Revista de Biología Tropical, ISSN: 2215-2075, Vol. 73 (S2): e64685, mayo 2025 (Publicado May. 15, 2025)

The role of nocturnal and diurnal pollinators in the commercial

production of Dragon Fruit crops in Costa Rica

Jordán Villegas-Murillo1; https://orcid.org/0009-0007-7809-9310

Mauricio Fernández Otárola1,2,*; https://orcid.org/0000-0001-9240-7569

1. School of Biology, University of Costa Rica, San José, Costa Rica. jvillegasmurillo@gmail.com

2. Biodiversity and Tropical Ecology Research Center (CIBET), University of Costa Rica, San José, Costa Rica

(*Correspondence: mauricio.fernandez@ucr.ac.cr)

Received 31-VIII-2024. Corrected 17-III-2025. Accepted 25-III-2025.

ABSTRACT

Introduction: Dragon fruit (or pitahaya) cultivation is of growing interest in Central America. Several species

of the genus Selenicereus (Cactaceae) are cultivated around the world, with S. costaricensis being the only species

native to this region. The pollination syndrome of the species partially matches chiropterophily and phalaenoph-

ily, but floral visitors and their effect on commercial fruit production is not well known.

Objective: To analyze the reproductive system, the floral visitors and their effect on fruit size in plantations of

S. costaricensis in Costa Rica.

Methods: We recorded the anthesis period, stigmatic receptivity, anther dehiscence and nectar production. We

conducted pollination experiments (manual self- and cross-pollination) and evaluated the effect of nocturnal and

diurnal natural pollination on fruit production and traits related to fruit size and weight. Flower visitors were

filmed, identified, quantified, and their behavior was documented.

Results: Plants were self-compatible. All pollination treatments produced viable fruits, but there were differ-

ences in fruit weight and size. Natural pollination, nocturnal and diurnal combined, produced the biggest fruits,

while self-pollination the smallest. The fruits generated by either nocturnal or diurnal pollinators were of good

commercial size. Hawk moths and bees were the main pollinators; no visits by bats were detected. The flowers

produced no measurable nectar volume. Mass visitation by bees in periods of 15–20 minutes at dawn was suf-

ficient to produce large fruits.

Conclusions: Diurnal and nocturnal pollinators are equally effective producing highly profitable fruits. The

management of Apis mellifera and native solitary bee species is recommended to increase production efficiency

in these plantations. Manduca rustica hawk moths seem the natural pollinators of S. costaricensis, and bats played

no role in the production of fruits.

Keywords: Pitahaya, reproductive biology, reproductive system, fruit production, hawk moth pollination, bee

pollination, Selenicereus costaricensis.

RESUMEN

El papel de los polinizadores diurnos y nocturnos en la producción comercial del cultivo

de pitahaya en Costa Rica

Introducción: El cultivo de la pitahaya es de creciente interés en Centroamérica. Varias especies del género

Selenicereus (Cactaceae) se cultivan alrededor del mundo, siendo S. costaricensis la única especie nativa de esta

https://doi.org/10.15517/rev.biol.trop..v73iS2.64685

SUPPLEMENT

SECTION: MUSEUM

2Revista de Biología Tropical, ISSN: 2215-2075 Vol. 73 (S1): e64685, mayo 2025 (Publicado May. 15, 2025)

INTRODUCTION

Animal pollination is highly important

at an ecological level (Kearns et al., 1998).

The number of flowering plant species pol-

linated by animals is > 87% worldwide and,

in tropical communities, it can reach 95%

(Ollerton et al., 2011).

Ecosystem services provided by pollinators

around the globe have been studied with the aim

of quantifying the contributions of animals to

human activities (Lautenbach et al., 2012; Porto

et al., 2020). Crop pollination has considerable

economic value (Ricketts et al., 2004; Giannini

et al., 2015; Borges et al., 2020; Porto et al.,

2020; Basualdo et al., 2022), and it is estimated

that two-thirds of crops require animal-assisted

pollination (Nabhan & Buchmann, 1997; Kevan

& Phillips, 2001). Moreover, several crops can

increase their production when supported by

animal-assisted pollination, which is beneficial

for the fruit and seed trade (Klein et al., 2007;

Garibaldi et al., 2013).

The notion of pollination syndromes is the

foundation of multiple studies regarding floral

diversity and pollination ecology, and their

accuracy in predicting plant pollinators has

been recently reviewed (Rosas-Guerrero et al.,

2014). However, plants may also benefit from

secondary pollinators (Johnson & Steiner,2000;

Fenster et al.,2004), and pollination syndromes

have been discussed several times, especially

when there is doubt or mismatch between pre-

dicted and observed pollinators (Waser et al.,

1996; Ollerton et al., 2009; Valverde-Espinoza

et al., 2021). How this mismatch affects the eco-

nomic revenue and management of commercial

crops requires further exploration. Different

groups of pollinators can benefit from different

management strategies in the local landscape in

agroecosystems or might depend on the pres-

ervation of natural spaces linked to plantations

(Klein et al., 2007).

In the Cactaceae family, chiropterophily is

commonly reported in species with nocturnal

anthesis with records for more than 150 species

(Vogel, 1968; Dobat & Peikert-Holle, 1985).

There are some cases where species considered

‘bat-pollinated’ have mixed systems, usually

involving other animals such as insects from

different orders, both diurnal and nocturnal

(Muchhala et al., 2009; Queiroz et al., 2015;

Queiroz et al.,2016; Muñiz et al., 2020). Pha-

laenophily (pollination by moths) has also

been reported in Cactaceae, and Haber (1983)

mentioned that hawk moths (Sphingidae) are

región. El síndrome de polinización de la especie coincide parcialmente con la quiropterofilia y la falenofilia, pero

los visitantes florales y su efecto en la producción comercial no son bien conocidos.

Objetivo: Analizar el sistema reproductivo, los visitantes florales y su efecto en el tamaño de los frutos en planta-

ciones de S. costaricensis en Costa Rica.

Métodos: Medimos el período de antesis, la receptividad estigmática, la dehiscencia de las anteras y la producción

de néctar. También realizamos experimentos de polinización (autopolinización manual y polinización cruzada

manual) y polinización natural nocturna y diurna, y evaluamos su efecto en la producción de frutos y en su

tamaño y peso. Los visitantes florales fueron filmados, identificados, cuantificados y su comportamiento fue

documentado.

Resultados: Las plantas fueron auto compatibles. Todos los tratamientos de polinización produjeron frutos via-

bles, pero hubo diferencias en su tamaño. La polinización natural, nocturna y diurna combinadas, produjo los fru-

tos más grandes y la autopolinización manual los más pequeños. Los frutos generados por polinizadores diurnos

o nocturnos fueron de un buen tamaño comercial. Las flores no produjeron un volumen de néctar medible. Las

visitas masivas de abejas en periodos de 15-20 min al amanecer fueron suficientes para producir frutos grandes.

