flowers of both Ipomoea taxa are
(sensu Fleming et al. 2009)
, whitish green and
have a tubular corolla. They are horizontally arranged in
I. marcellia, and in an upward, almost vertical position
in Ipomoea aff. marcellia, due to a curvature of the corolla
Individuals of I. marcellia occur close to the ground,
and have been frequently observed climbing on shrubs
of 1.5 and 2.0 m height. In turn, I. aff marcellia shows
high variation in height among the species it climbs on,
from shrubs close to the ground to trees (4.0 – 6.0 m).
We did not observe individuals of I. marcellia in supports
as high as individuals of I. aff. marcellia.
We developed the field study in Almas Farm (7828′45′′S
and 36854′18′′W), a 3505 ha private reserve located in the
Brazilian Caatinga. This area has the lowest rainfall
indices in Brazil (600 mm year21) and strong rainfall
. It has two marked seasons: a rainy
season, concentrated in the first 3 months of the year
and a dry season that lasts between 6 and 9 months.
A shrubby-arboreal, thorny, deciduous vegetation is
typical of the hyperxerophilic caatinga
(Rodal and Sampaio
Morphometry and floral anthesis
Field activities were carried out from July to October 2012,
covering the entire flowering period of the studied
Ipomoea taxa. To investigate differences in floral attributes
between the two Ipomoea taxa, we collected data on
length and diameter of the corolla tube, and length of
the style and pistil. To guarantee inter-individual floral
variability, we measured a single flower per individual
(n ¼ 20 individuals per taxon). We monitored anthesis
in different individuals (n ¼ 20 flowers per taxon) and
recorded the time of opening of flower buds, anther
dehiscence, stigmatic receptivity and senescence of
flowers. To test for stigmatic receptivity, one flower per
individual (n ¼ 15 individuals per taxon) had the pistil
dipped in hydrogen peroxide, and the stigma was
considered receptive when bubbles were released on the
. To determine whether the
flower is receptive during the visits by pollinators, the
measurements of stigmatic receptivity were carried out
in the beginning of anthesis and in the peak of visits by
hummingbirds and bats (n ¼ 5 flowers per time per
Nectar volume, concentration and amount of sugars
We recorded the total amount of nectar produced (mL),
concentration (%), amount of sugar (mg) and pattern of
production during the anthesis in both Ipomoea taxa,
Galetto and Bernardello (2005)
. We took
measurements of sugar volume using graduated microsyringes
(Hamilton, NV, USA) and measurements of concentration
with a handheld refractometer (0 – 50 %; Atago, Tokyo,
Japan). The measurements were carried out in all months
of flowering to include as many individuals as possible.
We selected individuals according to the availability of
flower buds and to avoid pseudoreplication, we excluded
individuals close to each other in measurements carried
out at the same time. The time and number of
measurements differed according to the period of daily anthesis of
each taxon: six measurements (volume, concentration
and milligrams of sugar) from 0900 to 0400 hours in
I. marcellia, and seven measurements (volume,
concentration and milligrams of sugar) from 1600 to 0700
hours in Ipomoea aff. marcellia. In each period, we
measured the accumulated volume of nectar in a group of
previously bagged flowers that could not be accessed
by flower visitors. And in each Ipomoea taxon, we used
a group of 10 – 20 flowers from different individuals per
measurement per time.
To analyze the reproductive system of the studied
Ipomoea, we carried out controlled pollination tests in the
(Radford et al. 1974)
. Frequently, the same individual
was used for more than one type of pollination
experiment (e.g. spontaneous self- and hand self-pollination).
However, to avoid pseudoreplication, one flower per
individual was used in similar experiments. Thus, flower buds
of different individuals were covered with voile bags,
and the following procedures were performed after the
beginning of anthesis: (i) spontaneous self-pollination
(I. marcellia, n ¼ 44; I. aff. marcellia, n ¼ 30)—without
hand pollen transfer to the stigma; (ii) hand
selfpollination (I. marcellia, n ¼ 20; I. aff. marcellia, n ¼
15)—with hand pollen transfer to the stigma of the same
flower; (iii) outcrossing intrataxa (I. marcellia, n ¼ 19;
I. aff. marcellia, n ¼ 22)—with hand pollen transfer
among individuals of the same taxon; and (iv) outcrossing
intertaxa (I. marcellia, n ¼ 11; I. aff. marcellia, n ¼ 16)—
with hand pollen transfer between taxa. After the
experiments, we bagged the flowers again and monitored the
production of fruits and seeds. Fruit set was measured
as the number of flowers setting fruit per treatment.
The seeds of the fruits were quantified for each
treatment, and for the analyzes, we have used the average
number of seeds per fruit.
Pollination by hummingbirds and bats: visit number, frequency and efficiency
We determined the taxonomic identity of flower visitors,
and number and frequency of visits to the flowers of both
Ipomoea taxa through diurnal and nocturnal
observations (n ¼ 10 individuals per taxon). As the number of
flowers can affect the visit rate, all individuals included
in the analysis had from one to two open flowers. We
carried out observations on different days ( 20 days),
in census lasting from 30 min to 2 h, during the flowering
season, in a total of 30 h for I. marcellia (diurnal
observation ¼ 13 h; nocturnal observation ¼ 17 h) and
26.5 h for Ipomoea aff. marcellia (diurnal observation ¼
11 h; nocturnal observation ¼ 15.5 h). Photographic
records were taken to help in the identification of flower
visitors and the visualization of the position of pollen
placement on the animal.
