Revista de Biología Tropical, ISSN: 2215-2075, Vol. 73: e2025456, enero-diciembre 2025 (Publicado Set. 17, 2025)

Light preferences, population structure, and pre-dispersal fruit predation

in the reverse-phenology tree Bonellia nervosa (Primulaceae)

Gerardo Avalos1,2*; https://orcid.org/0000-0003-2663-4565

Esteban Marín-Castillo1; https://orcid.org/0009-0005-7129-694X

Valeria Acevedo-Fernández1; https://orcid.org/0009-0006-4620-8597

Esteban Zamora-Villalobos1; https://orcid.org/0009-0002-1980-3738

Tadeo Aguilar-Bermúdez1; https://orcid.org/0009-0009-8055-687X

1. Escuela de Biología, Universidad de Costa Rica, San José, Costa Rica; gerardo.avalos@ucr.ac.cr (*Correspondence), este-

ban.marincastillo@ucr.ac.cr, raquel.acevedo@ucr.ac.cr, esteban.zamoravillalobos@ucr.ac.cr, tadeo.aguilar@ucr.ac.cr

2. The School for Field Studies, Center for Ecological Resilience Studies, P.O. Box 506 West Boxford, MA 01885, United

States of America; gavalos@fieldstudies.org

Received 11-XI-2024. Corrected 30-IV-2025. Accepted 02-IX-2025.

ABSTRACT

Introduction: Bonellia nervosa is a dry forest understory tree with reverse leafing phenology, as it produces

leaves during the dry season and is deciduous during the wet season. Being a phreatophytic species, it relies on

substantial root biomass to access groundwater and flush leaves during the dry season.

Objective: To assess the population structure of B. nervosa in Santa Rosa National Park, Costa Rica; analyze the

relationship between stem diameter and plant height; and examine how canopy structure and light availability

influence its local abundance. Also, to evaluate fruit maturity and pre-dispersal fruit predation during the late

wet season to understand its reproductive success.

Methods: The study was conducted along the Indio Desnudo trail (6.16 ha). Hemispherical photographs were

taken at each B. nervosa (n = 33) and at 10 random sites where the species was absent. Principal component

analysis was used to consolidate seven light and canopy structural variables. The Chapman-Richards model was

applied to examine the relationship between stem diameter and height. Fruit weight and seed count were ana-

lyzed and each fruit inspected for the presence of Tortricidae larvae.

Results: Adults dominated the population. An inflection point was identified at 70.6 cm in height, indicating

a shift from relatively rapid to progressively slower growth. Bonellia nervosa sites had higher canopy openness

(14.68 %) and transmitted more diffuse light (21.05 μmol/m²/s) than sites where the species was absent (12.42

% and 18.22 μmol/m²/s, respectively). Pre-dispersal fruit predation was low (16 % of fruits, n = 39) and by

Tortricidae larvae.

Conclusions: Seedling scarcity creates a recruitment bottleneck. The dominance of adults, the low number of

seedlings, and specific light preferences limit the plasticity and adaptive capacity of B. nervosa to environmental

changes. Future research should focus on the physiological mechanisms underlying B. nervosa’s reverse leaf phe-

nology and its resilience to climate change.

Keywords: phenological strategies; reverse phenology; pre-dispersal seed predation; tropical dry forests.

https://doi.org/10.15517/pmde0d11

TERRESTRIAL ECOLOGY

2Revista de Biología Tropical, ISSN: 2215-2075 Vol. 73: e2025456, enero-diciembre 2025 (Publicado Set. 17, 2025)

INTRODUCTION

Tropical dry forests rank among the most

endangered terrestrial ecosystems (Buchadas

et al., 2022; Miles et al., 2006; Murphy & Lugo,

1986; Sánchez-Azofeifa et al., 2013; Siyum,

2020), having been some of the first to be

extensively altered by recent human activity

(Buchadas et al., 2022; Janzen et al., 2016). As

a result, old-growth, mature tropical dry forests

are now rare, and most remaining areas are

disturbed or undergoing regeneration (Duan et

al., 2023; Griscom & Ashton, 2011; Quesada et

al., 2009). These forests endure a pronounced

seasonal drought (4-6 months) characterized by

high temperatures and significant water deficits

(Murphy & Lugo, 1995) and have evolved a dis-

tinctive phenology adapted to episodic growth

(Frankie et al., 1974; Janzen, 1967). During the

wet season, plants flush leaves and accumulate

reserves for reproduction during the following

dry period. As drought intensifies, most spe-

cies shed their leaves, reproduce, and release

seeds near the end of the dry period. With the

onset of the rains, seedlings establish and store

resources to withstand the prolonged drought

of the dry season (Janzen, 1967). While these

forests support a range of phenological strate-

gies to cope with extreme seasonality (Giraldo

& Holbrook, 2011; Lasky et al., 2016; Lópezara-

iza-Mikel et al., 2013), one particularly distinc-

tive adaptation is reverse phenology (Giraldo &

Holbrook, 2011; Janzen, 1970).

Bonellia nervosa is a dry forest under-

story tree exhibiting reverse leaf phenology.

Unlike most plants in this ecosystem, which

are deciduous during the prolonged dry season

and retain leaves in the wetter months (Frankie

et al., 1974; Murphy & Lugo, 1986; Murphy

& Lugo, 1995), B. nervosa produces a new

crop of leaves only during the dry period and

sheds them at the onset of the rains (Chaves &

RESUMEN

Preferencias de luz, estructura poblacional y depredación predispersión

de frutos en el árbol de fenología inversa Bonellia nervosa (Primulaceae)

Introducción: Bonellia nervosa es un árbol del sotobosque del bosque seco tropical con fenología foliar invertida

al producir hojas durante la estación seca y ser caducifolio durante la estación lluviosa. Esta especie freatofítica

depende de una biomasa radicular considerable para acceder al agua subterránea y desarrollar hojas durante la

estación seca.

Objetivo: Evaluar la estructura poblacional de B. nervosa en el Parque Nacional Santa Rosa, Costa Rica; analizar

la relación entre el diámetro del tallo y la altura de la planta; y examinar cómo la estructura del dosel y la dis-

ponibilidad de luz influyen en su abundancia local. También evaluar la madurez de los frutos y la depredación

predispersión al final de la estación lluviosa para entender su éxito reproductivo.