Conclusiones: Los polinizadores diurnos y nocturnos son igualmente efectivos produciendo frutos altamente

rentables. Se recomienda el manejo de Apis mellifera y especies de abejas solitarias nativas para aumentar la efi-

ciencia de producción en estas plantaciones. El esfíngido Manduca rustica parece ser el polinizadores natural de

S. costaricensis. Los murciélagos no jugaron ningún papel en la producción de frutos.

Palabras clave: Pitahaya, biología reproductiva, sistema reproductivo, producción de frutos, esfíngidos, abejas,

Selenicereus costaricensis.

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 73 (S2): e64685, mayo 2025 (Publicado May. 15, 2025)

pollinators of S. costaricensis in natural popula-

tions. Selenicereus costaricensis flowers present

ambiguous traits between chiropterophily and

phalaenophily (sensu van der Pijl, 1961). The

flowers are well exposed, large and strong,

creating good space for landing, with large

amounts of pollen, such as those of chirop-

terophilous species, but are also sweetly per-

fumed and do not contain large amounts of

nectar as sphingophilous flowers. However,

the peduncle is wide and does not form a clean

tube, such as in sphingophilous flowers, and

the bulky floral morphology avoids hovering as

a way to reach any reward. Another important

aspect to consider is that although anthesis

is nocturnal, preliminary observations have

shown that these flowers remain open for a

brief period after dawn, allowing visitation by

diurnal animals in plantations (MFO, personal

observations), but this seems not to happen in

natural populations (Haber, 1983).

Very few cacti species are commercially

planted for their fruits, with dragon fruits or

pitahayas (such as S. undatus, S. megalanthus,

and S. costaricensis) being the most relevant

product of this kind. Understanding the pol-

lination mechanisms of these species is impor-

tant for the management of plantations to

increase their productivity.

This work explored the contribution of

floral visitors to the pollination of commercially

planted individuals of Selenicereus costaricensis,

the Costa Rican pitahaya, a species of commer-

cial interest typically associated with nocturnal

pollinators. We characterize the impact that

different groups of pollinators have on crop

production and fruit traits related to their com-

mercial value. Our hypothesis is that secondary

pollinators (diurnal insects) contribute to the

pollination of this dragon fruit species and

have an impact on commercial production,

measurable through fruit weight and size. This

will lead to a better understanding of the pol-

lination dynamics of crop cacti, contributing to

the development of better production strategies

that will benefit local fruit producers.

MATERIALS AND METHODS

Study site: The study was conducted in

three commercial plantations in Costa Rica in

the surroundings of Liberia, Guanacaste prov-

ince: Curubandé (10°43’46.14”N, 85°25’31”W;

322 m asl; 0.5 ha) and El Salto (10°35’01.518”N

85°22’53.46”W; 305 m asl; 0.75 ha), both in

the lower premontane forest transitional to

dry forest life zone (Holdridge, 1967). The

third plantation, Plancillo, is in the borders of

the Central Valley (Alajuela province, Atenas:

9°58’54.96”N, 84°25’4.848”W; 970 m asl; 0.5

ha) in the transitional humid premontane forest

to dry forest life zone (Holdridge, 1967).

The Curubandé plantation is next to the

Rincón de la Vieja National Park and is sur-

rounded by large patches of forest and small

towns. El Salto is in a more disturbed area near

large farms, cattle ranches, and towns, but still

surrounded by small patches of gallery forests.

Both plantations consist of evenly distributed

wooden or cement poles where the plants

climb up to approximately 1.5 m from the

ground (Fig. 1). The Curubandé plantation is

composed mostly of genetically different plants

(plants with different origins according to the

farmers) and a small portion of clones; the El

Salto plantation is made up mostly of clones.

The Plancillo plantation consists of small living

trees grouped along several linear paths where

the cacti climb, ranging from ground level to

approximately 2 m tall. This plantation is in

a rural area and is surrounded by small cattle

farms, houses, streets and patches of gallery for-

ests. This plantation has a similar proportion of

properly identified genetically different plants

and vegetatively propagated clones.

The mean annual precipitation in Liberia

is 1 653 mm (range: 1 300-1 700). The mean

annual temperature is 27.5 °C (Villalobos &

Retana, 2 000). The annual precipitation at

Plancillo varies between 1 700 and 2 500 mm

per year (mean 2 000 mm). The average annual

temperature is 22 °C. The marked dry season

lasts from the end of November to mid-May,

and the rainy season lasts for the remainder of

the year.

4Revista de Biología Tropical, ISSN: 2215-2075 Vol. 73 (S1): e64685, mayo 2025 (Publicado May. 15, 2025)

The studied species: Selenicereus costari-

censis (F.A.C. Weber) S. Arias & N. Korotkova

(formerly in the genus Hylocereus) is a pitahaya

or dragon fruit, native from Nicaragua to Pana-

ma (Hammel, 2020). This epiphytic or hemiepi-

phytic cactus develops an elongated triangular

stem and is characterized by hermaphroditic,

white, aromatic flowers with herkogamy and

nocturnal anthesis that lasts for only one night.

The large funnel-shaped flowers are approxi-

mately 30 cm in length and diameter at the

corolla mouth, with a high number of stamens

(Fig. 2a). These plants are prone to pollination

by nocturnal animals (Haber, 1983).

In plantations from Costa Rica, flowering

occurs during the rainy season (in a five-month

period between May and September) in three

to seven massive bloom events, lasting approxi-

mately three nights each. Fruits are ellipsoidal

fuchsia false berries with fleshy bracts and

numerous small seeds imbedded in a fleshy

magenta or reddish purple mesocarp (Mizrahi

et al., 2002; Centurión-Yah et al., 2008; Fig. 2b).

The fruits of S. costaricensis are of commer-

cial interest for human consumption, ranging

between 0.1 and > 0.7 kg, and are of growing

economic importance in Mesoamerica.

All the following methodologies were

implemented in the three plantations. The

data were collected during eight mass bloom

events between 2019 and 2020 (Supplementary

Table S1).

Fig. 1. Commercial dragon fruit (pitahaya) plantations in (A) Curubandé, Liberia, Guanacaste (open flowers in the early

morning); (B) Plancillo, Atenas, Alajuela (late morning, after flower withering); and (C) El Salto, Liberia, Guanacaste (early

morning) in Costa Rica.

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 73 (S2): e64685, mayo 2025 (Publicado May. 15, 2025)

Anthesis period: The anthesis period was

divided into three stages: 1. opening of the

bud, 2. full extension of the flower tepals, and

exposure of the androecium and gynoecium,

and 3. corolla decay and wilting. The data from

345 flowers were recorded for flower opening,

and those from 325 flowers were recorded for

wilting. The flowers were marked in the peti-

ole with flagging tape, and observations were

made every 10 minutes from 18:00 h - 21:00 h

to record the start time and duration of stages 1

and 2. Observations resumed at 6:00 h and con-

tinued until 8:00 h for stage 3. The percentage

of flowers in each stage was calculated for each

observation time.

