Received: 23 December 2024 | Accepted: 30 March 2025

DOI: 10.1111/eea.13570

ORIGINAL ARTICLE
S p e c i a l I s s u e : Tr a i t S e l e c t i o n i n P r o d u c t i o n I n s e c t s

Selection for zoophagy influences biocontrol efficacy and fruit

damage by Dicyphus hesperus in greenhouses

François Dumont | Mireia Solà Cassi | Maud Lemay | Caroline Provost

Centre de Recherche Agroalimentaire de Abstract

Mirabel, Mirabel, Québec, Canada
The zoophytophagous predator Dicyphus hesperus Knight (Hemiptera: Miridae) is
Correspondence effective in the biological control of whiteflies in greenhouses, but it can also cause
François Dumont, Centre de Recherche damage to fruits and plants. Artificial selection on foraging behavior allows the
Agroalimentaire de Mirabel, 9850, Rue de
Belle-­Rivière, Mirabel, QC, Canada.
development of more zoophagous lines that have the potential to be both more
Email: [email protected] effective and less likely to cause damage. Moreover, highly zoophagous lines could
affect other biological control agents through increased intraguild predation or
Funding information
Agriculture and Agri-­Food Canada, Grant/
competition. This study tests the biological control efficacy against tobacco whitefly
Award Number: ASP-­141 Bemisia tabaci (Gennadius) (Hemiptera: Aleyrodidae) and damage by highly and lowly
zoophagous lines of D. hesperus in tomato greenhouses. The effect of these lines on
Encarsia formosa Gahan (Hymenoptera: Aphelinidae) parasitoid wasp populations was
also tested. In cage tests, we introduced D. hesperus from lowly or highly zoophagous
and non-­selected lines. In half of the cage, E. formosa was introduced. The ability of
predators and parasitoids to reduce B. tabaci populations was monitored for 12 weeks.
Tomatoes produced were harvested and graded according to damage by D. hesperus.
Highly zoophagous lines had a rapid and lasting impact on pest populations. Lowly
zoophagous lines take longer to achieve the same level of pest control as highly
zoophagous lines. Introductions of E. formosa also reduce populations, but without
interacting with D. hesperus. Dicyphus hesperus did not affect E. formosa abundance.
Lowly zoophagous lines generated higher proportions of damage. The results show
that artificial selection based on zoophagy produces more efficient and less damaging
lines in the greenhouse tomato crop. Over time, lines with low zoophagy compensated
for low individual efficiency by increasing their numbers. Highly zoophagous lines are
compatible with parasitoid wasps, which were little affected by D. hesperus.

KEYWORDS
biological control, feeding behavior, genetic improvement, heritability, Miridae, omnivorous predators,
parasitoid wasp, predator aggression, tobacco whitefly, tomato greenhouse

I NTRO DUC TIO N (Tuta) absoluta (Meyrick) (Lepidoptera: Gelechiidae) (De

Backer et al., 2014; Gerling et al., 2001; Mollá et al., 2009).
Zoophytophagous predators are effective greenhouse In North America, Dicyphus hesperus Knight (Hemiptera:
predators against whiteflies and other crop pests (Calvo Miridae) effectively reduces populations of green-
et al., 2016, 2018; McGregor et al., 1999; Mollá et al., 2009). house whitefly Trialeurodes vaporariorum (Westwood)
In Europe, Macrolophus pygmaeus (Rambur) (Hemiptera: (Hemiptera: Aleyrodidae) (Calvo et al., 2016, 2018; Dumont
Miridae) is marketed for biological control of the white- et al., 2021; McGregor et al., 1999). Nevertheless, these
fly Bemisia tabaci (Gennadius) (Hemiptera: Aleyrodidae) predators consume plant resources and can cause dam-
and the South American tomato leafminer Phthorimaea age to fruit or plants (Arnó et al., 2010; Castañé et al., 2011;

This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium,
provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.
© 2025 The Author(s). Entomologia Experimentalis et Applicata published by John Wiley & Sons Ltd on behalf of Netherlands Entomological Society.

Entomol Exp Appl. 2025;00:1–10.  wileyonlinelibrary.com/journal/eea | 1

 15707458, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eea.13570 by Cochrane Colombia, Wiley Online Library on [28/04/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2 |    DUMONT et al.