Métodos: El estudio se realizó a lo largo del sendero Indio Desnudo (6.16 ha). Se tomaron fotografías hemisféricas

de cada individuo de B. nervosa (n = 33) y en 10 sitios aleatorios donde la especie estaba ausente. Se utilizó análisis

de componentes principales para consolidar variables relacionadas con la luz y la estructura del dosel. Se aplicó

el modelo de Chapman-Richards para examinar la relación entre el diámetro del tallo y la altura. Se analizaron

el peso de los frutos y el número de semillas, y se inspeccionó cada fruto para detectar la presencia de larvas de

Tortricidae.

Resultados: La población estuvo dominada por adultos. Se identificó una transición de un crecimiento relati-

vamente rápido a uno progresivamente más lento a los 70.6 cm de altura. Los sitios con B. nervosa presentaron

mayor apertura del dosel (14.68 %) y mayor transmisión de luz difusa (21.05 μmol/m²/s) que los sitios donde la

especie estaba ausente (12.42 % y 18.22 μmol/m²/s, respectivamente). La depredación predispersión fue baja (16

% de los frutos, n = 39) y causada por larvas de Tortricidae.

Conclusiones: La escasez de plántulas genera un cuello de botella en el reclutamiento. El dominio de los adultos,

la escasez de plántulas y las preferencias específicas de luz restringen la plasticidad y capacidad de adaptación de

B. nervosa a cambios ambientales. Las investigaciones futuras deben centrarse en los mecanismos fisiológicos que

sustentan la fenología foliar invertida de B. nervosa y su resiliencia ante el cambio climático.

Palabras clave: estrategias fenológicas; fenología inversa; depredación predispersión; bosque tropical seco.

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 73: e2025456, enero-diciembre 2025 (Publicado Set. 17, 2025)

Avalos, 2008; Janzen, 1970). This unusual phe-

nological pattern requires a highly specialized

resource-use strategy (Janzen & Wilson, 1974).

In addition to concentrating leaf production at

the beginning of the drought period, B. nervosa

also flowers and initiates fruit development;

fruit maturation and seed dispersal occur in

the late rainy season (Chaves & Avalos, 2008).

The plant must also accumulate reserves for

fruit expansion and survival under the shaded

conditions of the forest understory during the

wetter months (Janzen & Wilson, 1974). This

strategy requires B. nervosa to allocate a sub-

stantial portion of its biomass to root develop-

ment (Chaves, 2002; Janzen, 1970), allowing it

to access the water table and clearly defining

it as a phreatophytic species. Similar leaf phe-

nology and biomass allocation strategies have

been observed in other phreatophytes, such as

Faidherbia albida in the arid forests of the Sahel

region in West Africa (Roupsard et al., 1999). A

reverse leafing phenology allows B. nervosa to

fix carbon during the dry season, maintaining

a competitive advantage relative to deciduous

plants. This strategy results in overall slow

growth (and growth in pulses, Schwinning &

Kelly, 2013), particularly during the seedling

stage, as the plant must first develop a suffi-

ciently large root system to tap the water table.

Bonellia nervosa establishes itself primar-

ily in shaded microhabitats, where light acts as

an environmental filter and plant abundance is

modulated by seed dispersal and seed predation.

In well-lit environments, the plant would incur

higher maintenance costs and face competition

from faster-growing species. Consequently, B.

nervosa thrives as a deciduous understory tree

under a closed canopy during the rainy season,

developing a full crown of leaves in the dry sea-

son when the canopy opens (Chaves & Avalos,

2008). Consequently, it is less common in

habitats with high light exposure—such as open

savannas, pastures, forest edges, or very young

secondary forests—or in areas with persistent,

dense shade throughout the rainy season, espe-

cially in forests dominated by evergreen species

with year-round canopies, such as old-growth,

mature dry forests. This pattern suggests that B.

nervosa performs best in intermediate shade or

moderately shaded environments, such as old

secondary forests (80-100 years), and is less fre-

quent in habitats with intense sunlight (Janzen

& Wilson, 1974) or dense shade, like riparian

or mature, late-successional forests. These latter

habitats have become rare in the tropical dry

forest ecosystem due to widespread deforesta-

tion and cattle ranching (Stan et al., 2024).

The pattern of habitat selection of B. ner-

vosa is congruent with our field observations

and what is reported in the literature (i.e.,

Enquist & Sullivan, 2001; Janzen, 1970; Janzen

& Wilson, 1974). Establishing under inter-

mediate shade has important implications for

its life history strategy and could influence its

population structure and reproductive strate-

gies (Avalos, 2019). Bonellia nervosa is found

in secondary forests and dry areas on hilltops

or cliffs providing competitive advantages at

the soil level (Enquist & Sullivan, 2001). At

Palo Verde National Park, we have observed it

along the Guayacán trail, which is dominated

by limestone outcrops. Under these conditions,

B. nervosa does not establish itself in purely

rocky areas but prefers flatter or slightly sloped

terrain that can accumulate enough soil for the

plant’s roots to reach the water table. The phe-

nological pattern of B. nervosa is unique, as no

other species in Costa Rica’s dry forests exhibits

reverse phenology. This strategy comes with

significant trade-offs, as B. nervosa is typically

scarce and confined to habitats with intermedi-

ate shade such as old secondary forests. Reverse

phenology likely puts this species at risk under

climate change scenarios, where increased cli-

matic variability could threaten the survival of

early life stages during heat waves or fluctuating

conditions of radiation and water availability.

The main objective of this study was to

examine the population structure of B. nervosa

in Santa Rosa National Park (SRNP). Specifi-

cally, we aimed to: (i) assess the abundance of

adults, juveniles, and seedlings in an old sec-

ondary forest, (ii) evaluate the influence of

canopy structure and light conditions on the

occurrence of B. nervosa during the peak of the

wet season when canopy cover is highest, and

4Revista de Biología Tropical, ISSN: 2215-2075 Vol. 73: e2025456, enero-diciembre 2025 (Publicado Set. 17, 2025)

(iii) analyze fruit maturity and the intensity of

pre-dispersal fruit predation five months into

the wet season. Given the slow and meta-

bolically demanding process of accumulating

root biomass and food reserves during early

life stages, we expected the population to be

dominated by adults, with relatively few seed-

lings. We hypothesized that population struc-

ture would be influenced by light availability,

as well as the activities of seed dispersers and

seed and seedling predators. Furthermore, we

anticipated that B. nervosa would establish most

successfully in moderately shaded environ-

ments, avoiding areas with either low canopy

cover (high light) or dense shade.

