Reproductive biology and flower-visitor interactions of two bromeliad species from the Brazilian Atlantic Forest

Abstract

Background and aims – The Bromeliaceae family has great importance in the maintenance of neotropical communities. In the Brazilian Atlantic Forest, bromeliads are among the major groups responsible for maintaining the local flora and fauna and participate in important ecological interactions with insects, anurans, and hummingbirds. This work reports on aspects of the reproductive biology and the interactions between two endemic bromeliad species from the Atlantic Forest (Aechmea bruggeri and Quesnelia indecora) and their floral visitors to assess the impact of these relationships on the reproductive success and conservation of these plants.
Material and methods – Reproductive phenology, floral biology, pollination experiments, and the reproductive success of both species were investigated. To determine the floral visitors, we made direct observations on flowers and collected floral visitors that could not be identified in the field.
Key results – Aechmea bruggeri and Quesnelia indecora presented the individual and population flowering phenological pattern classified as annual with intermediate duration. The species are partially and totally self-incompatible, respectively. Both species presented a varied visitation guild, and although Q. indecora presented flowers with ornitofilous characteristics, no hummingbirds were recorded for this species. The hummingbird Thalurania glaucopis was the main visitor for Aechmea bruggeri and the bee Trigona cf. braueri was the main visitor for Quesnelia indecora. Nectar thieving by lepidopterans was observed for both species. Pollen robbing by beetles and nectar robbing by bees were registered for Aechmea bruggeri and Quesnelia indecora, respectively. Fruit and seed set of both species were highly affected by herbivory, which may negatively affect their reproductive success.
Conclusion – Our work highlights the important role of bromeliads in neotropical communities, showing how floral visitors and plants interact by participating in maintaining biological diversity in the studied forest remnant.

Full text

Plant Ecology and Evolution 155 (1): 16–28, 2022
https://doi.org/10.5091/plecevo.84464
Reproductive biology and ower-visitor interactions of two bromeliad
species from the Brazilian Atlantic Forest
Matheus R. e Silva1,*, Bruno C. Barbosa2 & Ana Paula G. de Faria1,3
1Programa de Pós-Graduação em Biodiversidade e Conservação da Natureza, Universidade Federal de Juiz de Fora, Juiz de Fora, MG,
Brasil
2Laboratório de Ecologia Comportamental e Bioacústica, Universidade Federal de Juiz de Fora, Juiz de Fora, MG, Brasil
3Departamento de Botânica, Instituto de Ciências Biológicas, Universidade Federal de Juiz de Fora, Juiz de Fora, MG, Brasil
*Corresponding author: matheussilva_rezende@hotmail.com
RESEARCH ARTICLE
Background and aims – The Bromeliaceae family has great importance in the maintenance of neotropical
communities. In the Brazilian Atlantic Forest, bromeliads are among the major groups responsible for
maintaining the local ora and fauna and participate in important ecological interactions with insects,
anurans, and hummingbirds. This work reports on aspects of the reproductive biology and the interactions
between two endemic bromeliad species from the Atlantic Forest (Aechmea bruggeri and Quesnelia
indecora) and their oral visitors to assess the impact of these relationships on the reproductive success
and conservation of these plants.
Material and methods – Reproductive phenology, oral biology, pollination experiments, and the
reproductive success of both species were investigated. To determine the oral visitors, we made direct
observations on owers and collected oral visitors that could not be identied in the eld.
Key results – Aechmea bruggeri and Quesnelia indecora presented the individual and population owering
phenological pattern classied as annual with intermediate duration. The species are partially and totally
self-incompatible, respectively. Both species presented a varied visitation guild, and although Q. indecora
presented owers with ornitolous characteristics, no hummingbirds were recorded for this species. The
hummingbird Thalurania glaucopis was the main visitor for Aechmea bruggeri and the bee Trigona cf.
braueri was the main visitor for Quesnelia indecora. Nectar thieving by lepidopterans was observed for
both species. Pollen robbing by beetles and nectar robbing by bees were registered for Aechmea bruggeri
and Quesnelia indecora, respectively. Fruit and seed set of both species were highly aected by herbivory,
which may negatively aect their reproductive success.
Conclusion – Our work highlights the important role of bromeliads in neotropical communities, showing
how oral visitors and plants interact by participating in maintaining biological diversity in the studied
forest remnant.
Keywords – Aechmea bruggeri; Brazil; Bromeliaceae; cloud forests; orivory; hummingbirds; pollination;
Quesnelia indecora.
© 2022 Matheus R. e Silva, Bruno C. Barbosa, Ana Paula G. de Faria.
This article is published and distributed in Open Access under the terms of the Creative Commons Attribution License (CC BY 4.0), which
permits use, distribution, and reproduction in any medium, provided the original work (author and source) is properly cited.
Plant Ecology and Evolution is published by Meise Botanic Garden and Royal Botanical Society of Belgium
ISSN: 2032-3913 (print) – 2032-3921 (online)
INTRODUCTION
Bromeliaceae is a neotropical family with about 3653
species (Gouda et al. continuously updated). Almost half
of the bromeliad species occurs in the Brazilian territory,
especially in the Atlantic Forest, which represents the main
centre of endemism and diversity (Benzing 2000; Forzza
et al. 2015). The Atlantic Forest is the second largest
rain forest of South America and one of the world’s top
biodiversity hotspots (Ribeiro et al. 2011). In many areas of
the Brazilian Atlantic Forest, bromeliads oer shelter and
a breeding site for several species of invertebrates, such as