Gillespie et al., 2012; Gillespie & McGregor, 2000; Hamdi or may induce stronger competition between biocontrol
et al., 2013). Artificial selection of foraging behavior can agents. Thus, normally compatible biocontrol agents may
improve the ratio between the benefits and risks associ- no longer be compatible if zoophytophagous predators de-
ated with these predators (Chinchilla-­Ramírez et al., 2020; rived from artificial selection are used (Dumont et al., 2018).
Dumont et al., 2018, 2019; Dumont, Lucas, & Réale, 2017). In this study, we are testing, in the greenhouse, the ef-
Highly zoophagous lines can be developed to increase ficacy of highly and lowly zoophagous lines of D. hesperus
benefits, either through artificial selection from gen- in regulating populations of the tobacco whitefly, B. tabaci,
eration to generation or by selecting isofemale lines and the rate of damage to tomatoes caused by these lines.
(Bielza et al., 2020; Dumont et al., 2016, 2018; Nachappa In addition, we are testing the impact of the D. hesperus
et al., 2010). Furthermore, the tradeoff between zoophagy lines on populations of the parasitoid wasp Encarsia for-
and phytophagy observed in zoophytophagous predators mosa Gahan (Hymenoptera: Aphelinidae) and the compat-
could reduce the incidence of damage by selecting highly ibility of these two biological control agents when using
zoophagous individuals (Dumont et al., 2018; Dumont, different kinds of D. hesperus lines.
Lucas, & Réale, 2017).
A high level of zoophagy can increase pressure on prey
(Dumont et al., 2019; Nachappa et al., 2011), but also mod- M E TH O DS
ulate predator population dynamics through antagonis-
tic interactions, notably cannibalism (Dumont, Réale, & Study sites
Lucas, 2017). Thus, the efficiency of predators selected for
their high level of zoophagy can vary over time and modu- The project was carried out in the laboratories and
late predator–prey dynamics (Nachappa et al., 2011). Lines greenhouses of the Centre de recherche agroalimentaire de
with different traits can also be effective in biological con- Mirabel (CRAM) in Sainte-­Scholastique (Mirabel, Quebec).
trol, but use different strategies (Nachappa et al., 2011). For
example, Nachappa et al. (2011) observed that lines of the
predatory mite Phytoseiulus persimilis Athias-­Henriot (Acari: Populations and rearing
Phytoseiidae) selected for their high voracity, high conver-
sion rate, and propensity for dispersal equitably regulated The D. hesperus source population was reared in cloth
populations of the two-­spotted spider mite Tetranychus ur- cages containing six to nine mullein plants (depending on
ticae Koch (Acari: Tetranychidae). However, these lines had their size). Eggs of Ephestia kuehniella (Zeller) (Lepidoptera:
different population dynamics and spatial correlation with Pyralidae) were provided as a food source. Rearing was
prey. Highly voracious lines tended to eliminate all prey conducted under laboratory conditions at 25°C, 55%
from a food patch before exploring to find a new patch relative humidity, and 16 h of light. The source population
(Nachappa et al., 2010). Thus, this type of lines has an im- was established on October 28, 2020, from five different
pact on the abundance and distribution of its prey. sources: our existing laboratory colony, two biological
Damage caused by zoophytophagous predators control suppliers (Anatis Bioprotection and Natural Insect
could be modulated by behavioral differences (Dumont Control), and two greenhouses, one in Quebec and the
et al., 2018) and ecological conditions (Gillespie & other in Ontario, Canada.
McGregor, 2000; Sinia et al., 2004). Zoophytophagous Parasitoid wasps, E. formosa, were supplied by Anatis
predators cause more damage when their prey is scarcer Bioprotection Inc.
(Calvo et al., 2009). If highly zoophagous lines reduce pest The silverleaf whiteflies, B. tabaci, were reared on egg-
populations more efficiently, they would create favorable plant plants under the same climatic conditions as the
conditions for increased damage by zoophytophagous Dicyphus. The rearing population originated from col-
predators (i.e., low prey density). However, an increase in lections made in a greenhouse in Quebec in 2018 and
cannibalism could self-­regulate their population (Dumont, 2019. Biotype B was identified by genetic methods by the
Réale, & Lucas, 2017). Lines with a low level of zoophagy Laboratoire d'expertise et de diagnostic en phytoprotec-
could be specialized on a plant diet (Dumont et al., 2018; tion of the Ministère de l'Agriculture, des Pêcheries et de
Dumont, Lucas, & Réale, 2017) and have denser populations l'Alimentation du Québec (MAPAQ, Québec, Canada).
without the incidence of cannibalism (Dumont, Réale, &
Lucas, 2017). Thus, these lines could cause more damage to
crops than highly zoophagous lines (Dumont et al., 2018). Tests of zoophagy and artificial selection of
However, this hypothesis remains to be verified. Dicyphus hesperus
Artificial selection based on zoophagy is likely to in-
fluence interspecific interactions (Dumont et al., 2018). The methodology and results for tests of zoophagy and
Zoophytophagous predators are frequently used simul- artificial selection were published in Dumont et al. (2024).
taneously with other biological control agents (Bennett Zoophagy was assessed in virgin D. hesperus adults less
et al., 2009; McGregor & Gillespie, 2005). Highly zoophagous than 14 days old. Each individual was isolated in a cup
lines may have a higher propensity for intraguild predation with lettuce and Ephestia eggs until adulthood. After a
 15707458, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eea.13570 by Cochrane Colombia, Wiley Online Library on [28/04/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ZOOPHAGY'S IMPACT ON BIOCONTROL AND DAMAGE BY DICYPHUS HESPERUS    | 3