Analyzing the population structure, habi-

tat selection and predispersal fruit predation

of this species is crucial to understanding the

adaptive consequences of reverse phenology

(Stan & Sanchez-Azofeifa, 2019). The rarity

of the reverse leaf strategy may explain why it

has been largely excluded from phenological

analyses (Giraldo & Holbrook, 2011; Schwin-

ning & Kelly, 2013). This stresses the need to

develop effective conservation protocols that

account for the phenological diversity of tropi-

cal dry forests to improve our understanding of

their resilience in the face of climate change and

habitat degradation (Miles et al., 2006; Siyum,

2020; Stan & Sanchez-Azofeifa, 2019).

MATERIALS AND METHODS

Site description: This study was conduct-

ed in SRNP, Guanacaste, Costa Rica (10°50’04”

N & 85°36’45” W, 295 m.a.s.l) in September

2024, along the entire length of the Indio Des-

nudo trail. The forest in this area is classified

as a tropical premontane forest (Holdridge &

Grenke, 1971), with a canopy height of 25 m.

It is dominated by representative dry tree forest

species, including Bursera simaruba, Enterolo-

bium cyclocarpum, Manilkara chicle, and Caly-

cophyllum candidissimum. The average annual

rainfall is 1 423 mm, with September and Octo-

ber being the wettest months. The mean annual

temperature is 25.7 °C, and the relative humid-

ity averages 81 % (SRNP Climatic Records).

Established in 1974 to protect significant his-

torical sites, SRNP also initiated an ongoing

ecological restoration process (Griscom & Ash-

ton, 2011). SRNP maintains transitional areas

between tropical moist and tropical dry forests

(Kalacska et al., 2004). It is estimated that the

Indio Desnudo trail has undergone approxi-

mately 80 years of regeneration and is classified

as a late secondary forest, since it contains three

layers of vegetation with 50-90 % of the canopy

being occupied by evergreen species. This fol-

lows the terminology of Cao et al. (2015),

Castillo et al. (2011) and Kalacska et al. (2004).

Study species: Bonellia nervosa (C. Presl)

B. Ståhl & Källersjö (Primulaceae, formerly

classified as Jacquinia pungens, and then J. ner-

vosa in the Theophrastaceae, Morales, 2003)

has been described as an understory shrub

abundant in deciduous and semi-deciduous

forests throughout the Pacific coast of Meso-

america, from Southern Jalisco, Mexico, to

Northwestern Costa Rica (Ståhl, 1989; Ståhl &

Källersjö, 2004). However, we have observed

individuals reaching 4.86 m in height (this

study) and up to 7 m (pers. obs.), indicating

that B. nervosa should be considered a small

understory tree. Mature individuals can attain

not only these heights but also substantial stem

diameters (20-30 cm). This species is associ-

ated with tropical dry forests, although Morales

(2003) indicates it could reach moist forests.

The overall structure of B. nervosa is xerophytic

(Janzen, 1983) and drought-tolerant (Romero et

al., 2023). Leaves are coriaceous, with the apex

modified as a spine, and once with a complete

crown, the plant seems like a cactus with very

sharp spines pointing out (Janzen, 1970). The

leaves are produced synchronously at the start

of the dry season (Chaves & Avalos, 2008) but

begin to senesce and drop two weeks after the

beginning of the rainy season. Senescing leaves

often can remain in the plant for more than five

months into the rainy season (while finishing

reallocation of resources). The species remains

leafless–or if leaves remain, they are not func-

tional–throughout the rainy season. The spe-

cies is phreatophytic with a deep root system

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 73: e2025456, enero-diciembre 2025 (Publicado Set. 17, 2025)

(2 m tall individuals have roots reaching 3 m in

depth; Janzen, 1970; Janzen, 1983), which taps

the subsoil water allowing it to produce leaves

during the dry season while most of the plant

community is deciduous. Seventy-eight percent

of the biomass of very young seedlings (8.5 cm

tall, n = 15) goes into roots (Chaves, 2002).

There is limited data on water relations. For

instance, the range in predawn water potential

spanned -2.62 MPa during mid-March, one

of the widest ranges compared to the species

examined by Oberbauer (1985) in Palo Verde.

This indicates water stress during the peak of

the dry season. The fruits are round berries

(mean width = 2.6 ± 0.3 cm, mean length = 2.7

± 0.2 cm, mean weight = 8.48 ± 2.76 g, n = 39),

indehiscent and coriaceous, with the number

of seeds proportional to fruit size, ranging from

four to 24 (average 13 (5), n = 39).

Determination of population structure:

We conducted thorough visual surveys for

seedlings, juveniles, and adults of B. nervosa

along the road from the Park Administration to

the entrance of the Indio Desnudo trail near the

Santa Rosa historical site, extending 30 m into

the forest on both sides throughout the entire

trail (our sample area was approximately 6.16

ha). We meticulously searched for seedlings

(height < 0.6 m), juveniles (height > 0.6 and <

1.5 m), and adults (height > 1.5 m). For each

plant, we measured height and stem diam-

eter 10 cm above the ground and analyzed the

allometry of the height and stem diameter rela-

tionship using the Chapman-Richards model

(Domínguez-Calleros et al., 2017).

Relationship between canopy structure

and the presence of B. nervosa: We took hemi-

spherical photographs 1.5 m above the ground,

directly above or next to individual plants, and

at 10 random points along the Indio Desnudo

trail where B. nervosa was absent. We used a

Nikon Coolpix 5 000 camera with a fisheye

lens, mounted on a tripod with the lens ori-

ented vertically toward the canopy. The top of

the image was aligned with the magnetic North

as determined using a compass. Photographs

were taken under overcast skies, near dusk,

or after dawn, to minimize glare and prevent

direct sunlight from obscuring parts of the

image. We analyzed the photographs using Gap

Light Analyzer (GLA, version 2.0; Frazer et al.,

1999) and measured the percentage of visible

sky (% sky area), canopy openness (% canopy

openness), leaf area index (LAI, measured at

zenith angles of 0°-LAI 4- to 60° and 0° to 75°

-LAI 5-), and the percentage of direct, diffuse,

and total light transmitted through the canopy.