17
Silva, Barbosa & Faria, Reproductive ecology of two bromeliads from the Atlantic Forest
dipterans, Lepidoptera, Coleoptera, and Odonata (Marrero et
al. 1996; Basílio et al. 2015), as well as some amphibians,
including anuran species (Teixeira et al. 2002). Moreover,
these bromeliads oer important oral resources to support
pollinators, such as hummingbirds, and other occasional
visitors (Rocha et al. 2004).
Pollination by hummingbirds in Bromeliaceae comprises
about 60% of the genera (Givnish et al. 2014). These
vertebrates are the main pollinators of bromeliad species
from the Atlantic Forest (Varassin 2002). In some areas,
almost half of their owers are used as a food source for
these birds (Sazima et al. 1996; Buzato et al. 2000). In the last
years, many studies detailed these mutualistic interactions
(Sick 1985; Sazima et al. 1996; Buzato et al. 2000; Canela
& Sazima 2003a; Machado & Semir 2006; Piacentini &
Varassin 2007; Magalhães et al. 2018; Kessler et al. 2020).
Although some authors suggest a co-evolution between
hummingbirds and bromeliads, considering the relationship
between the beak morphology of these animals and the oral
morphology (Benzing 2000; Givnish et al. 2014), there is
still little evidence about this process. Kessler et al. (2020)
highlighted that there is insucient data on pollinators of
bromeliad species, in addition to the observation that the
same pollinators are shared by several bromeliads.
In addition to hummingbirds, other nectar feeding
animals with a diurnal behaviour, such as butteries and
bees, also benet from the sequential owering and oral
resources oered by bromeliad species (Varassin & Sazima
2000; Siqueira Filho & Machado 2001; Machado & Semir
2006). Bats form the second main group of vertebrates acting
like bromeliad pollinators, these animals being attracted by
oral scent and abundant nectar (Sazima et al. 1989; Benzing
et al. 2000; Aguilar-Rodríguez et al. 2019).
Compared to the number of studies that focused on
pollination biology of Bromeliaceae (e.g. Benzing et al.
2000; Kaehler et al. 2005; Carranza-Quiceno & Estévez-
Varón 2008; Scrok & Varassin 2011; Schmid et al. 2011;
Christianini et al. 2012; Hornung-Leoni et al. 2013; Rocca
& Sazima 2013; Aguilar-Rodríguez et al. 2014a, 2016,
2019; Marques et al. 2015; Velásquez-Noriega et al. 2020;
Milet-Pinheiro et al. 2021), research addressing the role of
oral visitors in non-mutualistic relations with bromeliads is
still scarce. Florivory has been reported for few species of
Bromeliaceae (Canela & Sazima 2003b; Grohme et al. 2007;
Cascante-Marín et al. 2009; Aguilar-Rodríguez et al. 2014b;
Saldarriaga 2014; Palacios-Mosquera et al. 2019; Freitas et
al. 2020), as well as examples of nectar robbing (González-
Gómez & Valdivia 2005; Fumero-Cabán & Meléndez-
Ackerman 2007, 2013).
Herbivory of fruit and/or seeds is another type of
ecological interaction that can severely reduce the tness of
the species. Studies on the interactions between bromeliads
and predators of fruit and seeds are also scarce (Nara &
Webber 2002; Cavallari 2004; Cascante-Marín et al. 2005;
Lenzi et al. 2006; Schmid et al. 2010; Filippon et al. 2012;
Saldarriaga 2014; Oliveira et al. 2021), although very
relevant in view of the impact on the fertility and viability of
populations, which is the basis for the stability of the species
in their natural habitat.
We focused on evaluating aspects of the reproductive
biology and the ower-visitor interactions of two endemic
Bromeliaceae species from a portion of the Atlantic Forest
located in southeastern Brazil. Over two consecutive years
of their owering and fruiting period, we investigated the
reproductive phenology, breeding systems, oral biology,
reproductive success, and ower-visitor interactions. Starting
from the premise that data on pollinators of bromeliads are
lacking, as highlighted by Kessler et al. (2020), we aimed
to improve the knowledge about these interactions through
the study of species whose pollination has not been studied
before. We also hypothesise that the reproductive success of
these species may be aected by non-mutualistic interactions,
such as pollen or nectar robbing and herbivory of fruits
and seeds. Finally, considering that the studied bromeliads
provide important ecosystem services, we contribute data
that could support the development of strategies for the
protection and conservation of endemic and/or endangered
species.
MATERIAL AND METHODS
Species and study area
Our study was carried out in the Reserva Particular do
Patrimônio Natural Chapadão da Serra Negra, a conservation
unit located in Minas Gerais state, southeastern Brazil, with
the coordinates 21°57′50″S, 43°48′1.0″W (g. 1). The area is
located in the Serra da Mantiqueira, a mountain range elected
by scientists as the 8th most irreplaceable protected area on
the planet and one of the ten most important locations for
biodiversity conservation (Le Saout et al. 2013). Altitude
in the study area ranges from 850 to 1200 m a.s.l. and the
climate is Cwa (Köppen), with dry winters and wet and
hot summers. The mean annual temperature is 20.6°C
with a mean annual rainfall of 1376 mm. Of the forest
physiognomies, dwarf cloud forests are the most common,
characterised by shrubs and small trees with a 3–5 m high
canopy (Oliveira-Filho et al. 2013).
We studied two Bromeliaceae species. The rst one
was Aechmea bruggeri Leme, an endemic species to the
Serra da Mantiqueira forest remnants of Minas Gerais
state, considered Critically Endangered in the state list of
threatened ora (Drummond et al. 2005). The other one was
Quesnelia indecora Mez, a species restricted to the Atlantic
Forest domain of southeastern Brazil, with distribution in
Minas Gerais and Espírito Santo states (Forzza et al. 2015).
Both species have leaf rosette forming a water tank. In
the study area, they occur terrestrially on leaf litter in the
forest understory, forming small (A. bruggeri) to large (Q.
indecora) clumps.
Reproductive phenology and oral biology
Reproductive phenophases were registered monthly
from January 2019 to December 2020 in the study area,
covering two consecutive owering and fruiting periods
for both species. The absence or presence of the following
phenophases was recorded for 31 individuals of A. bruggeri
and 67 individuals of Q. indecora: young inorescence,
oral buds, open owers, senescent owers, immature fruits,

18
Pl. Ecol. Evol. 155 (1), 2022
Figure 1 – Location of the study site. A. Portion of the Serra da Mantiqueira (indicated in red), located in the Minas Gerais state, southeastern
Brazil. Map created with QGIS v.3.20.3 ‘Odense’ (QGIS Development Team 2021). B. Location of RPPN Chapadão da Serra Negra (red
rectangle). C. The part of the RPPN where the reproductive studies and the oral visitor observations were done is indicated in yellow.
Sources of images B and C: Google Earth. Map data ©2020 Google. Images courtesy of ©2020 CNES Airbus via Google Earth.
and mature fruits. The classication of the phenological
owering patterns followed Gentry (1974) and Newstrom et
al. (1994).
The oral biology was investigated in two individuals
of A. bruggeri (n = 20 owers) and two individuals of Q.
indecora (n = 6 owers) collected in the eld and cultivated
in a greenhouse. We registered data about the number of
open owers per day, length and colour of the corolla,
anthesis hour, the period when owers remained open, nectar
volume, sugar concentration and mass. We measured nectar
volume with a graduated microsyringe of 50 µL (Hamilton,
NV, USA) from previously bagged owers. In each ower,
we performed one measurement at the moment of ower
opening and two more measurements taken every two hours
from anthesis. One measurement was taken from the owers
in the senescence stage, with the petals already withered.
The nectar sugar concentration was measured at the
moment of the ower opening using a hand refractometer
(0–33%; Atago, Tokyo, Japan), and the total amount of sugar
was calculated following Galetto & Bernardello (2005).
Pollination treatments and reproductive success
The breeding systems were investigated in four individuals
of A. bruggeri and ve of Q. indecora cultivated in a
greenhouse. These individuals were dierent from those that
were analyzed for oral biology. The following controlled
pollination experiments were carried out: (1) hand self-
pollination (n = 93 owers of A. bruggeri and n = 9 owers
of Q. indecora), where oral buds were bagged and the
owers that opened the next day were pollinated with pollen
from the same ower; (2) hand cross-pollination (n = 96
owers of A. bruggeri and n = 9 owers of Q. indecora),
where oral buds were emasculated and pollinated the
next day with pollen from other individuals of the same
species. Additionally, 14 individuals (n = 2236 owers) of A.
bruggeri and 33 individuals of Q. indecora (n = 288 owers)
were randomly selected in the eld for the natural (open)
pollination experiments.
For each controlled pollination experiment and the open
pollination experiment, fruit set was calculated as the number
of formed fruits divided by the number of tested owers.