Stigmatic receptivity and anther dehis-

cence: Observations were made every hour

on open flowers from 18:00 h until midnight,

since few changes were observed after 20:00

h. One more observation was done at 4:00 h

to check the late state of the flower. Anthers

dehiscence was inspected visually with a mag-

nifier lens (30X) to determine when the pollen

was available. To determine receptivity, two

small drops of hydrogen peroxide (H2O2) were

placed on the stigma, and any evidence of bub-

bling was observed with a magnifying glass

(30X) as an indication of enzymatic activity.

The stigma is large in S. costaricensis (see Fig.

2) and has a lobulated structure that allows

placing of small drops at each sampling time in

different sections of the stigma and tracking the

same flower for a long period. The percentages

of receptive flowers and flowers with available

pollen during each observation were calculated.

Nectar production: During all the sam-

pling nights (see Table S1), 10 flowers from

different plants were covered with voile exclu-

sion bags to prevent nectar extraction by visit-

ing animals. Two-cc syringes with needles and

glass capillaries were used for nectar extraction

at 20:00 h, 00:00 h, and 4:00 h to measure the

nectar volume throughout the night.

Pollination experiments: During four

nights in the Plancillo and El Salto plantations,

Fig. 2. Flower and fruit of Selenicereus costaricensis (Cactaceae). A. Longitudinal section of a flower. B. Ripe fruit dissected

(left) and in its original appearance (right); s= area where the pulp cubes were extracted for seed counting.

6Revista de Biología Tropical, ISSN: 2215-2075 Vol. 73 (S1): e64685, mayo 2025 (Publicado May. 15, 2025)

five groups of 8-10 flower buds (depending on

availability) were selected for five different pol-

lination treatments.

1. Manual self-pollination: To test for self-

compatibility, buds were marked and cove-

red with voile fabric bags (applied in all the

treatments) at 17:00 h to prevent any visi-

tors from contacting the flower. When the

stigma was receptive, anthers with pollen

from the same flower were rubbed against

the stigmatic surface, and the flower was

bagged again.

2. Manual cross-pollination: Buds were

emasculated using scissors and bagged at

17:00 h. When the stigma was receptive,

it was rubbed with anthers from another

genetically different plant (one individual)

and bagged. The bags were removed the

next morning when the flowers were com-

pletely withered, and the stigma had lost

receptivity.

3. Nocturnal pollination: To evaluate the

contribution of nocturnal pollination to

reproductive success, flowers were marked

and kept on display throughout the night

to make them available for nocturnal visi-

tors and bagged at 4:00 h to avoid visits

by daytime foragers. The bags were remo-

ved after the corolla wilted the following

morning.

4. Diurnal pollination: To evaluate the con-

tribution of pollinators that are active at

dawn and early morning, flower buds

were marked and bagged at 17:00 h. The

next day, bags were removed at 4:30 h to

allow the visit of potential pollinators until

flowers completely withered.

5. Control: Nonmanipulated flowers exposed

to nocturnal and diurnal pollination were

marked to estimate natural pollination suc-

cess (nocturnal and diurnal).

All the fruits produced in the treatments

were harvested and measured four weeks after

anthesis. Fruit size variables (weight, length,

width, and seed density) were measured. The

fruits were weighed with a Snowrex EQ 1200

balance with a precision of 0.02 g. The fruits

were cut longitudinally in two halves, and the

maximum length and width of the pulp were

measured with a digital caliper (Mitutoyo 500-

197-30B, precision of 0.02 mm). To determine

the density of the seeds, a 1 cm3 portion was

taken from the center of each fruit, and the

seeds were counted (Fig. 2b). This value was not

extrapolated to the whole fruit content because

the volume of the fruit pulp could not be cal-

culated with confidence. The average value of

each fruit trait variable among treatments was

compared with one-way ANOVA, and a pos-

teriori Tukey test was used to determine the

treatments that differed from each other.

Nocturnal and diurnal visitors: Three

digital cameras equipped with night vision

(Sony HDR-CX700) aimed at different clusters

of flowers (1.5 m apart) continuously video-

recorded from 19:00 h to 00:00 h and from 4:30

h to 7:00 h. In total, 87 flowers were recorded

(25 at Curubandé, 41 at Plancillo, and 21 at

El Salto). For each visit, we recorded the time

of occurrence, duration (when possible), and

whether the visitors had contacted the repro-

ductive organs. Insect visitors were collected

only after the cameras had finished filming. The

insects were identified to the lowest taxonom-

ic level possible, and their abundance (visit/

hour/flower) at the order level (Hymenoptera,

Lepidoptera and Coleoptera; other orders were

excluded due to their low abundance) was com-

pared between plantations using a GLM with a

negative binomial distribution.

One part of the diurnal visitation was mas-

sive and required a different methodology to

estimate the number of insects visiting the flow-

ers simultaneously. The mass visits lasted for

approximately 20 min. The videos were played,

and individual images were extracted at five

different times, evenly spaced, during the whole

mass visitation event (approximately minutes

1, 5, 10, 15, and 20). The number of insects

in every image was counted, and the average

number of individuals per minute for each

flower was calculated. To measure the average

duration of the visits, 15 random insects were

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 73 (S2): e64685, mayo 2025 (Publicado May. 15, 2025)

registered. The parameters described above

were compared for each site using Kruskal‒

Wallis tests. All analyses were performed in R

(R Core Team, 2016).

RESULTS

Floral biology: Flower buds began to open

around sunset, on average at 18:31 (±15:49

min), reached full anthesis approximately 1.25

hours later and withered after dawn at 6:41

(±18:29 min). The complete anthesis period

spanned approximately 12:09 h (±24:33 min).

No pollen was available at 19:00 h, but all flow-

ers had pollen available at 20:00 h. A similar

pattern was observed for stigmatic receptivity,

with a peak occurring at midnight (Supplemen-

tary Fig. S1).

Measurements for nectar production were

made on 80 flowers during all the samplings.

On no occasion was nectar obtained. Dissec-

tions of the flowers during each sampling night

did not show appreciable accumulation of nec-

tar in any floral structure.

Pollination experiments: All the flow-

ers used for all the treatments produced via-

ble fruits, and the plants were self-compatible

(n=26 per treatment, except for manual cross-

pollination, n=16). Fruit weight was highly cor-

related with fruit length (r=0.93, p<0.001) and

width (r=0.90, p<0.001), so we used weight as

the variable to measure the effect of pollination

treatments on fruit size. The fruits from flowers

exposed to natural pollination, that included

nocturnal and diurnal visitors, were on aver-

age the heaviest of all the treatments, whereas

fruits from manually self-pollinated flowers

presented the lowest weight (F4, 113 = 76.11;

p<0.001; Fig. 3a). There was also an effect of the

plantation on fruit size (F1., 113 = 4.68; p=0.033).