Intensity of predispersal fruit predation:

We inspected all B. nervosa plants for the pres-

ence of fruits and randomly selected 39 fruits.

Using Scout Pro Ohaus Portable Electronic Bal-

ance, we weighed the fresh fruits and seeds and

calculated the percentage of seed weight relative

to the total fruit weight. We also recorded the

number of fruits eaten by Tortricidae larvae

and calculated the percentage of fruit damage.

Additionally, we observed varying degrees of

seed maturation; some fruits contained visibly

unripe beige seeds, while others housed mature

brown seeds.

Statistical analyses: We applied the

Chapman-Richards nonlinear growth function

(Domínguez-Calleros et al., 2017; Zhao-Gang

& Feng-Ri, 2003) to model the relationship

between height and diameter. This flexible sig-

moidal model captures the trajectory of plant

growth, characterized by an initial phase of

rapid height increase followed by a gradual

deceleration as the plant approaches its maxi-

mum size. The Chapman-Richards model is

expressed as H = a (1-b × exp (-k × D))

m, where H is height, D is diameter, and a,

b, k, and m are parameters controlling the

asymptote, growth rate, and curve shape. This

approach allows for accurate estimation of

inflection points, reflecting critical shifts in

growth dynamics throughout plant´s ontogeny.

The height-diameter allometric relationships

provide insights into resource allocation strate-

gies, responses to environmental changes, and

transitions between life stages (i.e., inflection

6Revista de Biología Tropical, ISSN: 2215-2075 Vol. 73: e2025456, enero-diciembre 2025 (Publicado Set. 17, 2025)

points), which often indicate shifts from vegeta-

tive to reproductive growth.

In addition, to explore the relationships

among the seven light and canopy structure

variables mentioned earlier, we conducted a

principal component analysis (PCA). We select-

ed the first two components, which accounted

for 82.27 % of the variation, as response vari-

ables summarizing canopy structure. Subse-

quently, we performed a one-way ANOVA to

compare these components between locations

where B. nervosa was present and randomly

selected points where it was absent. Addition-

ally, we assessed the relationship between fruit

weight and seed count using an ordinary least

squares regression. We checked the adjustment

to the assumptions of normality and equality of

variances of parametric analyses. All statistical

analyses were conducted using R software (R

Core Team, 2024).

RESULTS

Population structure: We found 33 indi-

viduals ranging in height from 15 to 486 cm.

Three individuals were under 50 cm, main-

tained functional leaves, and were classified as

seedlings. The rest were classified as reproduc-

tive individuals since we found mature or rotten

fruits still attached to the plant. These results

support our expectation that seedlings will be

rare, and that the population will be dominated

by reproductive individuals (Fig. 1A). Stem

diameter and height were significantly related

(Fig. 1B). The inflection point was located at

1.1 cm in diameter and 70 cm in height (at this

size, the model estimates the switch from fast

to slower growth). The asymptotic maximum

value of height was 437.74 cm.

Canopy structure: Sites with B. ner-

vosa had significantly higher canopy open-

ness (14.68 %) than random sites (12.42 %;

F₁,₄₁ = 7.49, p = 0.009; Table 1, Fig. 2A),

as well as greater transmitted diffuse light

(21.05 vs. 18.22; F₁,₄₁ = 9.11, p = 0.004; Fig.

2B). However, these differences were mod-

est in magnitude. Leaf Area Index (LAI4 and

LAI5) was similar between groups, showing

no significant differences.

The percentage of transmitted direct light

and total transmitted light showed the strongest

correlation (0.96), followed by transmitted dif-

fuse light, canopy openness, and leaf area index

(LAI) at zenith angles of 0°-60° (0.89, LAI 4,

Table 2). Two principal components explained

82.27 % of the variation in canopy structure

(Table 3). The first component (PC1, 50.61 % of

the variation) was dominated by canopy open-

ness and transmitted diffuse light. The second

component (PC2, 31.66 %) was dominated by

transmitted direct light.

Fig. 1. A. Height distribution in the population of Bonellia nervosa. B. Chapman-Richards nonlinear growth model applied

to the relationship between height and diameter (R2 = 0.78, F3,29 = 34.62, p < 0.001) in 33 individuals of B. nervosa, SRNP,

Costa Rica.

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 73: e2025456, enero-diciembre 2025 (Publicado Set. 17, 2025)

One-way ANOVA was used to test the

hypothesis that B. nervosa prefers intermediate

shade conditions and compared the sites where

it was present relative to random sites where

it was absent using these two components

as response variables. PC1 scores (percentage

canopy openness and percentage trans diffuse)

were higher at B. nervosa sites (average = 0.42

± 1.81) compared to random sites (average =

-1.41 ± 1.38; F₁,₄₁ = 8.68, p = 0.005). There were

no significant differences in PC2 scores (per-

centage of trans direct) between B. nervosa sites

(average = 0.04 ± 1.61) and random sites (aver-

age = -0.12 ± 1.04; F₁,₄₁ = 0.09, p = 0.76; Fig. 2).

Fig. 2. Differences in light conditions between sites occupied by Bonellia nervosa and random sites in SRNP, Costa Rica. A.

Percentage of canopy openness. B. Percentage of transmitted diffuse light.

Table 1

Summary of canopy structure variables at the Indio Desnudo trail, SRNP, during the late wet season of 2024.

Site NPercentage

sky area

Percentage canopy

openness LAI 4 LAI 5 Percentage t

rans direct

Percentage

trans diffuse

B. nervosa 33 99.88 (0.01) 14.68 (2.42) 2.11 (0.36) 2.10 (0.33) 25.84 (6.19) 21.05 (2.76)

Random 10 99.88 (0.01) 12.42 (1.82) 2.36 (0.42) 2.30 (0.36) 22.76 (2.88) 18.22 (1.92)

Values are means (± 1 SD) of 43 hemispherical photos taken at random sites and at sites occupied by B. nervosa.