We captured bats visiting Ipomoea flowers with mist
nets (four mist nets, 7 × 2.5 m, Ecotone Inc., Spot,
Poland) for later identification, record of sites of pollen
deposition and the pollen load composition. The
identification of pollen types found on the body of bat specimens
was carried out using the reference collection of
chiropterophilous plant species found in the study area
et al. 2016)
. The pollen was removed from the body of
bats with adhesive tape and examined under light
microscope only in a quantitative manner, and we have
registered the presence/absence of pollen from both Ipomoea
taxa and from other plants species. The nets were opened
around blooming individuals during the peak of visits by
bats, between 1800 and 2100 hours. Sampling was
carried out from July to October, in a total of 48 h of capture
per Ipomoea taxon. Bats were identified in the field using
field guides or consulting specialists. One or two
specimens of each species were killed using a lethal chamber
with ether and deposited as vouchers in the zoological
collection of the Federal University of Pernambuco (UFPE).
We estimated and compared pollination services
by bats and hummingbirds using as indicators of the
number and frequency of flower visits and the quality of
those visits. The quality of visits is measured as the
number of fruits and seeds per fruit produced per flower in
experiments of diurnal and nocturnal selective exposure.
Each individual used in the treatment of bat and/or
hummingbird pollination was considered as an
experimental unit, that is, to perform the experiments
(nocturnal or diurnal exposure) we used only one flower per
plant individual, and a non-paired sample. Due to
differences in the anthesis period, diurnal exposure of flowers
varied according to taxon. In I. marcellia (n ¼ 24 flowers)
it was carried out in a single interval: 1000 – 1730 hours,
and in Ipomoea aff. marcellia (n ¼ 30 flowers) in two
intervals: 1600 – 1730 hours and 0500 – 0730 hours. All
flowers remained bagged except for the time of diurnal
exposure. Nocturnal exposure in I. marcellia (n ¼ 21
flowers) and Ipomoea aff. marcellia (n ¼ 27 flowers)
was carried out in the same interval: 1800 – 0430 hours.
All flowers remained bagged except for the time of
nocturnal exposure. We marked the flowers in both
experiments and monitored fruit set and seeds per fruit
set (Table 2).
We tested for differences in floral morphometry (e.g.
length and width of corolla and length of stamens and
pistil) and nectar characteristics (e.g. volume,
concentration and milligrams of sugar) between the two Ipomoea
taxa with a series of t-tests. In the treatments of
outcrossing (inter × intrataxa) and exclusion of pollinators
(bat × hummingbird pollination), we compared the fruit
set between treatments with a G test, taking into account
that this test is more suitable for samples with expected
values below five. We compared the number of seeds
formed in the controlled and natural pollination
treatments, and in the experiments of selective exclusion of
diurnal and nocturnal visitors with Kruskal – Wallis,
followed by a post hoc Dunn test. Whenever the statistic
test so required, the data were analyzed for normality
and homoscedasticity of variances. All tests followed
Sokal and Rohlf (1995)
Morphometry and flower anthesis
The flowers of I. marcellia had corolla, stamens and pistil
significantly larger than those of I. aff. marcellia (Table 1).
In addition, the edge of the corolla was more intensely
folded in I. marcellia than in I. aff. marcellia, which allows
higher exposure of anthers and stigma in the former
The time and total duration of anthesis differed
between taxa studied. The anthesis of I. marcellia started
in the morning, between 0900 and 1000 hours, and
lasted 20 h. The anthesis of Ipomoea aff. marcellia
started in the late afternoon, approximately at 1600
hours, and lasted 16 h. In both taxa, anthers were
dehiscent right after bud opening and the stigma remained
receptive throughout anthesis.
Nectar volume, concentration and amount of sugar
Nectar production was continuous and began with flower
opening: in I. marcellia between 0900 and 1000 hours and
in Ipomoea aff. marcellia at 1600 hours. The amount of
nectar decreased only at the end of anthesis: at 0400
hours in I. marcellia and 0700 hours in Ipomoea aff.
marcellia (Fig. 2). The volume, concentration and the total
amount of sugar in the nectar of Ipomoea aff. marcellia
were significantly higher than in I. marcellia (Table 1).
Both Ipomoea taxa were obligate xenogamous
: we observed fruit and seed set exclusively
in the treatment of hand cross-pollination (Table 2).
However, in spite of higher fruit and seed set in outcrossing
intrataxa than outcrossing intertaxa in I. marcellia (Z ¼
1.10; gl ¼ 1; P . 0.05) and in I. aff. marcellia (Z ¼ 2.14;
gl ¼ 1; P . 0.05), these differences were not statistically
significant. Natural seed set in both Ipomoea taxa did
not differ from outcrossing treatments (Table 2).
t ¼ 25.05; P , 0.0001
t ¼ 28.72; P , 0.0001
t ¼ 210.38; P , 0.0001
Treatments Ipomoea marcellia Ipomoea aff. marcellia
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
n Fruits (%) Seeds per fruit* n Fruits (%) Seeds per fruit*
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Spontaneous self-pollination 44 0 (0) 0 (0) 30 0 (0) 0 (0)
Pollination by hummingbirds and bats: visit number, frequency and efficiency
The two Ipomoea taxa shared the same diurnal and
nocturnal pollinator species (Figs 3 and 4). Heliomaster
squamosus (Family Trochilidae) was the only diurnal visitor
recorded (Fig. 3). This hummingbird accessed the flowers
of I. marcellia during a long daily period (0900 – 1730
hours). We recorded a total of 42 visits (1.90 visits h21).