19
Silva, Barbosa & Faria, Reproductive ecology of two bromeliads from the Atlantic Forest
Furthermore, we calculated the mean seed production for
each experiment. Indices of self-compatibility (SCI) were
estimated for each species (Lloyd & Schoen 1992). The self-
compatibility index was calculated based on the percentage
of fruit set (SCIf) or mean number of seeds per fruit (SCIs) via
hand self-pollination relative to the values from hand cross-
pollination. Values close to 1 are interpreted as complete
self-compatibility, and a value less than 0.75 is interpreted as
being due to at least partial self-incompatibility.
Floral visitors
The observations of oral visitors were realized in 24
individuals of A. bruggeri and in 60 individuals of Q.
indecora. The behaviour of oral visitors was registered
at the beginning of the morning (7:00), the middle of the
day (12:00), and at the end of the afternoon (16:00) to
cover dierent periods of foraging, staying for 15 to 30
min in front of clumps of individuals of the two species.
Due to diculties in accessing the study area at night, it
was not possible to observe and record nocturnal visitors.
We recorded the time, frequency of visits, and the visitors’
behaviour to determine their role in the interaction (e.g.
pollinator, oral resource robber, herbivore). The activity of
the oral visitors was recorded through photos and videos.
Invertebrate visitors were collected with an entomological
net, euthanised in a vial containing cotton impregnated with
ether, stored in 70% alcohol, and taken to the laboratory
for posterior identication by specialists. Vertebrates were
recorded through photos for posterior identication.
RESULTS
Reproductive phenology and oral biology
In Aechmea bruggeri, the owering period started at the end
of the rainy season and extended into the dry season. There
was an overlap between owering and fruiting periods, and
the mature fruits were also available during the dry season
(g. 2). According to Newstrom et al. (1994), the individual
and populational owering phenological pattern can be
classied as annual (only one major cycle per year) with
intermediate duration (ranging from one to ve months),
showing an asynchrony between the individuals. According
to Gentry (1974), owering of this species ts in the steady-
state pattern, whereby the plants produce a few owers a day
over an extended period of time (usually a month or more).
The reproductive potential of this species (total number
of owers produced per individual) was 159 ± 28, with
around 11 owers opened per day. The inorescences
present pink-reddish peduncle bracts, brown-greenish sepals,
and lilac petals, with the inorescence about 1 m above the
ground. The corolla is tubular, with a mean length of 1.37
± 0.11 cm. The owers open from the base to the top of the
inorescence (g. 3). Anthesis starts around noon, and the
owers remain open and receptive for 24 h. At the moment
of the ower opening, the mean volume of nectar produced
per ower was 9.79 ± 3.34 µL, with a sugar concentration
ranging from 30 to 33%. The total amount of sugar found
was 3.49 mg per ower (table 1). A decrease in the mean
volume of nectar was observed four hours after anthesis
(0.70 ± 0.75 µL), reaching zero in senescent owers.
Quesnelia indecora completed the owering and fruiting
cycle during the dry period of the year. Similar to A. bruggeri,
the individual and populational owering phenological
patterns of Q. indecora were annual, with individuals
owering asynchronously during one to ve months (g. 2).
According to Gentry (1974), owering of this species also
ts in the steady-state pattern.
The species presents a mean reproductive potential of 8
± 3 owers per individual, with one ower opening per day.
The owers have pinkish bracts and purple sepals and petals,
and the mean length of the corolla was 4.42 ± 0.2 cm, with
the peduncle of the inorescence recurved and bending down
to approximately 10 cm from the ground. Anthesis occurred
around 6:10, with approximately 26 h of ower availability.
Quesnelia indecora does not show an order of ower
opening along the inorescence (g. 4). At the moment of
ower opening, the mean volume of nectar was 7.58 ± 3.96
µL, with a sugar concentration ranging from 31 to 33%. The
total amount of sugar in the nectar was 2.79 mg per ower
(table 1). A decrease in the mean volume of nectar was also
observed four hours after the beginning of the anthesis (1.83
± 1.84 µL), reaching zero in senescent owers.
Pollination treatments and reproductive success
Aechmea bruggeri presented a higher fruit set from hand
cross-pollination (88%) than the hand self-pollination
Figure 2 – Reproductive phenology of Aechmea bruggeri and Quesnelia indecora populations during the years 2019 and 2020.

20
Pl. Ecol. Evol. 155 (1), 2022
A. bruggeri Q. indecora
Corolla colour Lilac Purple
Corolla length (cm) 1.37 (± 0.11)
n = 10 owers
4.42 (± 0.2)
n = 3 owers
Number of owers per inorescence 159 (± 28)
n = 14 inorescences
8 (± 3)
n = 33 inorescences
Number of open owers per day 11 (± 1)
n = 2 inorescences
2 (± 0.5)
n = 2 inorescences
Floral anthesis time 12:00 06:10
Duration of anthesis 24 h 26 h
Nectar volume (µL) 9.79 (± 3.34)
n = 20 owers
7.58 (± 3.96)
N = 6 owers
Variation of sugar concentration in nectar (%) 30–33
n = 10 owers
31–33
n = 3 owers
Total amount of sugar (mg) 3.49 2.79
Table 1 – Data on the oral biology of Aechmea bruggeri and Quesnelia indecora. X (± s) = mean (± SD); n = sample size.
Figure 3 – Aechmea bruggeri inorescence with dierent
phenological stages. Flower buds (yellow); pre-anthesis (+) and
anthesis (*) owers (black); senescence owers (►) and fruits (×)
(red). Photograph by Matheus Rezende e Silva.
(9.6%) treatment. The mean number of seeds from hand
self-pollination was 21.06 ± 2.71, and from hand cross-
pollination it was 32 ± 2.23. The SCIf and SCIs indexes
were 0.10 and 0.65, respectively, which indicates partial
self-incompatibility. Under natural conditions, A. bruggeri
presented a fruit set ranging from 0% (due to predation)
in 2019 to 87% in 2020 (table 2). From the 14 individuals
selected in the eld for the open pollination treatment, four
of them presented signs of fruit predation. For non-predated
fruits, the mean number of seeds formed was 32 ± 3.02 (table
2).
Quesnelia indecora individuals also showed a higher
fruit set from hand cross-pollination treatment (100%) than
hand self-pollination (0%). The mean number of seeds from
hand self-pollination was 0, and from hand cross-pollination
it was 92 ± 6.21. The SCIf and SCIs indexes were 0.0, which
indicates total self-incompatibility for this species. Under
natural conditions, the species presented a fruit set ranging
from 0% in 2019 to 100% in 2020 (table 2). Quesnelia
indecora individuals also suered high fruit predation.
From the 33 individuals investigated, 18 presented partial
or total destruction of the infructescences (g. 5B). It was
not possible to identify which visitor was responsible for the
damage, and further observations are needed to verify the
possible presence of nocturnal predators. For non-predated
fruits, the mean number of seeds formed was 90 ± 7.91 (table
2).
Floral visitors
Eight species of oral visitors were collected, six visiting
A. bruggeri and three visiting Q. indecora, with one species
shared by both plants.
Figure 4 – Quesnelia indecora inorescence with dierent
phenological stages. Floral bud (●); anthesis owers (*); senescence
owers (►). Photograph by Matheus Rezende e Silva.

21
Silva, Barbosa & Faria, Reproductive ecology of two bromeliads from the Atlantic Forest
Hand
cross- pollination
Hand
self-pollination
Natural (open)
pollination
Mean number of
seeds/fruit
A. bruggeri
88%
(85/96)
(n = 2)
9.6%
(9/93)
(n = 2)
0%* (2019)
87% (2020)
(995/2236)
(n = 14)
32 (± 3.02)
50 fruits
Q. indecora
100%
(8/9)
(n = 3)
0
(0/9)
(n = 2)
0%* (2019)
100% (2020)
(79/288)
(n = 33)
90 (± 7.91)
50 fruits
Table 2 – Fruit set and average number of seeds (mean ± SD) of Aechmea bruggeri and Quesnelia indecora for controlled and natural
pollination treatments carried out between 2019 and 2020. The number of plants used in each treatment is given in parentheses (n = number
of individuals). The number of formed fruits and the number of tested owers is given in parentheses (fruits/owers). The * indicates that the
result 0% are from totally predated inorescences.
Figure 5 – Fruit predation (circles). A. Aechmea bruggeri. Curculionidae larvae feeding on the fruits. B. Quesnelia indecora. Photographs
by Matheus Rezende e Silva.
Aechmea bruggeri visitors – Visits started in the early
morning, around 7:00. The only vertebrates were two
hummingbird species. The violet-capped woodnymph
Thalurania glaucopis Gmelin, 1788 (g. 6A–B) made
frequent visits after anthesis, usually between 13:00 and
16:00. We observed the presence of male and female
individuals who take turns during visits that last between
3 and 5 s, with intervals of about 10 min between them. In
search of nectar, they inserted their beak into the ower,
removed the nectar, and consequently also the pollen. They
foraged on all the open owers of one individual and then
moved to another plant.
Less frequently, the planalto hermit Phaethornis pretrei
Lesson & Dellatre, 1839 presented the same behaviour as
T. glaucopis, but its visits had a shorter duration and longer
intervals between them, since aggressive behaviour of T.
glaucopis towards P. pretrei were recorded. It was often seen
that T. glaucopis landed on branches close to A. bruggeri
individuals, preventing other hummingbirds from visiting
the owers, as happened for P. pretrei. As they approached
the owers, they were attacked and chased by T. glaucopis,
preventing them from visiting any A. bruggeri plants.
Concerning invertebrate visitors, in the early morning
hours, the presence of a large number of male and female
Drosophila sp. ies were observed visiting the owers in pre-
anthesis (g. 6C–D). Curculionidae beetles were registered
on the bracts of the inorescence. Always an average of
ve individuals per inorescence, the Curculionidae waited

22
Pl. Ecol. Evol. 155 (1), 2022
Figure 6 – Aechmea bruggeri and Quesnelia indecora visitors. A. Male Thalurania glaucopis. B. Female T. glaucopis. C–D. Individuals of
Drosophila sp. E–G. Individuals of Curculionidae. H–I. Strymon oreala. J, L. Trigona cf. braueri. K. Eurybia pergea. All photographs by
Matheus Rezende e Silva.