Table 2

Correlation matrix of 7 canopy structure variables in the SRNP Forest in 2024 (n = 43 hemispherical photos).

Percentage

sky area

Percentage

canopy openness LAI 4 LAI 5 Percentage

trans direct

Percentage

trans diffuse

Percentage sky area 1

Percentage canopy openness 0.17 1

LAI 4 -0.15 -0.76 1

LAI 5 -0.34 -0.71 0.89 1

Percentage trans direct -0.14 0.30 0.12 0.18 1

Percentage trans diffuse 0.10 0.89 -0.50 -0.40 0.59 1

Percentage trans total -0.06 0.54 -0.1 -0.009 0.96 0.80

Values represent Pearson correlation coefficients. Boldfaced values show significant correlations at p < 0.001. LAI = leaf

area index at a zenith angle of 0°-60° -LAI 4-, and at 0°-75° -LAI 5-, Trans direct = transmitted direct light, Trans diffuse =

transmitted diffuse light, Trans total = transmitted total light.

8Revista de Biología Tropical, ISSN: 2215-2075 Vol. 73: e2025456, enero-diciembre 2025 (Publicado Set. 17, 2025)

Predispersal fruit predation: There was a

positive relationship between fruit weight, seed

weight, and seed number (Fig. 3), indicating that

larger fruits contain more and heavier seeds. Of

the 39 fruits examined, six were infested by

moth larvae from the Family Tortricidae (the

only seed predator found), with one fruit hav-

ing all its seeds consumed. The percentage of

fruit predation was 15.38 %, although this likely

fluctuated throughout fruit development since

the last crop. Our observations were limited to

fruits that remained until September, but many

more were likely damaged earlier in the season.

DISCUSSION

The population structure, dominated by

adults, suggests that seedlings are rare, likely

due to multiple stressors. For instance, the high

energy demand required to store reserves for

surviving in the shade during the rainy season

may increase seedling mortality. Additionally,

seedlings are more susceptible to drought-

induced mortality, including cavitation and

hydraulic failure (Werden et al., 2023). Their

small size further increases vulnerability to

herbivory and pathogen damage, as they may

be unable to regenerate tissue or compen-

sate for losses effectively. In addition, seed-

lings are especially vulnerable to mechanical

damage from falling branches. Competition

with vines and other understory vegetation

may further limit their survival. Our previous

observations indicate that seedlings tend to

aggregate around adult plants, consistent with

the dispersion pattern typical of many tropi-

cal dry forest species (Hubbell, 1979), which

may increase their exposure to herbivores and

pathogens (Bhadouria et al., 2016). These com-

bined pressures suggest that the seedling stage

acts as a critical ecological filter influencing the

habitat preferences of B. nervosa. Long-term,

Fig. 3. A. Relationship between fruit weight and seed weight. B. Relationship between fruit weight and number of seeds in

B. nervosa, SRNP, Costa Rica.

Table 3

Coefficients of the eigenvalues and eigenvectors of the

PCA from seven variables obtained from the hemispherical

photograph analysis in the Indio Desnudo trail, SRNP.

Eigenvalue Variation

explained (%)

PC1 3.54 50.61

PC2 2.21 31.66

Cumulative variation 82.27

PC1 PC2

Percentage sky area 0.109 -0.27

Percentage canopy openness 0.50 -0.11

LAI 4 -0.38 0.40

LAI 5 -0.35 0.46

Percentage trans direct 0.27 0.54

Percentage trans diffuse 0.50 0.14

Percentage trans total 0.38 0.46

Boldfaced values represent dominant loadings of the

eigenvectors on each PC.

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 73: e2025456, enero-diciembre 2025 (Publicado Set. 17, 2025)

multi-season monitoring studies are necessary

to accurately estimate seedling abundance and

recruitment dynamics. A long-term popula-

tion growth study, projected over several years,

may be ambitious considering that B. nervosa

appears to grow slowly. The Chapman-Rich-

ards model indicates an inflection point at

approximately 70.6 cm in height, suggesting a

transition from slow to even slower growth. We

observed reproductive structures in an indi-

vidual of 1.5 m in height, indicating that B. ner-

vosa may begin reproducing at relatively small

sizes and possibly early in its life cycle, which

could shift resource allocation toward repro-

duction rather than growth and maintenance.

Since leaves are produced only during the dry

season, there may be minimal differences in

leaf quality and function between seedlings

and adults, although this hypothesis requires

further investigation.

Bonellia nervosa is not an abundant species

in the dry forests of SRNP. While it is capable of

colonizing open areas such as forest edges (Jan-

zen, 1970) and has occasionally been observed

in pastures, it is more frequently found in old

secondary forests within SRNP. Notably, it is

absent from the park’s only patch of primary

forest, which remains evergreen during the dry

season. This distribution suggests that B. ner-

vosa prefers intermediate light conditions and

avoids both the most open and the most shaded

environments. This pattern was also supported

by observations from A. Sánchez-Azofeifa, who

did not find B. nervosa in forests younger than

50 years and found only three individuals in an

80-year-old plot, further indicating its affin-

ity for late-successional stages (A. Sánchez-

Azofeifa, personal communication, 2024). Our

findings (five individuals/ha) are consistent

with those of Glander and Nisbett (1996), who

reported a very low density in Cañas, Costa

Rica (0.06 individuals/ha). These observations

suggest that B. nervosa may function as a late-

successional species and a potential indicator of

forest regeneration, especially in dry forest eco-

systems where successional dynamics are slow

and strongly shaped by water availability and

disturbance regimes (Griscom & Ashton, 2011;

Quesada et al., 2014). Its presence could there-

fore reflect structural and microclimatic condi-

tions associated with more advanced stages of

secondary succession.

A related species, B. macrocarpa, in Oaxa-

ca, Mexico, was classified as a late-successional

species by Romero et al. (2023). It typically

occurs in tropical dry forests aged 50 to 70 years

but has also been observed in early succes-

sional forests as young as two years old, but

in low numbers. Bonellia nervosa employs a

“long-lived pioneer” regeneration strategy,

enabling it to associate with old secondary for-

ests but limiting its abundance in early succes-

sional stages and primary tropical dry forests

(Rüger et al., 2023).