24-­h fasting period, adults were placed in a Petri dish with estimated by counting the number of individuals (by age
Ephestia eggs and agar cubes to observe their feeding class: L1–L3, L4–L5, and adults) present on five leaves of the
behavior. The number of Ephestia eggs consumed was mullein plant. To monitor whitefly populations, 20 tomato
counted after the predator was removed. leaflets per cage were collected at random and brought
Three line groups were established: highly zoophagous, back to the laboratory in Ziploc bags. In the laboratory,
lowly zoophagous, and control (no selection). Artificial se- three circular areas 1 cm in diameter were marked out
lection started with 287 adults (136 males and 151 females) on each leaflet with a cookie cutter. The total number
and continued from generation G1 to G4. The individuals of whitefly nymphs and pupae, as well as the number
with the highest or lowest zoophagy rates, depending on of whitefly pupae parasitized in the three zones, were
selection direction, were selected for each generation. counted for each leaflet.
Selection tests were spaced 6 weeks apart to allow the pro-
duction of a new generation. In the control lines with no
selection, each generation started with adults randomly Yield and tomato damage measurement
chosen from the previous generation.
After this procedure, the lines of each selection type Ripe tomatoes were harvested weekly from August 31 to
were grouped and then mass-­reared to conduct the green- October 28, 2022. Harvested tomatoes were brought back
house tests. to the laboratory. The total weight and number of tomatoes
harvested per cage and the number of tomatoes damaged
by D. hesperus per cage were determined.
Greenhouse experiment