23
Silva, Barbosa & Faria, Reproductive ecology of two bromeliads from the Atlantic Forest
for anthesis so that they could feed on the pollen (g. 6E–
G). For this, they used their legs to scrape the anthers to
remove pollen and discard them after feeding on the grains.
Larvae were also found in the fruits, causing partial or total
destruction (g. 5A).
The buttery Strymon oreala Hewitson, 1868 was
registered only twice, ying around the inorescence. They
landed on owers and inserted their proboscis into them
to collect nectar (g. 6H–I). The records were made in the
middle of the afternoon, between 13:00 and 16:00. The
stingless bee Trigona cf. braueri Friese, 1900 was registered
visiting only three times. This species foraged on the ower
in search of nectar located at the end of the corolla tube.
Because it is a small ower, the bee touched the anthers
and released the pollen, causing pollen transfer among
individuals.
Based on these behaviour patterns, it was possible to
classify Drosophila sp. and S. oreala as nectar thieving,
since they used the owers without necessarily transferring
pollen to other plants and without causing damage to oral
structures (Inouye 1980; Freitas 2018). The Curculionidae
beetles, which caused damage to the oral structures in
search of pollen, can be considered as pollen robbers. The
hummingbirds T. glaucopis and P. pretrei were considered as
pollinators since, to access the available resource, they end
up releasing pollen and transferring it to other plants. Trigona
cf. braueri showed both thieving and pollinator behaviour.
Quesnelia indecora visitors – For this species, three oral
visitors were registered: Trigona cf. braueri, Plebeia sp., and
Eurybia pergaea Geyer, 1832.
Trigona cf. braueri (g. 6J, L) was the most frequent,
with 23 visits. It was possible to register that, in addition to
accessing the nectar at the end of the tube through the corolla
entrance (consequently causing the transfer of pollen grains
to other individuals), this bee also damaged the base of the
ower to access the nectar resource (g. 3L). The same was
observed for the stingless bee Plebeia sp. The buttery E.
pergaea showed similar behaviour to S. oreala, using their
proboscis to access the nectar from the opened corolla (g.
6K). The dierence is that this species does not follow a
ower visitation pattern, as recorded for S. oreala.
Due to these behaviour patterns, E. pergaea was
classied as a nectar thieving, without transferring pollen to
other plants. Trigona cf. braueri and Plebeia sp. exhibited
behaviours of nectar robbers (causing damage to ower
structure) and occasionally behaved as pollinators, when
accessing nectar through the corolla opening.
DISCUSSION
There was no variation in the owering and fruiting seasons
for both species over the two consecutive reproductive
cycles. The continuous or steady-state phenology registered
for A. bruggeri and Q. indecora points out that these
bromeliads are an important food source for pollinators in
the study area. The steady-state owering pattern is often
related to plants visited by trapline foraging pollinators,
such as hummingbirds and many tropical bees, which are
characterised by having a repeated and xed visitation route
capable of covering long distances (Janzen 1971; Gentry
1974; Tello-Ramos et al. 2015). Visits by trapliners are
common in Bromeliaceae (Canela & Sazima 2003a; Kessler
et al. 2020) and have been reported for other species of
Aechmea (Canela & Sazima 2003a; Lenzi et al. 2006; Kamke
et al. 2011; Scrok & Varassin 2011; Pool-Chalé et al. 2018).
Aechmea bruggeri presented many of the oral traits
for ornithophily: scentless owers, tubular corolla, pink-
reddish inorescence peduncle bracts, and nectar secretion
during the whole diurnal anthesis (Faegri & van der Pijl
1979). However, the nectar sugar concentration is higher
and more related to buttery and/or moth pollination
(Krömer et al. 2008). Many bromeliad species are still little
known with respect to their insect pollinators (Krömer et
al. 2008). Although the lepidopteran S. oreala proved to be
a nectar thieving, a wide variety of other insects visited A.
bruggeri. Our results suggest that A. bruggeri is a generalist
plant, however, further analyses including observations of
nocturnal visitors will be important to elucidate the type of
pollination system of this species.
The individuals of A. bruggeri produced their largest
nectar volume at ower opening, and sugar concentration
did not change during anthesis. This dynamic of producing
a greater volume of nectar at the beginning of anthesis is
common among species of Bromeliaceae (Canela & Sazima
2003a; Machado & Semir 2006; Schmid et al. 2011; Aguilar-
Rodríguez et al. 2016). Nectar with high sugar concentrations
tends to attract and hold the attention of the plant visitor and,
when presented in small but sucient quantities, as observed
in A. bruggeri, it forces the animal to visit a maximum
number of owers, which may increase cross-pollination
rates (Baker 1975).
The average production of nectar by A. bruggeri (9.79
µL) is considered low, compared to other ornithophilous
species from the same genus with longer corolla tubes, such
as A. beeriana L.B.Sm. & M.A.Spencer (37 µL; Nara &
Webber 2002) and A. pectinata Baker (79.5 µL; Canela &
Sazima 2003a), besides bromeliads from other genera, such
as Billbergia horrida Regel and Tillandsia polystachia (L.) L.
(64.1 µL and 43.7 µL, respectively; Tagliati et al. 2018). Low
nectar volumes similar to those found in A. bruggeri were
observed for the short-corolla species A. caudata Lindm.
(15.5 µL; Kamke et al. 2011), where bees were registered as
an important visitor for pollination success, and A. bracteata
(Sw.) Griseb. (4.64 µL; Pool-Chalé et al. 2018), whose low
nectar production was directly related to the small size of the
corolla. Insects, mainly bees, are frequent ower visitors in
many short-corolla bromeliads with ornithophilous features
(Nara & Webber 2002; Araujo et al. 2004; Lenzi et al. 2006)
and might also have an important role in the pollination
system of these species due to their high frequency at the
owers. As bees were not observed visiting A. bruggeri
frequently, whose pollination was carried out mainly by
hummingbirds, the low nectar volume production for this
species is probably associated with its short corolla length.
Thalurania glaucopis was an eective pollinator of A.
bruggeri, corroborating the close mutualistic relationship of
this group of plants with hummingbirds of the Trochilidae
family (Sick 1985). Thalurania glaucopis was the main