Seed density showed a weaker correlation

with fruit weight (r=0.50, p<0.001) and varied

between the different pollination treatments

(F4, 108 = 15.73; p<0.001). There were more

seeds in the natural and nocturnal pollination

treatments than in the manual self-pollination

and diurnal pollination treatments. Manual

cross-pollination did not differ from any other

treatment (Fig. 3b). There was no effect of the

plantation on the seed density (F1, 108 = 0.02;

p<0.90).

Floral visitors: A total of 32.88 hours of

night video and 9.52 hours of day video were

analyzed.

1. Nocturnal visitors: A total of 118 indi-

vidual visits by insects were recorded.

Lepidoptera (40%), Coleoptera (36%),

and Hymenoptera (18%) were the most

frequent visitors (a list of the visitors is

provided in supplementary Table S2). Iso-

lated visits by Hemiptera and Diptera were

also recorded (6%), but in no case did

they contact the reproductive parts of the

flowers. The abundance of insects visi-

ting the flowers per hour varied between

plantations and between the orders of the

Fig. 3. Fruit traits of Selenicereus costaricensis (Cactaceae)

from pollination treatments in two crop farms from

Costa Rica. Data for: a. average weight. b. average seed

density. The error bars indicate the standard deviation. The

treatments with the same letter do not differ significantly

(Tukey test with a confidence level of 95%).

8Revista de Biología Tropical, ISSN: 2215-2075 Vol. 73 (S1): e64685, mayo 2025 (Publicado May. 15, 2025)

insects, and the interaction between these

factors was also significant (Table 1 and

Table 2).

Beetles generally entered flowers through

tepals and stay at the base of the corolla

for long periods; contact with reproduc-

tive structures was uncommon, but their

transit through flowers frequently moves

anther filaments and might contribute to

self-pollination. Regarding Lepidoptera,

almost all the visits were individuals from

the Sphingidae family, except for two occa-

sional visits by smaller unidentified moths.

Hawk moths flew through the corolla ope-

ning, entering and disappearing between

the reproductive structures on all occasions

(see supplementary video), as described by

Haber (1983). The average duration of

flower visits did not vary according to the

taxonomic group (χ2=1.49; d.f.=2; p>0.05).

No visit by bats or any other vertebrate was

recorded or observed in situ.

2. Diurnal visitors: During the samplings at

the Plancillo plantation, evening visits by

stingless bees (Trigona spp.) to unopened

flower buds were recorded. Visits by noc-

turnal bees (Megalopta amoena) were also

recorded at the two sites in the dry forest

between 20:00 h and 22:00 h and one

hour before sunrise (~ 4:00 h), when the

bees frequently contacted the stigma and

anthers. Xylocopa gualanensis also visi-

ted the flowers before sunrise, starting at

~4:30 h.

Massive honeybee (Apis mellifera) visita-

tion events involving thousands of indi-

viduals were registered in the plantations,

occurring just before dawn (starting at

approximately 4:45 –5:00 h). Groups of

bees collectively search for pollen for 15 to

20 minutes (see supplementary video). In

each event, once the resource was deple-

ted, the swarm abandoned the plantations.

In 66% of the cases, the bees contacted

the stigma, and in 34%, they came into

contact strictly with stamens or parts of

the perianth. The lowest visitation rate

during massive visits was recorded in El

Salto (22 bees/min), which differed signifi-

cantly from that at the other two sites (75

bees/min in Curubandé and 60 bees/min in

Table 1

Frequency of insect visitors to the flowers of S. costaricensis according to the plantation, order, and sampling effort (hours

recorded).

Plantation Order Visits Sampled hours Number of flowers

Plancillo Coleoptera 6 18.2 41

Hymenoptera 3

Lepidoptera 9

Curubandé Coleoptera 35 6.9 25

Hymenoptera 15

Lepidoptera 7

El Salto Coleoptera 1 12.5 21

Hymenoptera 3

Lepidoptera 31

Table 2

Results for a GLM comparing the frequency of visit per hour per flower by insects in the orders Hymenoptera, Lepidoptera

and Coleoptera, to El Salto, Plancillo and Curubandé plantations from Costa Rica.

Factor Df Deviance Residual deviance p

Order 2, 144 8.47 131.50 0.014

Plantation 2, 142 20.49 111.01 <0.001

Order x Plantation 4, 138 19.59 91.95 <0.001

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 73 (S2): e64685, mayo 2025 (Publicado May. 15, 2025)

Plancillo; χ2=39.05; d.f.=2; p<0.05). Much

lower abundances of other bee genera were

also detected during the morning before

wilting (Supplementary Table S2).

DISCUSSION

Floral Biology: Selenicereus costaricensis

presents nocturnal anthesis, as do other spe-

cies in the genus (Chang et al., 1997; Valien-

te-Banuet et al., 2007). Weiss et al. (1994)

proposed a general anthesis pattern for multiple

species of Selenicereus (reported as Hylocer-

eus; including S. costaricensis), and our study

revealed that S. costaricensis flowers were fully

open approximately one hour earlier than pre-

viously reported, indicating geographic and

population variability.

The flowers of S. costaricensis showed no

dichogamy and released pollen while their stig-

mas became receptive, approximately 1 h after

anthesis began. Overlap in the maturation of

reproductive organs has been described for S.

undatus in Mexican crops (Valiente-Banuet et

al., 2007). However, Weiss et al. (1994) reported

that in wild populations of S. costaricensis,

S. polyrhizus and S. undatus anthers dehisce

before flower opening. The same was reported

by Corona (2018) for S. ocamponis and S. pur-

pusii in Mexico.

Homogamy or adichogamy is an uncom-

mon trait among angiosperms with perfect

flowers and is present in only 14.4% of the spe-

cies (Bertin & Newman, 1993). The effects of

adichogamy on commercial production require

further investigation; a longer floral opening

period maximizes the probability of pollina-

tion (Fenster et al., 2004; Flores-Martínez et al.,

2013; Xu & Servedio, 2021), however, it may

also favor self-pollination which produces the

smallest fruits.

Reproductive system: Plants in the planta-

tions were self-compatible, differing from the

findings of Weiss et al. (1994), who reported

S. costaricensis as a self-incompatible species.

Compatibility is variable within the genus,

with self-compatible species such as S. undatus

(Weiss et al., 1994; Pushpakumara et al., 2005;

Valiente-Banuet et al., 2007) and self-incom-

patible species such as S. ocamponis, S. purpusii,

and S. polyrhizus (Weiss et al., 1994; Corona,

2018). A self-compatible plant is favorable from

the farmer’s perspective and can be related to

artificial selection during the domestication of

this species or to variations in the populations

of origin of the plants used in these commer-

cial plantations. However, herkogamy should

reduce natural self-pollination in S. costaricen-

sis, and fruit production is clearly favored by an

animal vector.