Greenhouse tests were conducted from July 19 to October Data analysis

28, 2022. They were conducted in 120 muslin cages
containing five plants: four cherry tomato plants and one A generalized linear mixed model (GLMM) for negative
mullein plant. binomial distribution was used to describe temporal
The whiteflies, B. tabaci, were introduced on July 19, variations in B. tabaci (immature stages) as a function
2022, a week before the D. hesperus to give the individu- of the “D. hesperus” and “E. formosa” line treatments
als a longer infestation period. A tomato plant previously (fixed effects). For the “time” variable (number of weeks
infested with 20 B. tabaci adults was introduced into each after introducing D. hesperus into the cages), linear and
cage in the experiment. Thus, of the four tomato plants polynomial relationships with 2 and 3 degrees were
in the cages, only one was initially infested by whiteflies. compared. Models were compared using likelihood ratio
These then infested the other plants. tests using the ANOVA function. The statistical significance
Two variables were controlled: D. hesperus lines and of fixed effects (explanatory variables) was tested using
the introduction of parasitoids (E. formosa). Treatments for a likelihood ratio test with the drop1 function in R v.4.4.3.
the Dicyphus variable were as follows: (1) control without Statistical differences between treatments of significant
Dicyphus; (2) line without selection; (3) low zoophagy line; fixed variables were estimated with a Tukey test using the
and (4) high zoophagy line. The low and high zoophagy glht function in the multcomp library (Hothorn et al., 2016).
lines were derived from an artificial selection experiment. The cage number was included as a random effect to
To increase the numbers of these lines, they were placed account for repeated measures over time.
in large muslin cages (48 cm × 48 cm × 48 cm) with five mul- The same approach was used to describe temporal vari-
lein plants several weeks before their introduction into the ations in the abundance of E. formosa (immature stages)
greenhouse. In each line with D. hesperus, four pairs were and their rate of parasitism on B. tabaci and D. hesperus. For
introduced per cage on July 26th. E. formosa abundance, the analyses exclude the first two
The parasitoids were introduced in half of the cages, observation periods because the parasitoid was absent or
while the other half served as the control group and did rare. For the rate of parasitism, a GLMER model for propor-
not contain parasitoids. Ten E. formosa pupae were intro- tional distribution (binomial) was used. In both models, the
duced on August 10, after the first monitoring of Dicyphus “E. formosa” variable was not included, and only cages with
populations. E. formosa were considered.
There were 15 replicates for each treatment combina- For temporal variations in D. hesperus, a different model
tion (Dicyphus x Encarsia). was run for small nymphs (N1–N3), large nymphs (N4–N5)
and adults. For large nymphs, the first two observation pe-
riods were excluded from the analyses because this devel-
Population monitoring opmental stage was absent or very rare at these times. For
the other developmental stages, all observation periods
Dicyphus hesperus populations were monitored every were included in the analyses.
2 weeks from August 10th to October 18, 2022, for a total of Tomato yield was estimated based on the total weight
11 weeks (six surveys). Dicyphus hesperus populations were of tomatoes harvested over the entire observation period.
 15707458, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eea.13570 by Cochrane Colombia, Wiley Online Library on [28/04/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4 |    DUMONT et al.

F I G U R E 1 Abundance of Bemisia tabaci nymphs and nymphs as a function of time and treatments with Encarsia formosa (A) and Dicyphus
hesperus (B) The lineplots show the median value and standard error (SE).

The rate of tomato damage caused by D. hesperus (num-

ber of damaged tomatoes divided by total number of to-
matoes) was tested with a linear generalized model (GLM)
for binomial distribution accounting for overdispersion of
the data (quasibinomial). The explanatory variables were
the “D. hesperus” and “E. formosa” lines. The glht function
was used to compare the levels of significant variables.
All statistical analyses were run in R v. 4.3.3.

R ESULTS

Effect of Dicyphus hesperus and Encarsia

formosa on Bemisia tabaci

Bemisia tabaci abundance varied over time according to a

quadratic function (GLMM, LRT3 = 150.01; p < 0.0001). The D.
hesperus (GLMM, LRT3 = 14.25; p = 0.003) (Figure 1A) and E.
formosa (LRT2 = 7.19; p = 0.007) (Figure 1B) lines influenced B.
tabaci abundance over the entire observation period, from
Weeks 2 to 12. The “high zoophagy” and “low zoophagy”
F I G U R E 2 Effect of the presence of Dicyphus hesperus on the lines significantly reduced B. tabaci abundance compared
number of immature Bemisia tabaci observed according to the with the control treatment (Figure 2). No interaction was
zoophagy of the lines. The boxplots show the median value and
observed between the D. hesperus and E. formosa lines
standard error (SE). Different letters indicate significant differences
between groups (Tukey–Kramer: p < 0.05). (LRT3 = 0.15; p = 0.985).

Effect of Dicyphus hesperus on Encarsia

The effect of the “D. hesperus” and “E. formosa” treatments formosa population
on yield was tested using analysis of variance (ANOVA). A
Tukey test was used to compare one by one the levels of E. formosa abundance varied according to a cubic
the significant explanatory variables. relationship with time (GLMM, LRT3 = 38.23; p < 0.0001)
 15707458, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eea.13570 by Cochrane Colombia, Wiley Online Library on [28/04/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ZOOPHAGY'S IMPACT ON BIOCONTROL AND DAMAGE BY DICYPHUS HESPERUS    | 5

F I G U R E 3 Abundance of immature Encarsia formosa (A) and parasitism rate (B) over time and according to Dicyphus hesperus line treatments The
lineplots show the median value and standard error (SE).