24
Pl. Ecol. Evol. 155 (1), 2022
pollinator for other Aechmea species, such as A. pectinata
(Canela & Sazima 2003a), A. lindenii (Lenzi et al. 2006),
and A. nudicaulis (Schmid et al. 2011), as well as for
several other genera of Bromeliaceae (Kessler et al. 2020).
Its territorial and agonistic behaviour towards other species
of hummingbirds was also reported by Canela & Sazima
(2003a) for A. pectinata. Although P. pretrei is considered
the most eective hummingbird pollinator for many other
Bromeliaceae species, such as A. constantinii (Mez) L.B.Sm.
(Rios et al. 2010), Alcantarea turgida (Versieux & Wanderley
2007), Tillandsia geminiora Brogn., T. polystachia (L.)
L., and T. stricta Sol. (Tagliati et al. 2018), the frequency
of visits of the A. bruggeri owers, and consequently, its
importance in the total pollination success of this species is
diminished by the agonistic behaviour of T. glaucopis.
Although hummingbird pollination has been reported for
some terrestrial species of Quesnelia, such as Q. arvensis
(Vell.) Mez, Q. humilis Mez, and Q. lateralis Wawra (Kessler
et al. 2020), in our study, Q. indecora was not visited by this
group of birds, even though it presented some oral traits
characteristic for ornithophily, such as tubular corolla, pink-
reddish inorescence peduncle and bracts, and high sugar
concentration (Hainsworth & Wolf 1976; Faegri & van der
Pijl 1979). One explanation for the absence of hummingbird
visits in this species would be related to its ower display.
During the inorescence development, the peduncle tends to
bend down and stay close to the ground, hiding the owers
in the understory vegetation and positioning them out of
the hummingbird’s visual eld. Blem et al. (1997) reported
that the hummingbird Selasphorus rufus Gmelin showed a
preference for sucrose sources ranging from 3 to 25 m in
height, and this behaviour was interpreted to avoid predation.
In addition, they point out that taller owers are more
visible in the animal’s eld of view. Henderson et al. (2001)
demonstrated that not only this same hummingbird species
but also others have the cognitive ability to remember
the location of a certain ower, showing a preference for
those that are located higher. Since in Q. indecora, the
inorescences were approximately 10 cm from the ground,
this may explain the non-visitation by hummingbirds. The
colour attraction for each group of visitors can also be taken
into account. While bees and butteries prefer colours of
the yellow-pink-violet-blue and yellow-blue-red-orange
spectrum (Faegri & van der Pijl 1979; Westerkamp 1997;
Weiss 2009), respectively, birds tend to be attracted to
colours of the red spectrum (Varassin & Amaral-Neto 2014),
which even when present in the bracts of Q. indecora, are
blending into the foliage.
Pollination by dipterans is widely distributed among the
basal angiosperms, being found in Cabombaceae, all families
of Austrobaileyales, some Annonaceae, Monimiaceae,
Lauraceae, Winteraceae, Saururaceae, Piperaceae, and
Aristolochiaceae (Endress 2010). The owers pollinated
by this group are usually hermaphrodite and protogynous,
with odour production and temperature regulation
(thermogenesis). Nectar production is not common and
other oral resources are available, such as pollen, heat,
shelter, and places for reproduction (Larson et al. 2001;
Endress 2010). The interactions between bromeliads and
ies have been scarcely investigated, with records only
for the genus Aechmea (Dejean & Olmsted 1997). Schmid
et al. (2011) observed the presence of dipterans of the
suborder Brachycera visiting A. nudicaulis, removing nectar
from extraoral nectaries present in the sepals. Dejean &
Olmsted (1997) observed a large diversity of dipteran larvae
inhabiting the phytotelm of A. bracteata. For A. bruggeri, the
Drosophila individuals do not act as pollinators but rather as
nectar thieving.
The beetles of the family Curculionidae caused serious
damage to the inorescences of A. bruggeri. In addition to
feeding on the pollen and damaging the anthers, numerous
larvae were found in the fruits, directly interfering with
the plant’s reproductive success since they prevent the full
development of fruits and seeds. Previous studies already
indicated that Curculionidae adults and larvae feed on
various reproductive and vegetative plant structures, causing
serious losses in bromeliad populations (Frank 1999).
Albertoni et al. (2016) listed 18 species of beetles associated
with Hohenbergia augusta (Vell.) E.Morren and Vriesea
friburguensis Mez, being the rst list of beetles associated
with bromeliad species.
Schmid et al. (2010) recorded the predation of bromeliad
inorescences by Lepidoptera larvae of the genus Strymon,
among them, S. oreala, which fed on the developing fruits
of A. caudata and A. lindenii. After feeding, the larvae went
to the rosette of the species to start their pupal stage. These
authors also highlighted the importance of the Bromeliaceae
family in maintaining the lepidopteran fauna in the Atlantic
Forest. Although we only recorded S. oreala feeding on
A. bruggeri nectar, further observations are necessary to
conrm whether the species also uses A. bruggeri during its
larval development.
The interactions between organisms are not xed
but changeable according to the circumstances of the
environment (Zhang et al. 2015). Some visitors of the same
species can play a dual role, with mutualistic and antagonistic
behaviours. In this work, the stingless bees Trigona cf.
braueri and Plebeia sp. are both pollinators and robbers,
with this last behaviour frequently observed in Q. indecora,
where they feed on the nectar by damaging the base of the
owers.
Nectar robbers may have direct and indirect eects on
plant reproductive success, from damaging reproductive
organs to removing oral rewards without the benet of
pollination (Irwin et al. 2010). However, the presence
of nectar robbing is not proof of negative tness eects
(Fumero-Cabán & Meléndez-Ackerman 2013). Some
studies, for example, have found that owers with less nectar
can have increased cross-pollination (Lasso & Naranjo
2003; Irwin et al. 2010; Pelayo et al. 2011; Rojas-Nossa et
al. 2015; Hazlehurst & Karubian 2016). Considering that
Q. indecora is totally dependent on pollinators for fruit and
seed set, our data on the reproductive success suggest that
the costs of loss of oral rewards by nectar robbing can vary
from maximum to minimum, given the fruit set under natural
conditions. While the antagonistic behaviour of the bees can
lead to the total absence of fruit set in certain periods of the
reproductive phenology of this plant, these visitors also play
a fundamental role as pollinators, allowing 100% fruit set