In those visits, the pollen was deposited on the head
(Fig. 3A and B) and neck of the hummingbird (Fig. 3C
and D). The visits of H. squamosus to Ipomoea aff.
marcellia occurred at two intervals (1700 – 1730 hours and
0530 – 0730 hours). We recorded 11 visits during the
whole-observation time (0.57 visits h21), in which the
pollen was exclusively deposited on the neck of the
hummingbird (Fig. 3E and F).
The main nocturnal visitors were the bats Glossophaga
soricina and Lonchophylla mordax (Glossophaginae).
Hawkmoth visits were occasional; we recorded one
visit of Agrius cingulata to both Ipomoea taxa. Bat visits
to flowers were concentrated between 1800 and
2000 hours. We recorded 11 visits (0.57 visits h21) to
I. marcellia, in which the pollen was deposited in the
region between the neck and breast of the bat (Fig. 4A
and B). We recorded 12 visits (0.64 visits h21) of these
bats to Ipomoea aff. marcellia, in which the pollen was
deposited on the chin and neck of the bats (Fig. 4C and D).
We collected 13 bat specimens with pollen of Ipomoea taxa,
L. mordax (eight individuals) and G. soricina (five individuals).
The pollen load found in 100 % of the bats analyzed was a
mixture of pollen of Ipomoea taxa with pollen of other
plants that overlap part of their flowering period. Among
those plants were Pilosocereus chrysostele (Cactaceae),
Pseudobombax marginatum (Bombacaceae) and
Encholirium spectabile (Bromeliaceae).
In Ipomoea aff. marcellia, pollination by bats resulted in
more fruits produced than in hummingbird pollination
(G ¼ 5.47; df ¼ 1; P ¼ 0.01). The average number of
seeds per fruit in I. aff. marcellia in natural pollination
did not differ of bat (Z ¼ 1.80; df ¼ 1; P , 0.05) or
hummingbird pollination (Z ¼ 1.91; df ¼ 1; P , 0.05).
However, seed set was higher in bat pollination when
compared with hummingbird pollination in this taxon
(Z ¼ 3.19; df ¼ 1; P , 0.05). Although we did not observe
significant differences among bat and hummingbird
pollination in terms of fruits set in I. marcellia (G ¼ 2.53; df ¼
1; P . 0.05), in terms of seed set, hummingbird
pollination was more effective and significantly different from
bat pollination (Z ¼ 3.00; df ¼ 1; P , 0.05). In addition,
the number of seeds per fruit in I. marcellia in the
treatment of natural pollination did not differ from that seen in
bat (Z ¼ 2.07; df ¼ 1; P . 0.05) or hummingbird
pollination (Z ¼ 1.59; df ¼ 1; P . 0.05).
In the present work, we have described two new cases of
shared pollination by vertebrates (bat and hummingbird)
in Ipomoea. Pollination by vertebrates has been reported
for the genus
(see McDonald et al. 2011)
Ipomoea consists of a predominantly insect pollinated
group, mainly by bees
(Galetto et al. 2002; Galetto and
Bernardello 2004; McDonald et al. 2011 and references
. Although bat pollination is suggested for
I. albivenia, I. ampullacea, I. arborescens and I. neei
(Dobat and Peikert-Holle 1985; Fleming et al. 2009)
has been rarely confirmed by direct observations
(Sa´ nchez and Medell´ın 2007; Caballero-Mart´ınez et al.
. Vertebrate pollination in I. murucoides, a
primarily chiropterophilous species
et al. 1978)
, was the only case in which diurnal versus
nocturnal pollinator efficiency in Ipomoea has been
(Caballero-Mart´ınez et al. 2012)
If pollination systems are conceptualized as a
continuum, we would have at one end very specialized
systems, such as in long tubular flowers of Centropogon
nigricans pollinated by a single species of bat
, and in another end, generalist systems with
plant species and its numerous floral visitors
and Steiner 2000; Fenster et al. 2004; Maldonado et al.