Hemispherical photographs show that B.

nervosa tends to occupy microsites with mod-

erate canopy cover, suggesting a preference for

intermediate light environments. The average

canopy openness at sites where B. nervosa

occurs was 14.16 %, with a LAI of 2.17. These

values fall between those of cloud forests (Syl-

vester & Avalos, 2013) and mature lowland

tropical rainforests, the latter typically exhibit-

ing canopy openness below 10 % and LAI val-

ues approaching six (Bakar et al., 2023; Clark et

al., 2008; Clark et al., 2021; Pfeifer et al., 2018).

During the dry season, the mature forest patch

at SRNP showed a canopy openness of 10.31 %

and an LAI of 2.78 (pers. obs.). These values

show that B. nervosa avoids dense shade and

full exposure and is instead associated with

intermediate light levels.

We consider our findings to be preliminary,

highlighting the need for further research dur-

ing the dry season, including a more through

seedling census and a detailed comparison

of leaf structure and function between seed-

lings and adults. The apparent restriction of

B. nervosa to older secondary forests, and its

absence from a broader range of habitats, sug-

gests limited phenotypic plasticity in coping

with environmental variation. This constraint

may significantly affect the species’ capacity to

adapt to climate change. For instance, Sánchez

et al. (2020) attempted to alter the species’ leaf

phenology through dry-season irrigation, but

10 Revista de Biología Tropical, ISSN: 2215-2075 Vol. 73: e2025456, enero-diciembre 2025 (Publicado Set. 17, 2025)

observed no significant changes, pointing to

a potentially low level of genetic variation in

response to water availability, although their

results were inconclusive. The unique inverted

phenology of B. nervosa underscores the need

for further investigation into the physiological

mechanisms determining its ecological strate-

gy, particularly in the face of increasing climate

variability (Schwinning & Kelly, 2013).

Bonellia nervosa shows a pulse in vegetative

and reproductive growth during the dry season.

The plant’s energy reserves must support the

production of a new canopy and the develop-

ment of flowers and fruits. The fruits continue

to expand and mature for at least five additional

months during the rainy season, further deplet-

ing the plant’s reserves. Additionally, the young

leaves of B. nervosa are exposed to a relatively

diverse array of herbivores during the dry sea-

son, which consume an average of 36.77 % of

their area (Chaves & Avalos, 2007). While the

fruit damage percentage observed in our study

may seem low, the level of infestation by Tor-

tricidae larvae should be monitored throughout

the fruit’s development. Adult moths likely lay

their eggs in the immature ovaries, with the

larvae developing as the fruit expands.

Chaves and Avalos (2008) reported that

only 3 % of the flowers became fruits. This low

reproductive success was associated with low

levels of flower synchrony. B. nervosa is thought

to be dispersed by rodents (Sumichrast’s vesper

rat, Nyctomys sumichrasti, Ceballos, 1990), as

well as by the White-tailed deer, Odocoileus vir-

ginianus, Jara-Guerrero et al., 2018). During the

2024 dry season, white-faced monkeys (Cebus

imitator) were also observed eating their ripe

fruits in SRNP and in Taboga Forest Reserve,

Guanacaste. Horses also have been seen eating

B. nervosa fruits in SRNP (O. Chaves, personal

communication, 2024). Possibly, there is no

shortage of seed dispersers, but the species still

shows low abundance.

Our study shows that B. nervosa has a popu-

lation structure dominated by adults. Seedlings

are scarce, likely because of the high energy

demands for reserve accumulation, vulnerabil-

ity to drought, and herbivory and mechanical

damages. The species occurs in microsites with

moderate canopy openness (~14 %), indicating

a preference for intermediate light conditions

and suggesting it functions as an indicator

of late-successional forest stages. Its absence

from very young and mature forests supports

this interpretation. Although predispersal fruit

predation was relatively low (~16 %), further

monitoring is needed to assess fruit losses over

the entire reproductive period. The species’

slow growth, narrow habitat preferences, and

distinctive reverse phenology point to lim-

ited phenotypic plasticity, justifying the need

for long-term, multi-seasonal studies to better

understand the recruitment dynamics and the

resilience of this species to climate change.

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.

REFERENCES

Avalos, G. (2019). Shade tolerance within the context of the

successional process in tropical rain forests. Revista de

Biología Tropical, 67(2), 53–77.

Bakar, A. A. A., Abd-Latif, Z., & Lee, W. K. (2023). Compa-

rison of canopy cover effects on rainfall interception

loss in tropical forest with homogenous and mixed

tree species. Jurnal Teknologi, 85(2), 1–9.

Bhadouria, R., Singh, R., Srivastava, P., & Raghubanshi, A.

S. (2016). Understanding the ecology of tree-seedling

growth in dry tropical environment: A management

perspective. Energy, Ecology and Environment, 1,

296–309.

Buchadas, A., Baumann, M., Meyfroidt, P., & Kuemmerle,

T. (2022). Uncovering major types of deforestation

frontiers across the world’s tropical dry woodlands.

Nature Sustainability, 5(7), 619–627.

Cao, S., Yu, Q., Sánchez-Azofeifa, A., Feng, J., Rivard, B.,

& Gu, Z. (2015). Mapping tropical dry forest succes-

sion using multiple criteria spectral mixture analysis.

ISPRS Journal of Photogrammetry and Remote Sen-

sing, 109, 17–29.

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 73: e2025456, enero-diciembre 2025 (Publicado Set. 17, 2025)

Castillo, M., Sánchez-Azofeifa, G. A., Croitoru, A., Rivard,

B., Calvo-Alvarado, J., & Dubayah, R. O. (2011).

Delineation of secondary succession mechanisms for

tropical dry forests using LiDAR. Remote Sensing of

Environment, 115, 2217–2231.

Ceballos, G. (1990). Comparative natural history of small

mammals from tropical forests in western Mexico.

Journal of Mammalogy, 71(2), 263–266.

Chaves, O. M. (2002). Efecto de la variación estacional en

la cantidad de luz y de las interacciones planta-herbí-

voro sobre la fenología inversa de Jacquinia nervosa

(Theophrastaceae) [Master’s thesis, Universidad de

Costa Rica]. Repositorio del SIBDI-UCR.