The unclear syndrome: Although the

flowers of S. costaricensis present characteristics

consistent with the chiropterophily pollina-

tion syndrome and bats have been reported as

pollinators of S. undatus (Valiente-Banuet et

al., 2007), there were no bat visits at any of

the study sites. The morphology of this flower

does not fully agree with phalaenophily; how-

ever, hawk moths were the most important

nocturnal visitors of S. costaricensis (~40% of

the visits), which agrees with observations by

Haber (1983) in natural populations. Instead

of hovering in front of the flowers, as in many

hawk moth adapted flowers, the insects entered

the flowers, disappearing between the anthers,

which allowed them to come into immedi-

ate contact with the reproductive organs (see

supplementary video). The constant flapping

of its wings during this process warrants pol-

lination. An analysis of floral fragrances of S.

costaricensis is needed to evaluate the presence

of specific compounds related to chiropteroph-

ily or phalaenophily.

Hawk moths consume nectar, but contrary

to what was reported by Weiss et al. (1994),

no evidence of nectar production was found

during this study. The absence of nectar has

also been reported for S. undatus in Mexico

(Valiente-Banuet et al., 2007). The same result

was obtained by Muñiz et al. (2019) for the

nectar production of flowers of S. costaricensis,

S. undatus, and S. polyrhizus. However, the large

visitation of hawk moths indicates that there

are either small amounts of nectar these moths

10 Revista de Biología Tropical, ISSN: 2215-2075 Vol. 73 (S1): e64685, mayo 2025 (Publicado May. 15, 2025)

can collect or pollination by deceit occurs in

commercial plantations. Natural populations

bearing nectar-producing flowers around plan-

tations could allow the maintenance of pol-

lination by a deceit mechanism, maintaining

large hawk moth visitation to plantations where

no nectar is produced. The study of natural

populations of S. costaricensis is necessary to

determine whether the lack of nectar is the

original trait.

Only Manduca rustica was collected in

the plantations, but our sample was small,

and it cannot be ruled out that other species

or even genera of hawk moths also visited S.

costaricensis. Moth species cannot be identi-

fied in video recordings and Haber (1983) also

observed other species of Manduca visiting S.

costarricensis. Landscape management around

pitahaya plantations can favor the populations

of Manduca, e.g. by planting host species.

The role of secondary pollinators: The

lack of nectar creates an unexpected scenario

in which pollen becomes the main reward for

visiting insects in a nocturnal flowering spe-

cies. However, in this case, nocturnal visitors

(excluding the few nocturnal bees) were not

observed collecting pollen. The extension of

flowering for a short period during the morn-

ing could be a safeguard ensuring abundant

pollination in cases where nocturnal visitors are

rare or absent.

Visits by bees were dominant during the

early morning, and bees are important pollina-

tors of S. costaricensis that favor cross-pollina-

tion. The important visitation by opportunistic

diurnal bees, also observed in plantations of

other species in the genus by Weiss et al. (1994)

and Valiente-Banuet et al. (2007), supports the

previously reported discordance between pre-

dicted and observed pollinators in angiosperms

based on traditional pollination syndromes

(Ollerton et al., 2009; Abrahamczyk et al., 2017;

Valverde-Espinoza et al., 2021).

Large carpenter bees (Xylocopa spp.) carry

large amounts of pollen on their bodies and

commonly contact the stigma when moving

between flowers. Weiss et al. (1994) previously

mentioned the potential of large Hymenoptera

as pollinators of the genus Selenicereus. Similar

behaviors were observed during the visits of

nocturnal Megalopta amoena bees.

Finally, Coleoptera (~36% of the visits)

cannot be considered important pollinators of

S. costaricensis, but they can potentially self-

pollinate flowers. For farmers, some beetles are

considered pests since they can damage buds

and open flowers, but none of the filmed beetles

exhibit harmful behavior on any occasion.

The differences between the visitation rates

to the flowers at the three study sites highlight

the importance of local effects on the frequency

and richness of floral visitors. Local effects are

also evident in the composition of the visitors at

each plantation with most of the collected spe-

cies being exclusive at every location.

Mass visits: The number of honeybees

(Apis mellifera) was high in all the samplings.

Muñiz et al. (2020) reported honeybees as the

main pollinators of S. polyrhizus and as a com-

plementary pollinator of S. undatus. Diurnal

bees have also been reported as the most effec-

tive pollinators of S. ocamponis and S. purpusii

(Corona, 2018). The frenetic behavior during

visits and the high number of bees visiting the

flowers in the early morning result in greater

pollen deposition and probably also cross-

pollination, becoming one of the main pollina-

tors of this crop. Generalist diurnal pollinators

can play an important role in the pollination of

cacti with nocturnal anthesis, especially when

nocturnal pollinators are low in abundance

(Weiss et al., 1994; Molina-Freaner et al., 2004;

Le Bellec et al., 2006). In fact, some studies

suggest that pollination of various dragon fruit

species, which is mediated mainly by bees, can

produce good results in terms of the production

of commercial quality fruit (Weiss et al., 1994;

Marques et al., 2011; Muñiz et al., 2020). In this

case, diurnal and nocturnal pollinators have

synergistic effects, explaining the larger dimen-

sions of fruits produced by natural pollination.

The management of Apis mellifera in the

vicinity or in conjunction with dragon fruit

plantations would clearly increase pollination

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 73 (S2): e64685, mayo 2025 (Publicado May. 15, 2025)

in this species due to its high abundance and

behavior in flowers, favoring commercial pro-

duction. Additionally, the maintenance of dead

logs of soft wood near plantations could ensure

the presence of nesting sites for carpenter bees

(Hanson et al., 2021). The traditional manage-

ment of Melipona beecheii for helping pollina-

tion in this cactus requires further study, given

that its presence in the studied plantations was

very low.

However, not all bees have a positive

impact. For example, bees of the genus Trigona

spp. were abundant visitors to the flowers of

the Plancillo plantation, but they are improb-

able pollinators. These bees reached the flowers

prior to anthesis (16:00 h) and pierced the pet-

als to steal pollen from the anthers, damaging

them before the stigmas became receptive.

Effect of pollinators on fruit traits: A

significant variation in fruit size and mass was

found between treatments. Naturally pollinated

flowers produced the largest fruits, supporting

the additive effects of nocturnal and diurnal

pollination. A greater diversity of visitors guar-

antees bigger fruits and greater profit for farm-

ers. Diverse pollinator assemblages improve

fruit quality in some important commercial

crops, such as coffee (Ricketts et al., 2004).