2013; Avila et al. 2015)
. In this continuum, our bat –
hummingbird pollination system would be closer to
those more specialized. Some bat – hummingbird
pollination systems have been considered as
(Muchhala et al. 2009)
or intermediate between
ornithophily and chiropterophily
(Sazima et al. 1994)
We support the latter view, since the co-flowering
Ipomoea taxa had, in general, a set of floral traits (e.g.
cup-shaped, whitish green and accessible flowers,
anthesis and nectar by day and night) that favours
pollination by only two groups of vertebrates, bats and
The whitish green colour of the flowers in the two
Ipomoea taxa is not a typically ornithophilous colour,
but more associated with chiropterophily
van der Pijl 1979; Tschapka and Dressler 2002; Willmer
. However, similarly to what was observed in
, it does not hinder
visits by hummingbirds. Another factor, day- and night-time
availability of nectar, also favours bat – hummingbird
(Sazima et al. 1994; Muchhala et al. 2009;
Christianini et al. 2013)
. This kind of pollination system
has a high energy cost, because both bats and
hummingbirds show high metabolic rates, due to their sizes and
(Nicolson and Fleming 2003; Tschapka
. Indeed, the nectar concentration of the two
studied Ipomoea was higher than that described for
(2 – 29 %; von Helversen 1993; Fleming
et al. 2009)
and ornithophilous flowers
(15 – 25 %;
Hainsworth and Wolf 1972; Baker 1975; Baker and Baker
1983; Proctor et al. 1996)
In addition, the period of day– night anthesis of studied
Ipomoea taxa is also another factor favouring both bat
and hummingbirds activities. Plant species with
prolonged anthesis (e.g. .24 h) are usually visited by diurnal
and nocturnal animals
(Sazima et al. 1994; Muchhala
et al. 2009; Caballero-Mart´ınez et al. 2012; Amorim et al.
2013; Christianini et al. 2013; Aguilar-Rodr´ıguez et al.
. In such cases, the duration
the beginning of anthesis
(Aguilar-Rodr´ıguez et al.
are important floral attributes that can provide
clues about the contribution, in terms of fruit and seed
set, of the different groups of pollinators (nocturnal and
diurnal). Thus, the differences in the beginning and in
the total duration of anthesis between the two Ipomoea
taxa can affect the interaction with their pollinators,
favouring hummingbirds in I. marcellia (daytime onset,
duration: 20 h) and bats in I. aff. marcellia (twilight
beginning; duration: 16 h).
In general, in species of plants using nocturnal and
diurnal pollinator services, complementarity between these
two groups of animals is noted in the pollination of these
(Muchhala et al. 2009; Amorim et al. 2013;
AguilarRodr´ıguez et al. 2015; Cruz-Neto et al. 2015)
effect has been rarely observed in terms of fruit and seed
set, as reported to Inga
(Avila et al. 2015)
. In the present
study, bats and hummingbirds played complementary
roles in the pollination of both Ipomoea. However, such
roles diverged among the two taxa. Although
hummingbirds presented higher frequency of visits (quantity
component) than bats, both seem to contribute equally to a
successful fruit set in I. marcellia. However, the higher
amount of seeds per fruit resulting from the pollination
by hummingbirds in this plant can result from the longer
duration of the diurnal anthesis, which favours
hummingbird visit rate. However, the frequency of visits should not
be so relevant to determine the importance of the
pollinator in I. aff. marcellia, as bats and hummingbirds showed
the same visit rates in this plant, but bat pollination was
significantly more successful, both in fruit and seed set.
When plant species share a flowering period and have
the same pollinators, loss of pollen, due to mixing with
foreign pollen, is a common condition
Feinsinger 1996; Muchhala and Jarrin-V 2002)
some factors may contribute to minimize pollen loss,
such as flowering period displacement (Heithaus et al.
1975) or specialization on different pollination strategies
(Christianini et al. 2013)
. The second factor was more
important in the studied Ipomoea, since intertaxa
differences in the morphometry and spatial orientation of
flowers allow to use slightly different parts of the
pollinator’s body for pollen deposition. The floral morphology
plays an important role in reducing competition for
(Dressler 1968; Tschapka et al. 2006; Muchhala
2008; Armbruster et al. 2014)
and favour reproductive
isolation among sympatric taxa
(Muchhala 2003; Muchhala
and Potts 2007; Christianini et al. 2013; Armbruster et al.
. Similarly to Burmeistera, which are also pollinated
by bats and hummingbirds
in floral morphology in Ipomoea played a central role
in pollinator specialization and made possible the
co-occurrence and pollinator sharing in these taxa.
Despite the self-incompatibility observed in both
Ipomoea, we observed fruit and seed production by intertaxa
outcrossing. Hybridization under natural conditions is
very rare in Ipomoea due to cross-incompatibility
et al. 2009)
. However, under controlled conditions some
Ipomoea species of economic interest have been able
to generate hybrids
(Abel and Austin 1981; Cao et al.
. When congeneric species have contact with each
other, hybridization may not occur due to the existence
of pre- and post-mating barriers
Temporal displacement of flowering peaks, strong
pollinator specificity and high flower constancy in the shared
pollinators can contribute to restrict hybridization
(Marques et al. 2007)
. However, we did not observe any
of these barriers in the two studied Ipomoea. We believe,
though, that two other barriers can negatively affect
the success of intertaxa outcrossing by limiting the
formation of hybrids under natural conditions in these two
taxa: differences in the floral morphometry and
In the case of the studied Ipomoea, two main factors
may make a bat–hummingbird pollination system
advantageous. The first factor is related to the quality of the
pollen deposited on the stigma of Ipomoea flowers. Due to
their high energy requirements, bats and hummingbirds
need to visit several flowers, and, therefore, can show
mixed pollen load
(Murcia and Feinsinger 1996; Muchhala
and Jarrin-V 2002)
. During part of the flowering season of
Ipomoea taxa, other chiropterophilous and ornithophilous
plants were also in blossom (Quirino and Machado 2014).