(Figure 3A). Over the entire E. formosa observation period Temporal variations in adult abundance followed a
(Weeks 6 to 12), D. hesperus had no significant effect on E. four-­stage polynomial relationship (GLMM, LRT2 = 170.97;
formosa abundance (GLMM, LRT3 = 6.45; p = 0.09). p < 0.0001) (Figure 4C). Adult D. hesperus were more abun-
The rate of parasitism of E. formosa on B. tabaci varied over dant in the “low zoophagy” treatment than in the “no se-
time according to a cubic function (GLMM, LRT3 = 885.28; lection” treatment (GLMM, LRT2 = 6.47; p = 0.04) (Figure 5C).
p < 0.0001) (Figure 3B). Dicyphus hesperus lines had no ef- The “high zoophagy” treatment was indistinguishable
fect on this rate of parasitism (GLMM, LRT3 = 4.48; p = 0.21). from the “low zoophagy” and “no selection” treatments.
The presence of E. formosa had no effect on the abundance
of adult D. hesperus (GLMM, LRT1 = 0.04; p = 0.85).
Dicyphus hesperus populations

Temporal variations in the abundance of small nymphs Effect of Dicyphus hesperus and Encarsia
are described by a cubic function (GLMM, LRT3 = 27.57; formosa on yield and damage
p < 0.0001) (Figure 4A). Small nymphs (N1–N3) had two
peak periods, from 4 to 6 weeks after the first introduction Yield (in kg) was higher in cages with the “high zoophagy”
of D. hesperus and at 12 weeks. On average, over the whole and “low zoophagy” treatments compared with the “con-
observation period, the abundance of small nymphs trol” and “no selection” treatments (ANOVA, F3,1075 = 11.03;
(N1–N3) of D. hesperus was higher in the “low zoophagy” p < 0.0001) (Figure 6A). Encarsia formosa had no effect on
treatment than in the “no selection” treatment (GLMM, tomato plant yield (ANOVA, F1,1075 = 0.95; p = 0.33) (Figure 6B).
LRT2 = 6.22; p = 0.04) (Figure 5A). The “high zoophagy” The rate of damage to tomatoes by D. hesperus was
treatment had an abundance intermediate between the higher in the “low zoophagy” treatment than in the “high
“low density” and “no selection” treatments. The presence zoophagy” or “no selection” treatments (GLM, LRT2 = 117.78;
of E. formosa had no effect on the abundance of D. hesperus p < 0.0001) (Figure 7A). This rate was also higher in the pres-
(GLMM, LRT1 = 0.60; p = 0.44). ence of E. formosa than in its absence (GLM, LRT1 = 13.52;
From Weeks 6 to 12, the relationship between the abun- p = 0.0002) (Figure 7B).
dance of large D. hesperus nymphs and time followed a
three-­ stage polynomial function (GLMM, LRT3 = 28.30;
p < 0.0001) (Figure 4B). Large nymphs had two peak peri- D ISCUSSIO N
ods at 6 and 12 weeks. Line treatments (GLMM, LRT2 = 2.86;
p = 0.24) (Figure 5B) and E. formosa (GLMM, LRT1 = 3.13; Phenotypic and genotypic variations in the feeding behav-
p = 0.08) had no significant effect on the abundance of ior of biological control agents influence ecological interac-
large nymphs (L4–L5) of D. hesperus. tions and, consequently, the economic value of these agents
 15707458, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eea.13570 by Cochrane Colombia, Wiley Online Library on [28/04/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
6 |    DUMONT et al.

F I G U R E 4 Abundance of N1–N3 nymphs (A), N4–N5 nymphs (B), and adult Dicyphus hesperus (C) according to line treatments The lineplots show
the median value and standard error (SE).

F I G U R E 5 Average number of N–N3 nymphs (A), N4–N5 nymphs (B), and adult Dicyphus hesperus (C) according to the zoophagy of the lines (w/o
select: without selection; low zoo: lowly zoophagous lines; high zoo: highly zoophagous lines). The boxplots show the median value and standard
error (SE). Different letters indicate significant differences between groups (Tukey–Kramer: p < 0.05).

in agricultural environments (Dumont et al., 2018, 2019). For lines of D. hesperus rapidly regulated the B. tabaci pest popu-
zoophytophagous predators, such as D. hesperus, the pro- lations. Over a longer period, the low zoophagous lines had
pensity to consume prey (zoophagy) or plant resources a comparable efficacy to the high zoophagous lines. The
(phytophagy) is a determining factor in the benefit/risk ratio low zoophagy lines could compensate for their low individ-
associated with these biological control agents (Dumont ual efficiency by being superior in number compared with
et al., 2018, 2019). Our results show that highly zoophagous the other types of lines. However, this abundance of lowly
 15707458, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eea.13570 by Cochrane Colombia, Wiley Online Library on [28/04/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ZOOPHAGY'S IMPACT ON BIOCONTROL AND DAMAGE BY DICYPHUS HESPERUS    | 7