Furthermore, reports indicate that the qual-

ity of some commercial fruits increased when

assisted by native pollinators (Hoehn et al.,

2008; Garibaldi et al., 2013). In agreement with

this rationale, we found that exclusive self-pol-

lination, the outcome expected in a defaunated

environment, produces the smallest fruits. The

effect of spontaneous self-pollination should be

explored to determine management strategies

under these conditions.

Diurnal pollination alone produces large

fruits. Weiss et al. (1994) obtained similar

results with a much larger diurnal visitation

period. Here, we demonstrate that 15–20 min-

utes of massive diurnal visitation is sufficient

for successful fruit production.

This finding confirms that pollinators play

an important role in the production of fruits

of higher commercial value for this species

since the movement of pollen and interbreed-

ing increases crop production. Muñiz et al.

(2020) reported a similar effect when con-

ducting experiments with S. polyrhizus and S.

undatus in plantations in northeastern Brazil.

In fact, pollination mediated by animal vectors

can increase the production of various crops,

either by increasing the quantity or weight of

fruits (Klein et al., 2003; Ricketts et al., 2004;

Rosa et al., 2011; Rizzardo et al., 2012; Milfont

et al., 2013; Taber & Olmstead, 2016).

The reduced size of self-pollinated fruit, as

observed here, is a common result of inbreeding

depression (Charkesworth & Charkesworth,

1999; Waser & Williams, 2001; Chautá-Mellizo

et al., 2012); however, more studies are needed

to confirm this hypothesis. Self-fertilization can

affect the quality of export crops by disfavor-

ing their desirable characteristics, such as size,

shape, or quantity of seeds (Chautá-Mellizo et

al., 2012). Clonal production is the main way

to increase the number of plants in dragon fruit

plantations in Costa Rica, but this practice can

reduce the size of the fruits. The creation of

genetically diverse plantations is recommended

to ensure larger fruits with the greatest com-

mercial value.

Finally, the variation in seed density

between the applied treatments was more irreg-

ular than that of the other metrics. It has been

reported that for S. undatus, the type of pol-

lination does not influence the number of seeds

(Muñiz et al., 2020). The same study indicated

that, in the case of S. polyrhizus, the number of

seeds is greater when the flowers are pollinated

by morning visitors such as honeybees, which

contrasts with our results. Differences in exter-

nal conditions and the abundance of visitors

may explain the differences between the results

of both studies.

Conclusions: In this study, the self-com-

patibility of S. costaricensis was reported for the

first time. Fruit traits are favored by cross-polli-

nation, resulting in fruits of higher commercial

size. An additive effect of nocturnal and diurnal

pollinators produced fruits with the greatest

commercial value.

12 Revista de Biología Tropical, ISSN: 2215-2075 Vol. 73 (S1): e64685, mayo 2025 (Publicado May. 15, 2025)

This study documented the pollinators of

S. costaricensis and their importance in com-

mercial production in the original distribution

of the species. The participation of bats in the

pollination process is ruled out, while hawk

moths and morning swarms of bees (Apis mel-

lifera) appear to be the main pollinators. Other

organisms, such as bees from the genera Xylo-

copa and Megalopta, as well as various beetles,

appear to be secondary or complementary

pollinators. This result demonstrates the con-

tribution of morning pollinators, which could

easily be overlooked since many cacti are tra-

ditionally associated with nocturnal pollination

syndromes, with moths and bats as the main

pollinators. Enrichment of the landscape with

hawk moth host plants near plantations might

benefit their populations and the pollination

services they provide. Management of honey-

bees can complement natural pollinators, and

their use is a promising strategy for increasing

fruit set in defaunated environments. Finally,

increasing the genetic diversity of pitahaya

plants within plantations would favor cross-

pollination and its positive effects on fruit size.

Ethical statement: The authors declare

that they all agree with this publication and

made significant contributions; that there is no

conflict of interest of any kind; and that we fol-

lowed all pertinent ethical and legal procedures

and requirements. All financial sources are fully

and clearly stated in the acknowledgments sec-

tion. A signed document has been filed in the

journal archives.

See supplementary material

a20v73s2-suppl1

ACKNOWLEDGMENT

We thank María del Pilar Rivas, Marvin

Araya (and his family), and Melvin Alpizar for

allowing us to work in their plantations. We

also thank Jorge Lobo and Paul Hanson for

their contributions and suggestions throughout

this project; P. Hanson for identifying most of

the visiting insects; and J. Lobo for the use of

the cameras. Finally, we thank Bronia Alfaro,

José Miguel Valverde, Sua Jiménez, and Beatriz

Antillón for their assistance with fieldwork.

This work was funded by Vicerrectoría de

Investigación, Universidad de Costa Rica, proj-

ect B7287.

REFERENCES

Abrahamczyk, S., Lozada-Gobilard, S., Ackermann, M.,

Fischer, E., Krieger, V., Redling, A., & Weigend, M.

(2017). A question of data quality–testing polli-

nation syndromes in Balsaminaceae. PloS One,

12(10), e0186125. https://doi.org/10.1371/journal.

pone.0186125

Basualdo, M., Cavigliasso, P., De Avila Jr, S. R., Aldea-

Sánchez, P., Correa-Benítez, A., Harms, J.M, Ramos,

A.K., Rojas-Bravo, V., & Salvarrey, S. (2022). Current

status and economic value of insect-pollinated

dependent crops in Latin America. Ecological Eco-

nomics, 196, 107395. https://doi.org/10.1016/j.

ecolecon.2022.107395

Bertin, R. I., & Newman, C. M., (1993). Dichogamy in

angiosperms. The Botanical Review, 59, 112–152.

https://doi.org/10.1007/BF02856676

Borges, R. C., Brito, R. M., Imperatriz-Fonseca, V. L., &

Giannini, T. C. (2020). The value of crop produc-

tion and pollination services in the eastern amazon.

Neotropical Entomology, 49, 545–556. https://doi.

org/10.1007/s13744-020-00791-w.

Centurión-Yah, A., Solís-Pereira, S., Saucedo-Veloz, C.,

Báez-Sañudo, R., & Sauri-Duch, E. (2008). Cambios

físicos, químicos y sensoriales en frutos de pitahaya

(Hylocereus undatus) durante su desarrollo. Revista

Fitotecnia Mexicana, 31(1), 1–5.

Chang, F., Yen, C., Chen, Y., & Chang, L. (1997). Flowering

and fruit growth of pitaya (Hylocereus undatus Britt.

and Rose). Taichung District Agricultural Improve-

ment Station, Annual Report, 38, 293–299.

Charkesworth, B. & Charkesworth, D. (1999). The

genetic basis of inbreeding depression. Genetics

Research, 74(3), 329–340. https://doi.org/10.1017/

S0016672399004152

Chautá-Mellizo, A., Campbell, S. A., Bonilla, M. A., Thaler,

J. S., & Poveda, K. (2012). Effects of natural and arti-

ficial pollination on fruit and offspring quality. Basic

and Applied Ecology, 13(6), 524–532.