Chaves, O., & Avalos, G. (2007). Is the inverse leafing phe-

nology of the dry forest understory shrub Jacquinia

nervosa (Theophrastaceae) a strategy to escape her-

bivory? Revista de Biología Tropical, 54(3), 951–963.

Chaves, O. M., & Avalos, G. (2008). Do seasonal chan-

ges in light availability influence the inverse leafing

phenology of the neotropical dry forest understory

shrub Bonellia nervosa (Theophrastaceae)? Revista de

Biología Tropical, 56(1), 257–268.

Clark, D. B., Olivas, P. C., Oberbauer, S. F., Clark, D. A., &

Ryan, M. G. (2008). First direct landscape-scale mea-

surement of tropical rain forest Leaf Area Index, a key

driver of global primary productivity. Ecology Letters,

11(2), 163–172.

Clark, D. B., Oberbauer, S. F., Clark, D. A., Ryan, M. G., &

Dubayah, R. O. (2021) Physical structure and biologi-

cal composition of canopies in tropical secondary and

old-growth forests. PloS ONE, 16(8), e0256571.

Domínguez-Calleros, P. A., Rodríguez-Flores, F. D. J.,

Lizárraga-Mendiola, L., Jiménez-Gómez, M. A., &

Navar, J. (2017). Aplicaciones y ejemplos de modelos

de crecimiento diamétrico para árboles tropicales.

Ecosistemas y Recursos Agropecuarios, 4(11), 265–274.

Duan, M., Bax, C., Laakso, K., Mashhadi, N., Mattie,

N., & Sanchez-Azofeifa, A. (2023). Characterizing

transitions between successional stages in a tropical

dry forest using LiDAR techniques. Remote Sensing,

15(2), 479.

Enquist, B. J., & Sullivan, J. J. (2001). Vegetative key and

descriptions of tree species of the tropical dry forests

of upland Sector Santa Rosa, Área de Conservación

Guanacaste, Costa Rica. Tucson, USA.

Frankie, G. W., Baker, H. G., & Opler, P. A. (1974). Compa-

rative phenological studies of trees in tropical wet and

dry forests in the lowlands of Costa Rica. Journal of

Ecology, 62(3), 881–919.

Frazer, G. W., Canham, C. D., & Lertzman, K. P. (1999). Gap

Light Analyzer (GLA), version 2.0: Imaging software

to extract canopy structure and gap light transmission

indices from true-colour fisheye photographs, user’s

manual and program documentation. Simon Fraser

University, British Columbia, and the Institute of

Ecosystem Studies, Millbrook, New York, USA.

Giraldo, J. P., & Holbrook, N. M. (2011). Physiological

mechanisms underlying the seasonality of leaf senes-

cence and renewal in seasonally dry tropical forest

trees. In R. Dirzo, H. S. Young, H. A. Mooney, & G.

Ceballos (Eds.), Seasonally dry tropical forests: ecology

and conservation (pp. 129–140). Island Press.

Glander, K. E., & Nisbett, R. A. (1996). Community struc-

ture and species density in tropical dry forest associa-

tions at Hacienda La Pacifica in Guanacaste Province,

Costa Rica. Brenesia, 45(46), 113–142.

Griscom, H. P., & Ashton, M. S. (2011). Restoration of

dry tropical forests in Central America: A review of

pattern and process. Forest Ecology and Management,

261(10), 1564–1579.

Holdridge, L. R., & Grenke, W. C. (1971). Forest environ-

ments in tropical life zones: A pilot study. Pergamon

Press.

Hubbell, S. P. (1979). Tree dispersion, abundance, and

diversity in a tropical dry forest: That tropical trees are

clumped, not spaced, alters conceptions of the organi-

zation and dynamics. Science, 203(4387), 1299–1309.

Janzen, D. H. (1967). Synchronization of sexual reproduc-

tion of trees within the dry season in Central Ameri-

ca. Evolution, 21(3), 620–637.

Janzen, D. H. (1970). Jacquinia pungens, a heliophile from

the understory of deciduous forest. Biotropica, 2(2),

112–19.

Janzen, D. H. (1983). Jacquinia pungens (Burriquita, Siem-

pre Viva, Siempre Verde, False Evergreen Needle

Bush). In D. H. Janzen (Ed.), Costa Rican Natural

History (pp. 265–266). University of Chicago Press.

Janzen, D. H., & Wilson, D. E. (1974). The cost of being

dormant in the tropics. Biotropica, 6(4), 260–262.

Janzen, D. H., Hallwachs, W., & Kappelle, M. (2016). Bio-

diversity conservation history and future in Costa

Rica: The case of Área de Conservación Guanacaste

(ACG). In M. Kappelle (Ed.), Costa Rican Ecosystems

(pp. 290–341). University of Chicago Press.

Jara-Guerrero, A., Escribano-Avila, G., Espinosa, C. I., De

la Cruz, M., & Méndez, M. (2018). White-tailed deer

as the last megafauna dispersing seeds in Neotropical

dry forests: the role of fruit and seed traits. Biotropica,

50(1), 169–177.

Kalacska, M., Sanchez-Azofeifa, G. A., Calvo-Alvarado, J.

C., Quesada, M., Rivard, B., & Janzen, D. H. (2004).

Species composition, similarity and diversity in three

successional stages of a seasonally dry tropical forest.

Forest Ecology and Management, 200(1-3), 227–247.

12 Revista de Biología Tropical, ISSN: 2215-2075 Vol. 73: e2025456, enero-diciembre 2025 (Publicado Set. 17, 2025)

Lasky, J. R., Uriarte, M., & Muscarella, R. (2016). Syn-

chrony, compensatory dynamics, and the functional

trait basis of phenological diversity in a tropical dry

forest tree community: effects of rainfall seasonality.

Environmental Research Letters, 11(11), 115003.

Lópezaraiza-Mikel, M., Quesada, M., Álvarez-Añorve, M.,

Ávila-Cabadilla, L., Martén-Rodríguez, S., Calvo-

Alvarado, J., & Vázquez-Ramírez, A. (2013). Phenolo-

gical patterns of tropical dry forests along latitudinal

and successional gradients in the Neotropics. In R.

K. Murphy, & A. Sánchez-Azofeifa (Eds.), Tropical

dry forests in the Americas (pp. 119–146). CRC Press.