25
Silva, Barbosa & Faria, Reproductive ecology of two bromeliads from the Atlantic Forest
in other periods. For A. bruggeri, the foraging behaviour of
T. cf. braueri does not seem to aect hummingbirds as the
eective pollinators of this species. As Aguilar-Rodríguez et
al. (2016) stated, the secondary pollinators may be important
as a ‘fail-safe’ system by which to guarantee the pollination
of some species.
Other plant-animal interactions such as herbivory, can
negatively aect the reproductive success of the species,
interfering in important processes (e.g. seed dispersal) for the
establishment of new individuals in their natural habit. In the
study area, seed dispersal of A. bruggeri is severely aected
by the predation of fruits by Curculionidae beetles. Fruit and
seed herbivory negatively aecting the reproductive success
have also been registered for other bromeliad species, such
as A. beeriana Smith & Spencer (Nara & Webber 2002),
A. lindenii (Lenzi et al. 2006), A. nudicaulis (L.) Griseb.
(Schmid et al. 2010), and Puya nitida Mez (Saldarriaga
2014). Failures in seed production and dispersal can bring
great risks to endangered species, especially for those which
have restricted distributions and high habitat specicity, such
as A. bruggeri.
Conservation eorts for the studied species should focus
primarily on habitat preservation and their pollinators.
Future studies involving observations of nocturnal visitors
and pollinator exclusion experiments will be important to
deepen the knowledge about the pollination ecology of these
species. However, considering the ecological importance
of the Bromeliaceae family in the Atlantic Forest, this
work reinforces how the interactions between animals and
bromeliads are important to sustain the biological diversity
of these forest remnants.
ACKNOWLEDGEMENTS
The authors would like to thank the collaborators who made
this study possible, in particular Leonardo Moreira Campos
Lima and Lúcio Moreira Campos Lima, managers of the
RPPN Chapadão da Serra Negra. SISBIO is acknowledged
for granting authorization to collect the species (SISBIO #
70113-2), and the Fundação de Amparo à Pesquisa do Estado
de Minas Gerais (FAPEMIG) is thanked for granting a MSc
scholarship to the rst author.
REFERENCES
Aguilar-Rodríguez P.A., MacSwiney G.M.C., Krömer T., García-
Franco J.G., Knauer A. & Kessler M. 2014a. First record of bat-
pollination in the species-rich genus Tillandsia (Bromeliaceae).
Annals of Botany 113: 1047–1055.
https://doi.org/10.1093/aob/mcu031
Aguilar-Rodríguez P.A., MacSwiney G.M.C., Krömer T. & García-
Franco J.G. 2014b. Pollen consumption by free-living mice.
Acta Theriologica 59: 361–365.
https://doi.org/10.1007/s13364-013-0164-7
Aguilar-Rodríguez P.A., Krömer T., García-Franco J.G. &
MacSwiney G.M.C. 2016. From dusk till dawn: nocturnal
and diurnal pollination in the epiphyte Tillandsia heterophylla
(Bromeliaceae). Plant Biology 18(1): 37–45.
https://doi.org/10.1111/plb.12319
Aguilar-Rodríguez P.A., Krömer T., Tschapka M., García-Franco
J.G., Escobedo-Sarti J. & MacSwiney G.M.C. 2019. Bat
pollination in Bromeliaceae. Plant Ecology & Diversity 12(1):
1–19. https://doi.org/10.1080/17550874.2019.1566409
Albertoni F.F., Steiner J. & Zillikens A. 2016. The associated
beetle fauna of Hohenbergia augusta and Vriesea friburgensis
(Bromeliaceae) in southern Brazil. Journal of Natural History
50(47–48): 2917–2939.
https://doi.org/10.1080/00222933.2016.1218079
Araujo A.C., Fischer E. & Sazima M. 2004. As bromélias na região
do Rio Verde. In: Marques O. & Duleba W. (eds) Estação
Ecológica Juréia-Itatins: ambiente físico, ora e fauna: 162–
171. Holos Editora, Ribeirão Preto.
Baker H.G. 1975. Sugar concentrations in nectars from
hummingbird owers. Biotropica 7(1): 37–41.
https://doi.org/10.2307/2989798
Basílio G.A., Barbosa D.E.F., Furtado S.G., Silva F.R. & Menini
Neto L. 2015. Community ecology of epiphytic Bromeliaceae
in a remnant of Atlantic Forest in Zona da Mata, Minas Gerais
State, Brazil. Hoehnea 42(1): 21–31.
https://doi.org/10.1590/2236-8906-68/2013
Benzing D.H. 2000. Bromeliaceae: prole of an adaptive radiation.
Cambridge University Press, Cambridge.
https://doi.org/10.1017/CBO9780511565175
Benzing D.H., Luther H. & Bennett B. 2000. Reproduction and
life history. In: Benzing D.H. (ed.) Bromeliaceae: prole of
an adaptive radiation: 245–326. Cambridge University Press,
Cambridge. https://doi.org/10.1017/CBO9780511565175.008
Blem C.R., Blem L.B. & Cosgrove C.C. 1997. Field studies of
rufous hummingbird sucrose preference: does source height
aect test results? Journal of Field Ornithology 68(2): 245–252.
Buzato S., Sazima M. & Sazima I. 2000. Hummingbird-pollinated
oras at three Atlantic forest sites. Biotropica 32(4): 824–841.
https://doi.org/10.1111/j.1744-7429.2000.tb00621.x
Canela M.B.F. & Sazima M. 2003a. Aechmea pectinata: a
hummingbird-dependent bromeliad with inconspicuous owers
from the rainforest in south-eastern Brazil. Annals of Botany
92(5): 731–737. https://doi.org/10.1093/aob/mcg192
Canela M.B.F. & Sazima M. 2003b. Florivory by the crab Armases
angustipes (Grapsidae) inuences hummingbird visits to
Aechmea pectinata (Bromeliaceae). Biotropica 35(2): 289–294.
https://www.jstor.org/stable/30044936
Carranza-Quiceno J.A. & Estévez-Varón J.V. 2008. Ecología de
la polinización de Bromeliaceae en el dosel de los bosques
neotropicales de Montaña. Boletín Cientíco Centro de Museos
Museo de Historia Natural 12: 38–47.
Cascante-Marín A., Wolf J.H.D. & Oostermeijer J.G.B. 2009.
Wasp orivory decreases reproductive success in an epiphytic
bromeliad. Plant Ecology 203(1): 149–153.
https://doi.org/10.1007/s11258-008-9522-y
Cascante-Marín A., Oostermeijer J.G.B., Wolf J.H.D. & den Nijs
J.C.M. 2005. Reproductive biology of the epiphytic bromeliad
Werauhia gladioliora in a premontane tropical forest. Plant
Biology 7: 203–209. https://doi.org/10.1055/s-2005-837584
Cavallari M.M. 2004. Estrutura genética de populações de
Encholirium (Bromeliaceae) e implicações para sua
conservação. Master’s thesis, Universidade de São Paulo
(ESALQ/USP), Brazil.
Christianini A.V., Forzza R.C. & Buzato S. 2012. Divergence
on oral traits and vertebrate pollinators of two endemic
Encholirium bromeliads. Plant Biology 15(2): 360–368.
https://doi.org/10.1111/j.1438-8677.2012.00649.x