Those plants deposit pollen on the same sites of the
pollinator’s bodies where both Ipomoea did. Indeed, all bats
collected in Farm Almas had pollen from other
chiropterophilous plants (J. Queiroz, unpubl. data). Although we did
not capture hummingbirds during fieldwork, we observed
H. squamosus visiting flowers of Melocactus sp., which
probably led to mixed pollen too. This mixed pollen pattern
can have negative implications to pollination, by
decreasing pollinator effectiveness and consequently favouring
generalization in detriment of specialization
and Jarrin-V 2002; Muchhala et al. 2009)
A second factor is related to an apparent advantage
of bat – hummingbird pollination systems in
environments where pollinator populations can undergo
(Waser et al. 1996)
or be small
et al. 2003)
. In the Caatinga, where rainfall changes
largely between years
populations can undergo seasonal variations and be less
abundant during droughts
(Aguiar and Martins 1997; Duarte
Junior and Schlindwein 2005)
. As the Ipomoea taxa
studied here bloom in the dry season, trusting the
pollination service to a single group of pollinators can be risky.
When specialization in a particular pollinator group
results in ecological dependence, the risk of extinction
is higher (Johnson 2010). Hence, in both cases (pollinator
‘infidelity’ and fluctuations in pollinator populations) the
natural selection favour shared pollination systems.
Both studied Ipomoea taxa blossom in the same period of
the year, occur simultaneously in mixed vegetation patches,
share the same vertebrate pollinators and could probably
hybridize. However, we consider that some factors should
minimize pollen exchange among those taxa and,
consequently, favour their reproductive isolation, namely,
variations in flower morphology and morphometry, floral
orientation, anthesis time and pollen deposition on distinct
places of the pollinator body, both in bats (I. marcellia:
breast/I. aff. marcellia: chin and throat) and hummingbirds
(I. marcellia: throat and forehead/I. aff. marcellia: chin).
Although we classified the two studied Ipomoea in a bat–
hummingbird pollination system, those plants seem to
have different degrees of specialization: the floral traits
of I. aff. marcellia are more consistent with
chiropterophily, which should explain the higher effectiveness of bats
in the pollination of this taxon, and most attributes of
I. marcellia, as well as its larger flower size and longer
anthesis, do not seem to restrict either hummingbirds
or bats. Hence, I. aff. marcellia can be seen as
comparatively more specialist than I. marcellia.
Sources of Funding
The work was supported by PELD-CNPq (Long-Term
Ecological Program) and by FACEPE (APQ-1096-2.03/08).
Contributions by the Authors
The authors contributed equally in all activities of this
work, its conception, data collection and the writing of
Conflict of Interest Statement
We thank Drs Rosangela Bianchini and Enrico Bernard for
identifying the plants and the bats, respectively; Dr Tarcila
Nadia and Elivania Barral MSc. for the help on some of
the statistical analyzes; Arthur Domingos MSc. for the
drawings of the Figs 3 and 4. Roberto Lima for logistical
support in the field; Dr Marcos Me´ ndez and an anonymous
reviewer for their important contributions to the
manuscript; the owners and managers of RPPN Fazenda Almas
for allowing us to develop the field work in the area;
Peld/CNPq and FACEPE (APQ-1096-2.03/08) for financial
Abel WE, Austin DF. 1981. Introgressive hybridization between
Ipomoea trichocarpa and Ipomoea lacunosa (Convolvulaceae).
Bulletin of the Torrey Botanical Club 108:231 –239.
Aguiar CML , Martins CF . 1997 . Relative abundance, diversity and bees phenology (Hymenoptera, Apoidea) at caatinga , Sa˜ o Joa˜ o do Cariri , Para´ıba, Brazil. Iheringia 83 : 151 - 163 .
Aguilar-Rodr´ıguez PA , Kr o¨mer T, Garc´ıa- Franco JG , MacSwiney GMC . 2015 . From dusk till dawn: nocturnal and diurnal pollination in the epiphyte Tillandsia heterophylla (Bromeliaceae) . Plant Biology; doi:10 .1111/plb.12319.
Amorim FW , Galetto L , Sazima M. 2013 . Beyond the pollination syndrome: nectar ecology and the role of diurnal and nocturnal pollinators in the reproductive success of Inga sessilis (Fabaceae) . Plant Biology 15 : 317 - 327 .
Armbruster WS , Shi XQ , Huang SQ . 2014 . Do specialized flowers promote reproductive isolation? Realized pollination accuracy of three sympatric Pedicularis species . Annals of Botany 113 : 331 - 340 .
Aslan A , Hart P , Wu J , Aslan CE . 2014 . Evaluating the qualitative effectiveness of a novel pollinator: a case study of two endemic Hawaiian plants . Biotropica 46 : 732 - 739 .
Austin DF . 1998 . Convolvulaceae. Morning Glory family . Journal of the Arizona-Nevada Academy of Science 30 : 61 - 83 .
Avila R , Pinheiro M , Sazima M. 2015 . The generalist Inga subnuda subsp. luschnathiana (Fabaceae): negative effect of floral visitors on reproductive success? Plant Biology 17 : 728 - 733 .
Baker HG . 1975 . Sugar concentrations in nectars from hummingbird flowers . Biotropica 7 : 37 - 41 .