Miles, L., Newton, A. C., DeFries, R. S., Ravilious, C., May,

I., Blyth, S., Kapos, V., & Gordon, J. E. (2006). A global

overview of the conservation status of tropical dry

forests. Journal of Biogeography, 33(3), 491–505.

Morales, J. F. (2003). Bonellia. In B. F. Hammel, M. H.

Grayum, N. Zamora (Eds.), Manual de plantas de

Costa Rica Vol VIII (pp. 407–408). St. Louis Missouri

Botanical.

Murphy, P. G., & Lugo, A. E. (1986). Ecology of tropical

dry forest. Annual Review of Ecology and Systematics,

17, 67–88.

Murphy, P. G., & Lugo, A. E. (1995). Dry forests of Central

America and the Caribbean. In S. H. Bullock, H. A.

Mooney, & E. Medina (Eds.), Seasonally dry tropical

forests (pp. 9–34). Cambridge University Press.

Oberbauer, S. F. (1985). Plant water relations of selec-

ted species in wet and dry tropical lowland forests

in Costa Rica. Revista de Biología Tropical, 33(2),

137–142.

Pfeifer, M., Gonsamo, A., Woodgate, W., Cayuela, L., Mar-

shall, A. R., Ledo, A., Paine, T. C., Marchant, R., Burt,

A., Calders, K., & Courtney-Mustaphi, C (2018).

Tropical Forest canopies and their relationships with

climate and disturbance: results from a global dataset

of consistent field-based measurements. Forest Ecosys-

tems, 5(1), 1–14.

Quesada, M., Álvarez-Añorve, M., Ávila-Cabadilla, L., Cas-

tillo, A., Lopezaraiza-Mikel, M., Martén-Rodríguez,

S., Rosas-Guerrero, V., Sáyago, R., Sánchez-Monto-

ya, G., Contreras-Sánchez, J. M., Balvino-Olvera, F.,

Olvera-García, S. R., Lopez-Valencia, S., & Valdespi-

no-Vázquez, N. (2014). Tropical dry forest ecological

succession in Mexico: synthesis of a long-term study.

In G. Sánchez-Azofeifa, J. S. Powers, G. W. Fernan-

des, & M. Quesada (Eds.), Tropical dry forests in the

Americas: ecology, conservation, and management (pp.

17–33). CRC Press.

Quesada, M., Sanchez-Azofeifa, G. A., Alvarez-Anorve, M.,

Stoner, K. E., Avila-Cabadilla, L., Calvo-Alvarado,

J., Castillo, A., Espírito-Santo, M. M., Fagundes, M.,

Fernandes, G. W., & Gamon, J. (2009). Succession and

management of tropical dry forests in the Americas:

Review and new perspectives. Forest Ecology and

Management, 258(6), 1014–1024.

R Core Team (2024). R: A Language and Environment

for Statistical Computing [Software]. R Foundation

for Statistical Computing, Vienna, Austria. https://

www.R-project.org/

Romero, E., Terrazas, T., González, E. J., & Meave, J. A.

(2023). Wood anatomical spectrum of co-occurring

species in early and late-successional tropical dry

forest communities. Trees, 37(6), 1609–1625.

Roupsard, O., Ferhi, A., Granier, A., Pallo, F., Depommier,

D., Mallet, B., Joly, H. I., & Dreyer, E. (1999). Reverse

phenology and dry-season water uptake by Faidher-

bia albida (Del.) A. Chev. in an agroforestry parkland

of Sudanese west Africa. Functional Ecology, 13(4),

460–472.

Rüger, N., Schorn, M. E., Kambach, S., Chazdon, R. L.,

Farrior, C. E., Meave, J. A., Muñoz, R., Van Breugel,

M., Amissah, L., Bongers, F., & Craven, D. (2023).

Successional shifts in tree demographic strategies in

wet and dry Neotropical forests. Global Ecology and

Biogeography, 32(6), 1002–1014.

Sánchez-Azofeifa, A., Powers, J. S., Fernandes, G. W., &

Quesada, M. (2013). Tropical dry forests in the Ame-

ricas: ecology, conservation, and management. CRC

Press.

Sánchez, O., Quesada, M., Dirzo, R., & Schlichting, C. D.

(2020). A field experiment to determine the effect of

dry-season irrigation on vegetative and reproductive

traits in the wet-deciduous tree Bonellia nervosa. Jour-

nal of Tropical Ecology, 36(1), 29–35.

Schwinning, S., & Kelly, C. K. (2013). Plant competition,

temporal niches and implications for productivity

and adaptability to climate change in water-limited

environments. Functional Ecology, 27(4), 886–897.

Ståhl, B. (1989). A synopsis of Central American Theophras-

taceae. Nordic Journal of Botany, 9(1), 15–30.

Ståhl, B., & Källersjö, M. (2004). Reinstatement of Bonellia

(Theophrastaceae). Novon, 14(1), 115–118.

Stan, K., & Sanchez-Azofeifa, A. (2019). Tropical dry forest

diversity, climatic response, and resilience in a chan-

ging climate. Forests, 10(5), 443.

Stan, K. D., Sanchez-Azofeifa, A., & Hamann, H. F. (2024).

Widespread degradation and limited protection of

forests in global tropical dry ecosystems. Biological

Conservation, 289, 110425.

Sylvester, O., & Avalos, G. (2013). Influence of light condi-

tions on the allometry and growth of the understory

palm Geonoma undata subsp. edulis (Arecaceae) of

neotropical cloud forests. American Journal of Botany,

100(12), 2357–2363.

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 73: e2025456, enero-diciembre 2025 (Publicado Set. 17, 2025)

Siyum, Z. G. (2020). Tropical dry forest dynamics in the

context of climate change: syntheses of drivers, gaps,

and management perspectives. Ecological Processes,

9(1), 1–16.

Werden, L. K., Averill, C., Crowther, T. W., Calderón-Mora-

les, E., Toro, L., Alvarado, J. P., & Powers, J. S. (2023).

Below-ground traits mediate tree survival in a tropical

dry forest restoration. Philosophical Transactions of

the Royal Society B, 378(1867), 20210067.

Zhao-Gang, L., & Feng-Ri, L. I. (2003). The generalized

Chapman-Richards function and applications to tree

and stand growth. Journal of Forestry Research, 14,

19–26.