Corona, C. (2018). Biología reproductiva de Hylocereus

ocamponis e H. purpusii (Cactaceae) [Tesis para obte-

ner el grado de Maestro en Ciencias en Biosistemática

y Manejo de Recursos Naturales y Agrícolas]. Univer-

sidad de Guadalajara.

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 73 (S2): e64685, mayo 2025 (Publicado May. 15, 2025)

Dobat, K., & Peikert-Holle, T. (1985). Blüten und fleder-

mäuse: bestäubung durch fledermäuse und flughunde

(chiropterophilie). Waldemar Kramer.

Fenster, C., Armbruster, W., Wilson, P., Dudash, M. &

Thomson, J. (2004). Pollination syndromes and floral

specialization. Annual Review of Ecology, Evolution

and Systematics, 35, 375–403. https://doi.org/10.1146/

annurev.ecolsys.34.011802.132347

Flores-Martínez, A., Manzanero, G., Golubov, J. & Mandu-

jano, M. (2013). Biología floral de Mammillaria hui-

tzilopochtli, una especie rara que habita acantilados.

Botanical Sciences, 91(3), 349–356.

Garibaldi, L. A., Steffan-Dewenter, I., Winfree, R., Aizen,

M. A., Bommarco, R., Cunningham, S. A., Kremen,

C., Carvalheiro, L. G., Harder, L. D., Afik, O., Barto-

meus, I., Benjamin, F., Boreux, V., Cariveau, D., Cha-

coff, N. P., Dudenhöffer, J. H., Freitas, B. M., Ghazoul,

J., Greenleaf, S., ... Klein A. M. (2013). Wild pollina-

tors enhance fruit set of crops regardless of honey

bee abundance. Science, 339, 1608–1611. https://doi.

org/10.1126/science.1230200

Giannini, T. C., Cordeiro, G. D., Freitas, B. M, Saraiva, A.

M., & Imperatriz-Fonseca, V. L. (2015). The depen-

dence of crops for pollinators and the economic

value of pollination in Brazil. Journal of Economic

Entomology, 108(3), 849–857. https://doi.org/10.1093/

jee/tov093

Haber, W. A. (1983). Hylocereus costaricensis (pitahaya

silvestre), wild pitahaya. In D.H. Janzen (Ed.), Costa

Rican Natural History (pp. 252–253). University of

Chicago Press.

Hammel, B. E. (2020). Cactaceae. In B. E. Hammel, M. H.

Grayum, C. Herrera & N. Zamora (Eds.), Manual de

Plantas de Costa Rica, Vol. IV, Parte 2: Dicotiledóneas

(Balanophoraceae-Clethraceae) (pp.251–293). Mis-

souri Botanical Garden Press.

Hanson, P., Otárola, M. F., Lobo, J., Frankie, G., Coville, R.,

Aguilar-Monge, I., Acuña Cordero, M., Herrera-Gon-

zález, E. (2021). Abejas de Costa Rica. Editorial UCR.

Hoehn, P., Tscharntke, T., Tylianakis, J., & Steffan-Dewen-

ter, I. (2008). Functional group diversity of bee polli-

nators increases crop yield. Proceedings of the Royal

Society B: Biological Sciences, 275(1648), 2283–2291.

https://doi.org/10.1098/rspb.2008.0405

Holdridge, L. R. (1967). Life Zone Ecology. Tropical Science

Center.

Johnson S. D., & Steiner K. E. (2000). Generalization versus

specialization in plant pollination systems. Trends

in Ecology & Evolution, 15(4), 140–143. https://doi.

org/10.1016/S0169-5347(99)01811-X

Kearns, C. A., Inouye, D. W., & Waser, N. M. (1998).

Endangered mutualisms: the conservation of plant-

pollinator interactions. Annual Review of Ecology and

Systematics, 29(1), 83–112. https://doi.org/10.1146/

annurev.ecolsys.29.1.83

Kevan, P. G., & Phillips, T. P. (2001). The economic impacts

of pollinator declines: an approach to assessing the

consequences. Conservation Ecology, 5(1), 8. http://

www.consecol.org/vol5/iss1/art8/

Klein, A. M., Steffan-Dewenter, I., & Tscharntke, T. (2003).

Bee pollination and fruit set of Coffea arabica and C.

canephora (Rubiaceae). American Journal of Botany,

90(1), 153–157. https://doi.org/10.3732/ajb.90.1.153

Klein, A. M., Vaissiere, B. E., Cane, J.H., Steffan-Dewenter,

I., Cunningham, S. A., Kremen, C., & Tscharntke, T.

(2007). Importance of pollinators in changing lands-

capes for world crops. Proceedings of the Royal Society

B: Biological Sciences, 274(1608), 303–313. https://doi.

org/10.1098/rspb.2006.3721

Lautenbach, S., Seppelt, R., Liebscher, J., & Dormann, C.

F. (2012). Spatial and temporal trends of global polli-

nation benefit. PLoS ONE, 7(4): e35954. https://doi.

org/10.1371/journal.pone.0035954

Le Bellec, F., Vaillant, F., & Imbert, E. (2006). Pitaha-

ya (Hylocereus spp.): a new fruit crop, a market

with a future. Fruits, 61(4), 237–250. https://doi.

org/10.1051/fruits:2006021

Marques, V., Moreira, R., Ramos, J., Araújo, N., & Silva, F.

(2011). Fenologia reprodutiva de pitaia vermelha no

município de Lavras, MG. Ciência Rural, 41, 984–987.

https://doi.org/10.1590/S0103-84782011005000071

Milfont, M., Rocha, E., Lima, A., & Freitas, B. (2013).

Higher soybean production using honeybee and wild

pollinators, a sustainable alternative to pesticides and

autopollination. Environmental Chemistry Letters, 11,

335–341. https://doi.org/10.1007/s10311-013-0412-8

Mizrahi, Y., Nerd, A., & Sitrit, Y. (2002). New fruits for arid

climates. In J. Janick & A. Whipkey (Eds.), Trends in

New Crops and New Uses. (pp. 378–384). ASHS Press.

Molina-Freaner, F., Rojas-Martínez, A., Fleming, T., &

Valiente-Banuet, A. (2004). Pollination biology of

the columnar cactus Pachycereus pecten-aboriginum

in north- western México. Journal of Arid Envi-

ronments, 56(1), 117–127. https://doi.org/10.1016/

S0140-1963(02)00323-3

Muchhala, N., Caiza, A., Vizuete, J. C., & Thomson, J. D.