Baker HG , Baker I. 1983 . Floral nectar sugar constituents in relation to pollinator type . In: Jones CE , Little RJ , eds. Handbook of experimental pollination biology . New York: Van Nostrand Reinhold, 117 - 141 .
Bullock SH , Ayala R , Baker I , Baker HG . 1987 . Reproductive biology of the tree Ipomoea wolcottiana (Convolvulaceae) . Madron˜ o 34 : 304 - 314 .
Buril MT , Delgado G , Barbosa MRV , Alves M. 2013 . Convolvulaceae do cariri paraibano, PB, Brasil . Revista Nordestina de Biologia 21 : 3 - 26 .
Butanda-Cervera A , Va´ squez -Yanes C , Trejo L. 1978 . La polinizacio´ n quiroptero´ fila . Una revisio´ n bibliogra´ fica. Biotica 3 : 29 - 35 .
Buzato S , Sazima M , Sazima I. 1994 . Pollination of three species of Abutilon (Malvaceae) intermediate between bat and hummingbird flower syndromes . Flora 189 : 327 - 334 .
Caballero-Mart´ınez LA , Aguilera-Go´ mez LI , Rivas-Manzano IV , Aguilar-Ortigoza CJ , Lamus-Molina V. 2012 . Biolog´ıa floral y polinizacio´ n de Ipomoea murucoides Roem . & Schult. (Convolvulaceae) en Ixtapan del Oro , Estado de Me´ xico (Me´xico) . Anales de Biolog´ıa 43 : 65 - 76 .
Cao Q , Zhang A , Ma D , Li H , Li Q , Li P. 2009 . Novel interspecific hybridization between sweetpotato (Ipomoea batatas (L.) Lam .) and its two diploid wild relatives . Euphytica 169 : 345 - 352 .
Caruso CM , Alfaro M. 2000 . Interspecific pollen transfer as a mechanism of competition: effect of Castilleja linariaefolia pollen on seed set of Ipomopsis aggregata . Canadian Journal of Botany 78 : 600 - 606 .
Christianini AV , Forzza RC , Buzato S. 2013 . Divergence on floral traits and vertebrate pollinators of two endemic Encholirium bromeliads . Plant Biology 15 : 360 - 368 .
Cruden RW . 1977 . Pollen-ovule ratios: a conservative indicator of breeding systems in flowering plants . Evolution 31 : 32 - 46 .
Cruz-Neto O , Machado IC , Galetto L , Lopes AV . 2015 . The influence of nectar production and floral visitors on the female reproductive success of Inga (Fabaceae): a field experiment . Botanical Journal of the Linnean Society 177 : 230 - 245 .
Darwin C. 1862 . On the various contrivances by which British and foreign orchids are fertilised by insects and on the good effects of intercrossing . London: John Murray, 365 p.
Dobat K , Peikert-Holle T. 1985 . Blu¨ ten und Flederma¨ use . Frankfurt: Waldemar Kramer Verlag, 356 p.
Dressler RL . 1968 . Pollination by euglossine bees . Evolution 22 : 202 - 210 .
Duarte Junior JA , Schlindwein C. 2005 . The highly seasonal hawkmoth fauna (Lepidoptera: Sphingidae) of the caatinga of northeast Brazil: a case study in the state of Rio Grande do Norte . Journal of the Lepidopterist Society 59 : 212 - 218 .
Faegri K , Van der Pijl L. 1979 . The principles of pollination ecology . London: Pergamon Press.
Fenster CB , Armbruster WS , Wilson P , Dudash MR , Thomson JD . 2004 . Pollination syndromes and floral specialization . Annual Review of Ecology, Evolution, and Systematics 35 : 375 - 403 .
Ferreira PPA , Sima˜ o -Bianchini R , Miotto STS . 2013 . Three new species of Convolvulaceae Juss . from South America. Phytotaxa 135 : 27 - 34 .
Fleming TH , Geiselman C , Kress WJ . 2009 . The evolution of bat pollination: a phylogenetic perspective . Annals of Botany 104 : 1017 - 1043 .
Fulton M , Hodges SA . 1999 . Floral isolation between Aquilegia formosa and Aquilegia pubescens . Proceedings of the Royal Society B: Biological Sciences 266 : 2247 - 2252 .
Galetto L , Bernardello G. 2004 . Floral nectaries, nectar production dynamics and chemical composition in six Ipomoea species (Convolvulaceae) in relation to pollinators . Annals of Botany 94 : 269 - 280 .
Galetto L , Bernardello G. 2005 . Rewards in flowers: nectar . In: Dafni A , Kevan PG , Husband BC , eds. Practical pollination biology . Cambridge, ON: 261 - 312 .
Galetto L , Fioni A , Calvin˜ o A. 2002 . Exito reproductivo y calidad de los frutos en poblaciones del extremo sur de la distribucio´ n de Ipomoea purpurea (Convolvulaceae) . Darwiniana 40 : 25 - 32 .
Gottsberger G , Camargo JMF , Silberbauer-Gottsberger I. 1988 . A beepollinated tropical community: the beach dune vegetation of Ilha de Sa˜ o Luis, Maranha˜ o, Brazil . Botanische Jahrbu¨ cher fu¨ r Systematik 109 : 469 - 500 .