26
Pl. Ecol. Evol. 155 (1), 2022
Dejean A. & Olmsted I. 1997. Ecological studies on Aechmea
bracteata (Swartz) (Bromeliaceae). Journal of Natural History
31: 1313–1334. http://doi.org/10.1080/00222939700770741
Drummond G.M., Martins C.S., Machado A.B.M., Sebaio F.A. &
Antonini Y. 2005. Biodiversidade em Minas Gerais: um atlas
para sua conservação. Second edition. Fundação biodiversitas,
Belo Horizonte.
Endress P.K. 2010. The evolution of oral biology in basal
angiosperms. Philosophical Transactions of The Royal Society
B 365: 411–421. https://doi.org/10.1098/rstb.2009.0228
Faegri K. & van der Pijl L. 1979. Principles of pollination ecology.
Elsevier, Oxford. https://doi.org/10.1016/C2009-0-00736-3
Filippon S., Fernandes C.D., Ferreira D.K., et al. 2012. Produção de
frutos para uso medicinal em Bromelia antiacantha (caraguatá):
fundamentos para um extrativismo sustentável. Revista
Brasileira de Plantas Medicinais 14: 506–513.
https://doi.org/10.1590/S1516-05722012000300013
Forzza R.C., Costa A., Siqueira Filho J.A., et al. 2015. Bromeliaceae
in lista de espécies da Flora do Brasil. Jardim Botânico do Rio
de Janeiro. Available from
http://floradobrasil.jbrj.gov.br/jabot/floradobrasil/FB66
[accessed 2 Dec. 2020].
Frank J.H. 1999. Bromeliad-eating weevils. Selbyana 20(1): 40–48.
Freitas L. 2018. Precisamos falar sobre o uso impróprio de recursos
orais. Rodriguésia 69(4): 2223–2228.
https://doi.org/10.1590/2175-7860201869446
Freitas L., Ribeiro P.C.C., Cancio A.S., et al. 2020. Population
demography, genetic variation and reproductive biology of
two rare and endangered Neoregelia species (Bromeliaceae).
Botanical Journal of the Linnean Society 192(4): 787–802.
https://doi.org/10.1093/botlinnean/boz110
Fumero-Cabán J.J. & Meléndez-Ackerman E.J. 2007. Relative
pollination eectiveness of oral visitors of Pitcairnia
Angustifolia (Bromeliaceae). American Journal of Botany
94(3): 419–424. https://doi.org/10.3732/ajb.94.3.419
Fumero-Cabán J.J. & Meléndez-Ackerman E.J. 2013. Eects of
nectar robbing on pollinator behaviour and plant reproductive
success of Pitcairnia angustifolia (Bromeliaceae). Plant
Species Biology 28(3): 224–234.
https://doi.org/10.1111/j.1442-1984.2012.00388.x
Galetto L. & Bernardello G. 2005. Rewards in owers: nectar. In:
Dafni A., Kevan P.G., Husband B.C. (eds) Practical pollination
biology: 231–325. Enviroquest, Ltd, Ontario.
Gentry A.H. 1974. Flowering phenology and diversity in tropical
Bignoniaceae. Biotropica 6(1): 64–68.
https://doi.org/10.2307/2989698
Givnish T.J., Barfuss M.H.J., Van Ee B., et al. 2014. Adaptive
radiation, correlated and contingent evolution, and net species
diversication in Bromeliaceae. Molecular Phylogenetics and
Evolution 71(1): 55–78.
https://doi.org/10.1016/j.ympev.2013.10.010
González-Gómez P.L. & Valdivia C.E. 2005. Direct and indirect
eects of nectar robbing on the pollinating behaviour of
Patagona gigas (Trochilidae). Biotropica 37(4): 693–696.
https://doi.org/10.1111/j.1744-7429.2005.00088.x
Gouda E.J., Butcher D. & Gouda C.S. continuously updated.
Encyclopaedia of bromeliads. Version 4. University Botanic
Gardens, Utrecht. Available from
http://bromeliad.nl/encyclopedia/ [accessed 25 Nov. 2020].
Grohme S., Steiner J. & Zillikens A. 2007. Destruction of oral
buds in the bromeliad Vriesea friburgensis by the phytophagous
larvae of the wasp Eurytoma sp. in southern Brazil
(Hymenoptera: Eurytomidae). Entomologia Generalis 30(2):
167–172. https://doi.org/10.1127/entom.gen/30/2007/167
Hainsworth F.R. & Wolf L.L. 1976. Nectar characteristics and food
selection by hummingbirds. Oecologia 25(2): 101–113.
https://doi.org/10.1007/BF00368847
Hazlehurst J.A. & Karubian J.O. 2016. Nectar robbing impacts
pollinator behaviour but not plant reproduction. Oikos 125(11):
1668–1676. https://doi.org/10.1111/oik.03195
Henderson J., Hurly T.A. & Healy S.D. 2001. Rufous
hummingbirds’ memory for ower location. Animal Behaviour
61(5): 981–986. https://doi.org/10.1006/anbe.2000.1670
Hornung-Leoni C.T., González-Gómez P.L. & Troncoso A.J. 2013.
Morphology, nectar characteristics and avian pollinators in ve
Andean Puya species (Bromeliaceae). Acta Oecologica 51: 54–
61. https://doi.org/10.1016/j.actao.2013.05.010
Inouye D.W. 1980. The terminology of oral larceny. Ecology
61(5): 1251–1253. https://doi.org/10.2307/1936841
Irwin R.E., Bronstein J.L., Manson J.S. & Richardson L. 2010.
Nectar robbing: ecological and evolutionary perspectives.
Annual Review of Ecology, Evolution, and Systematics 41: 271–
292. https://doi.org/10.1146/annurev.ecolsys.110308.120330
Janzen D.H. 1971. Euglossine bees as long-distance pollinators of
tropical plants. Science 171(3967): 203–205.
https://doi.org/10.1126/science.171.3967.203
Kaehler M., Varassin I.G. & Goldenberg R. 2005. Polinização
em uma comunidade de bromélias em oresta atlântica alto-
montana no estado do Paraná, Brasil. Revista Brasileira de
Botânica 28(2): 219–228.
https://doi.org/10.1590/s0100-84042005000200003
Kamke R., Schmid S., Zillikens A., Lopes B.C. & Steiner J. 2011.
The importance of bees as pollinators in the short corolla
bromeliad Aechmea caudata in southern Brazil. Flora 206(8):
749–756. https://doi.org/10.1016/j.ora.2010.12.004
Kessler M., Abrahamczyk S. & Krömer T. 2020. The role
of hummingbirds in the evolution and diversication of
Bromeliaceae: unsupported claims and untested hypotheses.
Botanical Journal of the Linnean Society 192(4): 592–608.
https://doi.org/10.1093/botlinnean/boz100
Krömer T., Kessler M., Lohaus G. & Schmidt-Lebuhn A.N. 2008.
Nectar sugar composition and concentration in relation to
pollination syndromes in Bromeliaceae. Plant Biology 10: 502–
511. https://doi.org/10.1111/j.1438-8677.2008.00058.x
Lasso E. & Naranjo M.E. 2003. Eect of pollinators and nectar
robbers on nectar production and pollen deposition in Hamelia
patens (Rubiaceae). Biotropica 35(1): 57–66.
https://doi.org/10.1111/j.1744-7429.2003.tb00262.x
Larson B.M.H., Kevan P.G. & Inouye D.W. 2001. Flies and
owers: taxonomic diversity of anthophiles and pollinators. The
Canadian Entomologist 133(4): 439–465.
https://doi.org/10.4039/Ent133439-4
Le Saout S., Homann M., Shi Y., et al. 2013. Protected areas and
eective biodiversity conservation. Science 342(6160): 803–
805. https://doi.org/10.1126/science.1239268
Lenzi M., de Matos J.Z. & Orth A.I. 2006. Variação morfológica e
reprodutiva de Aechmea lindenii (E.Morren) Baker var. lindenii
(Bromeliaceae). Acta Botanica Brasilica 20(2): 487–500.
https://doi.org/10.1590/S0102-33062006000200024
Lloyd D. & Schoen D. 1992. Self and cross-fertilization in plants. I.
Functional dimensions. International Journal of Plant Sciences
153(3): 358–369. https://doi.org/10.1086/297040
Machado C.G. & Semir J. 2006. Fenologia da oração e biologia
oral de bromeliáceas ornitólas de uma área da Mata Atlântica