(2009). A generalized pollination system in the tro-

pics: bats, birds and Aphelandra acanthus. Annals of

Botany, 103(9), 1481–1487. https://doi.org/10.1093/

aob/mcn260

Muñiz, J., Bomfim, I., Corrêa, M., & Freitas, B. (2019).

Floral biology, pollination requirements and beha-

vior of floral visitors in two species of pitaya. Revis-

ta Ciência Agronômica, 50, 640–649. https://doi.

org/10.5935/1806-6690.20190076

14 Revista de Biología Tropical, ISSN: 2215-2075 Vol. 73 (S1): e64685, mayo 2025 (Publicado May. 15, 2025)

Muñiz, J., Bomfim, I., Corrêa, M., & Freitas, B. (2020).

Complementary bee pollination maximizes yield and

fruit quality in two species of self-pollinating pita-

ya. Revista Ciência Agronômica, 51(4), e20207106.

https://doi.org/10.5935/1806-6690.20200072

Nabhan, G. P., & Buchmann, S. L. (1997). Services provided

by pollinators. In G. C. Daily (Ed.), Nature’s Servi-

ces: Societal Dependence on Natural Ecosystems. (pp.

133–150). Island Press.

Ollerton, J., Alarcón, R., Waser, N. M., Price, M. V, Watts, S.,

Cranmer, L., Hingston, A., Peter, C. I., & Rotenberry,

J. (2009). A global test of the pollination syndrome

hypothesis. Annals of Botany, 103, 1471–1480. https://

doi.org/10.1093/aob/mcp031

Ollerton, J., Winfree, R., & Tarrant, S. (2011). How

many flowering plants are pollinated by ani-

mals?. Oikos, 120(3), 321–326. https://doi.

org/10.1111/j.1600-0706.2010.18644.x

Porto, R. G., Almeida, R. F., Cruz-Neto, O., Tabarelli, M.,

Viana, B. F., Peres, C. A., & Lopes, A. V. (2020). Polli-

nation ecosystem services: a comprehensive review of

economic values, research funding and policy actions.

Food Security, 12, 1425–1442. https://doi.org/10.1007/

s12571-020-01043-w

Pushpakumara, D., Gunasena, H., & Karyawasam, M.

(2005). Flowering and fruiting phenology, pollination

vectors and breeding system of dragon fruit (Hyloce-

reus spp.). Sri Lankan Journal of Agricultural Science,

42, 81–91.

Queiroz, J. A., Quirino, Z. G. M., Lopes, A. V., & Machado,

I. C. (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. https://doi.

org/10.1016/j.jaridenv.2015.09.015

Queiroz, J. A., Quirino, Z. G. M, & Machado, I. C. (2015).

Floral traits driving reproductive isolation of two

co-flowering taxa that share vertebrate pollinators.

AoB Plants, 7, plv127. https://doi.org/10.1093/aobpla/

plv127

R Core Team. (2016). R: a language and environment for

statistical computing. R Foundation for Statistical

Computing.

Ricketts, T. H., Daily, G. C., Ehrlich, P. R., & Michener,

C. D. (2004). Economic value of tropical forest to

coffee production. Proceedings of the National Aca-

demy of Sciences, 101(34), 12579–12582. https://doi.

org/10.1073/pnas.0405147101

Rizzardo, R., Milfont, M., Silva, E., & Freitas, B. (2012).

Apis mellifera pollination improves agronomic

productivity of anemophilous castor bean (Rici-

nus communis). Anais da Academia Brasileira de

Ciências, 84(4), 1137–1145. https://doi.org/10.1590/

S0001-37652012005000057

Rosa, A., Blochtein, B., & Lima, D. (2011). Honey bee

contribution to canola pollination in Southern

Brazil. Scientia Agricola, 68, 255–259. https://doi.

org/10.1590/S0103-90162011000200018

Rosas-Guerrero, V., Aguilar, R., Martén-Rodríguez, S.,

Ashworth, L., Lopezaraiza-Mikel, M., Bastida, J. M.,

& Quesada, M. (2014). A quantitative review of polli-

nation syndromes: do floral traits predict effective

pollinators? Ecology Letters, 17, 388–400. https://doi.

org/10.1111/ele.12224

Taber, S. K., & Olmstead, J. W. (2016). Impact of cross-and

self-pollination on fruit set, fruit size, seed number,

and harvest timing among 13 southern highbush

blueberry cultivars. HortTechnology, 26(2), 213–219.

https://doi.org/10.21273/HORTTECH.26.2.213

Valiente-Banuet, A., Gally, R., Arizmendi, M., & Casas, A.

(2007). Pollination biology of the hemiepiphytic cac-

tus Hylocereus undatus in the Tehuacán Valley, Mexi-

co. Journal of Arid Environments, 68(1), 1–8. https://

doi.org/10.1016/j.jaridenv.2006.04.001

Valverde-Espinoza, J. M., Chacón-Madrigal, E., Alvarado-

Rodríguez, O., & Dellinger, A. S. (2021). The pre-

dictive power of pollination syndromes: passerine

pollination in heterantherous Meriania macrophylla

(Benth.) Triana (Melastomataceae). Ecology and Evo-

lution, 11(20), 13668–13677. https://doi.org/10.1002/

ece3.8140

Villalobos, R., & Retana, J. (2000). Costa Rica: Instituto

Meteorologico Nacional. In J.W. Jones (Coordina-

tor), Comparative assessment of agricultural uses of

ENSO-based climate forecasts in Argentina, Costa Rica

and Mexico, final report (pp. 49–95). Inter-American

Institute for Global Change Research.

Vogel, S. (1968). Chiropterophilie in der neotropischen

flora: neue mitteilungen I. Flora oder Allgemeine

botanische Zeitung. Abt. B, Morphologie und Geobo-

tanik 157(4), 562–602. https://doi.org/10.1016/

S0367-1801(17)30097-2

Waser, N. M., Chittka, L, Price, M. V., Williams, N. M., &

Ollerton, J. (1996). Generalization in pollination sys-

tems, and why it matters. Ecology, 77(4), 1043–1060.

https://doi.org/10.2307/2265575

Waser, N. M., & Williams, C. F. (2001). Inbreeding and

Outbreeding, In C.W. Fox, D.A. Roff, & D.J. Fair-

bairn (Eds.), Evolutionary Ecology: Concepts and Case

Studies (pp. 84–96). Oxford Academic. https://doi.

org/10.1093/oso/9780195131543.003.0011

Weiss, J., Nerd, A., & Mizrahi, Y. (1994). Flowering behavior

and pollination requirements in climbing cacti with

fruit crop potential. HortScience, 29(12), 1487–1492.

https://doi.org/10.21273/HORTSCI.29.12.1487

Xu, K., & Servedio, M. (2021). The evolution of flower

longevity in unpredictable pollination environments.

Journal of Evolutionary Biology, 34(11), 1781–1792.

https://doi.org/10.1111/jeb.13936