Grant V. 1949 . Pollination systems as isolating mechanisms in Angiosperms . Evolution 3 : 82 - 97 .
Grant V. 1992 . Floral isolation between ornithophilous and sphingophilous species of Ipomopsis and Aquilegia . Proceedings of the National Academy of Sciences of the USA 89 : 11828 - 11831 .
Hainsworth FR , Wolf LL . 1972 . Crop volume, nectar concentration and hummingbird energetics . Comparative Biochemistry and Physiology Part A: Physiology 42 : 359 - 366 .
Heithaus ER , Fleming TH , Opler PA . 1975 . Foraging patterns and resource utilization in seven species of bats in a seasonal tropical forest . Ecology 56 : 841 - 854 .
Herrera CM. 1987 . Components of pollinator “quality”: comparative analysis of a diverse insect assemblage . Oikos 50 : 79 - 90 .
Johnson SD . 2010 . The pollination niche and its role in the diversification and maintenance of the southern African flora . Philosophical Transactions of the Royal Society B: Biological Sciences 365 : 499 - 516 .
Johnson SD , Steiner KE . 2000 . Generalization versus specialization in plant pollination systems . Trends in Ecology & Evolution 15 : 140 - 143 .
Kay KM . 2006 . Reproductive isolation between two closely related hummingbird-pollinated Neotropical gingers . Evolution 60 : 538 - 552 .
Lloyd DG , Yates JMA . 1982 . Intrasexual selection and the segregation of pollen and stigmas in hermaphrodite plants, exemplified by Wahlenbergia albomarginata (Campanulaceae) . Evolution 36 : 903 - 913 .
Machado CSI , Sazima M. 1987 . Estudo comparativo da biologia floral em duas espe´ cies invasoras: Ipomoea hederifolia e I. quamoclit (Convolvulaceae) . Revista Brasileira de Biologia 47 : 425 - 436 .
Maldonado MB , Loma´ scolo SB, Va´ zquez DP . 2013 . The importance of pollinator generalization and abundance for the reproductive success of a generalist plant . PLoS ONE 8 : e75482 .
Marques I , Rossell o´-Graell A , Draper D , Iriondo JM . 2007 . Pollination patterns limit hybridization between two sympatric species of Narcissus (Amaryllidaceae) . American Journal of Botany 94 : 1352 - 1359 .
Marte´ n-Rodr´ıguez S , Almarales-Castro A , Fenster CB . 2009 . Evaluation of pollination syndromes in Antillean Gesneriaceae: evidence for bat, hummingbird and generalized flowers . Journal of Ecology 97 : 348 - 359 .
McDonald JA , McDill JR , Hansen DR , Simpson BB . 2011 . A phylogenetic assessment of breeding systems and floral morphology of North American Ipomoea (Convolvulaceae) . Journal of the Botanical Research Institute of Texas 5 : 150 - 177 .
Morales CL , Traveset A. 2008 . Interspecific pollen transfer: magnitude, prevalence and consequences for plant fitness . Critical Reviews in Plant Sciences 27 : 221 - 238 .
Muchhala N. 2003 . Exploring the boundary between pollination syndromes: bats and hummingbirds as pollinators of Burmeistera cyclostigmata and B. tenuiflora (Campanulaceae) . Oecologia 134 : 373 - 380 .
Muchhala N. 2006a . The pollination biology of Burmeistera (Campanulaceae): specialization and syndromes . American Journal of Botany 93 : 1081 - 1089 .
Muchhala N. 2006b . Nectar bat stows huge tongue in its rib cage . Nature 444 : 701 - 702 .
Muchhala N. 2008 . Functional significance of interspecific variation in Burmeistera flower morphology: evidence from nectar bat captures in Ecuador . Biotropica 40 : 332 - 337 .
Muchhala N , Jarrin-V P . 2002 . Flower visitation by bats in cloud forests of western Ecuador . Biotropica 34 : 387 - 395 .
Muchhala N , Potts MD . 2007 . Character displacement among batpollinated flowers of the genus Burmeistera: analysis of mechanism, process and pattern . Proceedings of the Royal Society B: Biological Sciences 274 : 2731 - 2737 .
Muchhala N , Thomson JD . 2010 . Fur versus feathers: pollen delivery by bats and hummingbirds and consequences for pollen production . The American Naturalist 175 : 717 - 726 .
Muchhala N , Caiza A , Vizuete JC , Thomson JD . 2009 . A generalized pollination system in the tropics: bats, birds and Aphelandra acanthus . Annals of Botany 103 : 1481 - 1487 .
Murcia C , Feinsinger P. 1996 . Interspecific pollen loss by hummingbirds visiting flower mixtures: effects of floral architecture . Ecology 77 : 550 - 560 .
Nicolson SW , Fleming PA . 2003 . Nectar as food for birds: the physiological consequences of drinking dilute sugar solutions . Plant Systematics and Evolution 238 : 139 - 153 .
Olesen JM , Dupont YL , Ehlers BK , Hansen DM . 2007 . The openness of a flower and its number of flower-visitor species . Taxon 56 : 729 - 736 .
Paz JRL , Gimenes M , Pigozzo CM . 2013 . Three diurnal patterns of anthesis in Ipomoea carnea subsp. fistulosa (Convolvulaceae): implications for temporal, behavioral and morphological characteristics of pollinators? Flora 208 : 138 - 146 .