27
Silva, Barbosa & Faria, Reproductive ecology of two bromeliads from the Atlantic Forest
do Sudeste brasileiro. Revista Brasileira de Botânica 29(1):
163–174. https://doi.org/10.1590/S0100-84042006000100014
Magalhães A.F.P., Maruyama P.K., Tavares L.A.F. & Martins R.L.
2018. The relative importance of hummingbirds as pollinators
in two bromeliads with contrasting oral specializations and
breeding systems. Botanical Journal of the Linnean Society
188(3): 316–326. https://doi.org/10.1093/botlinnean/boy050
Marques J.S., Tagliati M.C. & Faria A.P.G. 2015. Diurnal versus
nocturnal pollination success in Billbergia horrida Regel
(Bromeliaceae) and the rst record of chiropterophily for the
genus. Anais da Academia Brasileira de Ciências 87(2): 835–
842. https://doi.org/10.590/0001-3765201520140340
Marrero C., Garcia J.E., Farrera J., et al. 1996. Invertebrados
asociados a las bromélias en el Parque Nacional General Cruz
Carrilho (Guaramacal edo. Trujillo Venezuela). BioLlania 12:
143–144.
Milet-Pinheiro P., Domingos-Melo A., Olivera J.B., Schubert M.,
Dötterl S. & Machado I.C. 2021. A semivolatile oral scent
marks the shift to a novel pollination system in Bromeliads.
Current Biology 31: 860–868.
https://doi.org/10.1016/j.cub.2020.11.012
Nara A.K. & Webber A.C. 2002. Biologia oral e polinização de
Aechmea beeriana (Bromeliaceae) em vegetação de baixio na
Amazônia Central. Acta Amazonica 32(4): 571–588.
https://doi.org/10.1590/1809-43922002324588.
Newstrom L.E., Frankie G.W. & Baker H.G. 1994. A new
classication for plant phenology based on owering patterns
in lowland tropical rain forest trees at La Selva, Costa Rica.
Biotropica 26(2): 141–159. https://doi.org/10.2307/2388804
Oliveira A.M.S., Peixoto P.H.P., Barbosa B.C. & Faria A.P.G.
2021. Reproductive success, herbivory and ex situ conservation
of Neoregelia ibitipocensis (Bromeliaceae): an endemic and
endangered species from the Atlantic Forest. Australian Journal
of Botany 69(4): 237–246. https://doi.org/10.1071/BT21011
Oliveira-Filho A.T., Fontes M.A.L., Viana P.L., Valente A.S.M.,
Salimena F.R.G. & Ferreira F.M. 2013. O mosaico de
tosionomias do Parque Estadual do Ibitipoca. In: Forzza
R.C., Menini Neto L., Salimena F.R.G. & Zappi D.C. (eds)
Flora do Parque Estadual do Ibitipoca e seu entorno: 54–93.
Editora UFJF, Juiz de Fora.
Palacios-Mosquera Y., Mondragón D. & Santos-Moreno A.
2019. Vertebrate orivory of vascular epiphytes: the case of
a bromeliad. Brazilian Journal of Biology 79(2): 201–207.
https://doi.org/10.1590/1519-6984.176023
Pelayo R.C., Rengifo C. & Soriano P.J. 2011. Avian nectar robbers
of Passiora mixta (Passioraceae): do they have a positive
eect on the plant? Interciencia 8: 587–589.
Piacentini V.D.Q. & Varassin I.G. 2007. Interaction network and the
relationships between bromeliads and hummingbirds in an area
of secondary Atlantic rain forest in southern Brazil. Journal of
Tropical Ecology 23(6): 663–671.
https://doi.org/10.1017/S026646740700449X
Pool-Chalé M., Ramírez-Morillo I., Carnevali Fernández-Concha G.
& Hornung-Leoni C.T. 2018. Reproductive biology of Aechmea
bracteata (Sw.) Griseb. (Bromelioideae: Bromeliaceae). Plant
Biology 20(1): 113–120. https://doi.org/10.1111/plb.12645
Ribeiro M.C., Martensen A.C., Metzger J.P., Tabarelli M., Scarano
F. & Fortin M.-J. 2011. The Brazilian Atlantic Forest: a
shrinking biodiversity hotspot. In: Zachos F.E. & Habel J.C.
(eds) Biodiversity hotspots: distribution and protection of
conservation priority areas: 405–434. Springer, Berlin.
https://doi.org/10.1007/978-3-642-20992-5_21
Rios P.A.F., Da Silva J.B. & Moura F.D.B.P. 2010. Visitantes orais
de Aechmea constantinii (Mez) L. B. Sm. (Bromeliaceae) em
um remanescente da Mata Atlântica do Nordeste Oriental.
Biotemas 23(4): 29–36.
https://doi.org/10.5007/2175-7925.2010v23n4p29
Rocca M.A. & Sazima M. 2013. Quantity versus quality: identifying
the most eective pollinators of the hummingbird-pollinated
Vriesea rodigasiana (Bromeliaceae). Plant Systematics and
Evolution 299(1): 97–105.
https://doi.org/10.1007/s00606-012-0706-5
Rocha C.F.D., Cogliatti-Carvalho L., Nunes-Freitas A.F., et al.
2004. Conservando uma larga porção da diversidade biológica
através da conservação de Bromeliaceae. Vidalia 2(1): 52–72.
Rojas-Nossa S.V., Sánchez J.M. & Navarro L. 2015. Eects of
nectar robbing on male and female reproductive success of a
pollinator-dependent plant. Annals of Botany 117(2): 291–297.
https://doi.org/10.1093/aob/mcv165
Saldarriaga A.F. 2014. Estrategias de la reproducción sexual de
Puya nitida (NT) Mez. (Bromeliaceae) en el Parque Nacional
Natural Chingaza (Cundinamarca, Colombia). Master’s thesis,
Universidad Nacional de Colombia, Colombia.
Sazima I., Vogel S. & Sazima M. 1989. Bat pollination of
Encholirium glaziovii, a terrestrial bromeliad. Plant Systematics
and Evolution 168(3–4): 167–179.
https://doi.org/10.1007/BF00936097
Sazima I., Buzato S. & Sazima M. 1996. An assemblage of
hummingbird-pollinated owers in a montane forest in
southeastern Brazil. Botanica Acta 109(2): 149–160.
https://doi.org/10.1111/j.1438-8677.1996.tb00555.x
Schmid S., Schmid V.S., Kamke R., Steiner J. & Zillikens A. 2010.
Association of three species of Strymon Hübner (Lycaenidae:
Theclinae: Eumaeini) with bromeliads in southern Brazil.
Journal of Research on the Lepidoptera 42(2003): 50–55.
Schmid S., Schmid V.S., Zillikens A., Harter-Marques B. & Steiner
J. 2011. Bimodal pollination system of the bromeliad Aechmea
nudicaulis involving hummingbirds and bees. Plant Biology
13(SUPPL. 1): 41–50.
https://doi.org/10.1111/j.1438-8677.2010.00348.x.
Scrok G.J. & Varassin I.G. 2011. Reproductive biology and
pollination of Aechmea distichanta Lem. (Bromeliaceae). Acta
Botanica Brasilica 25(3): 571–576.
https://doi.org/10.1590/S0102-33062011000300009
Sick H. 1985. Ornitologia brasileira, uma introdução. Editora
Universidade de Brasília, Brasília.
Siqueira Filho J.A. & de Machado I.C.S. 2001. Biologia reprodutiva
de Canistrum aurantiacum Morren (Bromeliaceae) em
remanescente da Floresta Atlântica, Nordeste do Brasil. Acta
Botanica Brasilica 15(3): 427–443.
https://doi.org/10.1590/S0102-33062001000300011
Tagliati M.C., de Oliveira H.C. & de Faria A.P.G. 2018. Fenologia
reprodutiva, recursos orais e polinização de espécies de
bromeliaceae em um remanescente urbano de Floresta Atlântica
do sudeste brasileiro. Diversidade e Gestão 2(2): 139–150.
Teixeira R.L., Schineider J.A.P. & Almeida G.I. 2002. The
occurrence of amphibians in bromeliads from a southeastern
Brazilian restinga habitat, with special reference to
Aparasphenodon brunoi (Anura, Hylidae). Brazilian Journal of
Biology 62(2): 263–268.
https://doi.org/10.1590/S1519-69842002000200010
Tello-Ramos M.C., Hurly T.A. & Healy S.D. 2015. Traplining in
hummingbirds: ying short-distance sequences among several
locations. Behavioural Ecology 26(3): 812–819.
https://doi.org/10.1093/beheco/arv014

28
Pl. Ecol. Evol. 155 (1), 2022
Varassin I.G. 2002. Estrutura espacial e temporal de uma
comunidade de Bromeliaceae e seus polinizadores em Floresta
Atlântica no sudeste do Brasil. PhD thesis, Universidade
Estadual de Campinas, Brazil.
Varassin I.G. & Sazima M. 2000. Recursos de Bromeliaceae
utilizados por beija-ores e borboletas em Mata Atlântica no
Sudeste do Brasil. Boletim do Museu de Biologia Mello Leitão
11/12: 57–70.
Varassin I.G. & do Amaral-Neto L.P. 2014. Atrativos. In: Rech A.R.,
Agostini K., Oliveira P.E. & Machado I.C. (eds) Biologia da
polinização: 151–170. Editora Projeto Cultural, Rio de Janeiro.
Velásquez-Noriega P., Mayta C., Cuba E., García E.E., Montaño-
Centellas F. & Krömer T. 2020. Floral ecology and oral
visitors of Puya atra (Bromeliaceae), a Bolivian endemic plant.
Ecología en Bolivia 55(1): 36–45.
Versieux L.M. & Wanderley M.G.L. 2007. Two new species of
Alcantarea (Bromeliaceae, Tillandsioideae) from Brazil.
Brittonia 59(1): 57–64. https://www.jstor.org/stable/4497346
Weiss M.R. 2009. Vision and learning in some neglected
pollinators: Beetles, ies, moths, and butteries. In: Chittka L.
& Thomson J.D. (eds) Cognitive ecology of pollination: 171–
190. Cambridge University Press, New York.
https://doi.org/10.1017/CBO9780511542268.010
Westerkamp C. 1997. Keel blossoms: bee owers with adaptations
against bees. Flora 192(2): 125–132.
https://doi.org/10.1016/S0367-2530(17)30767-3
Zhang Z., Yan C., Krebs C.J. & Stenseth N.C. 2015. Ecological
non-monotonicity and its eects on complexity and stability
of populations, communities and ecosystems. Ecological
Modelling 312: 374–384.
https://doi.org/10.1016/j.ecolmodel.2015.06.004
Communicating editor: Renate Wesselingh.
Submission date: 10 May 2021
Acceptance date: 13 Oct. 2021
Publication date: 30 Mar. 2022