Pick RA , Schlindwein C. 2011 . Pollen partitioning of three species of Convolvulaceae among oligolectic bees in the Caatinga of Brazil . Plant Systematics and Evolution 293 : 147 - 159 .
Prado DE . 2003 . As Caatingas da Ame´ rica do Sul . In: Leal IR , Tabarelli M , Silva JMC , eds. Ecologia e Conservac¸ a˜ o da Caatinga . Recife: Universidade Federal de Pernambuco, 3 - 73 .
Proctor M , Yeo P , Lack A. 1996 . The natural history of pollination . London: Harper Collins.
Queiroz JA , Quirino ZGM , Lopes AV , Machado IC . 2016 . Vertebrate mixed pollination system in Encholirium spectabile: a bromeliad pollinated by bats, opossum and hummingbirds in a tropical dry forest . Journal of Arid Environments 125 : 21 - 30 .
Quirino ZGM , Machado IC . 2014 . Pollination syndromes in a Caatinga plant community in northeastern Brazil: seasonal availability of floral resources in different plant growth habits . Brazilian Journal of Biology 74 : 62 - 71 .
Radford AE , Dickinson WCJ , Massey R , Bell CR . 1974 . Vascular plant systematic . New York: Harper & How.
Rathcke BJ . 1983 . Competition and facilitation among plants for pollination . In: Real L, ed. Pollination biology. New York: Academic Press, 305 - 329 .
Rocca MA , Sazima M. 2013 . Quantity versus quality: identifying the most effective pollinators of the hummingbird-pollinated Vriesea rodigasiana (Bromeliaceae) . Plant Systematics and Evolution 299 : 97 - 105 .
Rodal MJN , Sampaio EVSB . 2002 . A vegetac¸a˜ o do bioma Caatinga . In: Sampaio EVSB , Giulietti AM , Virg´ınio J , Rojas CFLG , eds. Vegetac¸ a˜ o e flora da Caatinga . Recife: Universidade Federal de Pernambuco, 11 - 24 .
Sa´ nchez R , Medell´ın RA. 2007 . Food habits of the threatened bat Leptonycteris nivalis (Chiroptera: Phyllostomidae) in a mating roost in Mexico . Journal of Natural History 41 : 1753 - 1764 .
Sazima M , Sazima I , Buzato S. 1994 . Nectar by day and night: Siphocampylus sulfureus (Lobeliaceae) pollinated by hummingbirds and bats . Plant Systematics and Evolution 191 : 237 - 246 .
Schemske DW , Horvitz CC . 1984 . Variation among floral visitors in pollination ability: a precondition for mutualism specialization . Science 225 : 519 - 521 .
Sokal RR , Rohlf FJ . 1995 . Biometry. New York: WH Freeman & Co.
Stebbins GL . 1970 . Adaptive radiation of reproductive characteristics in angiosperms, I: pollination mechanisms . Annual Review of Ecology and Systematics 1 : 307 - 326 .
Templeton AR . 1989 . The meaning of species and speciation: a genetic perspective . In: Otte D , Endler JA , eds. Speciation and its consequences . Sunderland , MA: Sinauer, 3 - 27 .
Tschapka M. 2004 . Energy density patterns of nectar resources permit coexistence within a guild of Neotropical flower-visiting bats . Journal of Zoology 263 : 7 - 21 .
Tschapka M , Dressler S. 2002 . Chiropterophily: on bat-flowers and flower-bats . Curtis's Botanical Magazine 19 : 114 - 125 .
Tschapka M , Dressler S , Von Helversen O. 2006 . Bat visits to Marcgravia pittieri and notes on the inflorescence diversity within the genus Marcgravia (Marcgraviaceae) . Flora 201 : 383 - 388 .
Von Helversen O. 1993 . Adaptations of flowers to the pollination by glossophagine bats . In: Barthlott W, Naumann CM , SchmidtLoske K , Schuchmann KL , eds. Animal - plant interactions in tropical environments . Bonn: Zoologisches Forschungsinstitut und Museum Alexander Koenig , 41 - 59 .
Waser NM . 1983 . Competition for pollination and floral character differences among sympatric plant species: a review of evidence . In: Jones CE , Little RJ , eds. Handbook of experimental pollination biology . New York: Academic Press, 277 - 293 .
Waser NM , Chittka L , Price MV , Williams NM , Ollerton J. 1996 . Generalization in pollination systems, and why it matters . Ecology 77 : 1043 - 1060 .
Willmer P. 2011 . Pollination and floral ecology . Princeton: Princeton University Press.
Wolff D , Braun M , Liede S. 2003 . Nocturnal versus diurnal pollination success in Isertia laevis (Rubiaceae): a sphingophilous plant visited by hummingbirds . Plant Biology 5 : 71 - 78 .
Zeisler M. 1938 . U¨ ber die Abgrenzung der eigentlichen Narbenfla¨ che mit Hilfe von Reaktionen . Beiheft Botanisches Zentralblatt 58 : 308 - 318 .
Zych M , Goldstein J , Roguz K , Stpiczyn´ ska M. 2013 . The most effective pollinator revisited: pollen dynamics in a spring-flowering herb . Arthropod-Plant Interactions 7 : 315 - 322 .