F I G U R E 6 Yield (kg) of tomato plants per cage according to the zoophagy of the Dicyphus hesperus lines (A) and the presence or absence of
Encarsia formosa (B). The boxplots show the median value and standard error (SE). Different letters indicate significant differences between groups
(Tukey–Kramer: p < 0.05).

zoophagous individuals caused more damage to tomatoes including zoophagy and cannibalism (Chang, Ng, &
than to highly zoophagous or non-­selected lines. Thus, the Li, 2017; Chang, Teo, et al., 2017; Dumont et al., 2018;
highly and lowly zoophagous lines differed in their effects Michalko et al., 2021; Royer et al., 2022; Schausberger &
on population and prey dynamics, and on tomato damage. Croft, 2000). Aggression is one of the dimensions of animal
The relationship between the benefits and risks associated personality (Careau et al., 2010; Réale et al., 2007; Toscano
with this zoophytophagous predator in greenhouse tomato et al., 2016) and is described as an agonistic interaction
production is therefore modulated by artificial selection toward a conspecific (Réale et al., 2007). Although mea-
based on their level of zoophagy. sured in a social context, aggression can be expressed
Intraguild trophic interactions and competition between in interactions with prey (Belgrad & Griffen, 2016; Chang,
zoophytophagous predators and parasitoid wasps are likely Ng, & Li, 2017; Chang, Teo, et al., 2017) or predators (Dall
to influence the efficacy of biological control in systems et al., 2012; Dochtermann & Dingemanse, 2013; Dufty
with several biological control agents (Bennett et al., 2009; Jr., 1989; Melotto et al., 2019; Michalko & Řežucha, 2018;
McGregor & Gillespie, 2005; Perdikis et al., 2014; Velasco-­ Sih & Bell, 2008). For example, Chang, Ng, and Li (2017)
Hernández et al., 2013). In our experiment, lines of D. hespe- and Chang, Teo, et al. (2017) observed that aggressive
rus and E. formosa had no additive effect on the regulation Portia labiata (Thorell) spiders (Arachnida: Salticidae)
of B. tabaci populations. Furthermore, D. hesperus lines had are more reactive than non-­ aggressive individuals in
no significant effect on E. formosa abundance or parasitism the presence of their preferred prey. Moreover, aggres-
rates. Similarly, the presence of E. formosa had no effect on siveness probably influences different predator behav-
the abundance of D. hesperus. Our results, therefore, do not iors through pleiotropy (Laine & van Oers, 2017; Martins
suggest that potential intraguild predation by D. hesperus & Bhat, 2014; Sih et al., 2004; Sih & Bell, 2008). In a so-
on E. formosa has a significant impact. Similarly, competition cial context, aggressiveness can modulate population
between the two biological control agents appears negligi- dynamics and density (Briffa et al., 2015; Modlmeier
ble over a 12-­week period. Nevertheless, although E. formosa et al., 2015; Sih et al., 2012). In particular, highly zooph-
plays a role in regulating B. tabaci, their presence does not agous lines are likely to engage in more cannibalism
influence tomato plant yield. Conversely, D. hesperus lines (Dumont, Réale, & Lucas, 2017), which can affect de-
with high and low zoophagy yielded higher tomato yields. mography in insect populations (Fox, 1975; Richardson
Artificial selection for D. hesperus zoophagy does not, there- et al., 2010). Dumont, Réale, and Lucas (2017) observed
fore, generate a particular increase in intraguild predation, that strongly zoophagous lines of Campylomma verbasci
does not compromise compatibility between D. hesperus (Meyer-­Dür) (Hemiptera: Miridae) cannibalize more than
and E. formosa, and does not specifically affect plant yield. lowly zoophagous lines. Our greenhouse results seem to
The level of aggressiveness of individuals is likely to confirm this trend. The population dynamics of highly
influence various ecological behaviors and interactions, zoophagous lines of D. hesperus are marked by a rapid
 15707458, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eea.13570 by Cochrane Colombia, Wiley Online Library on [28/04/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
8 |    DUMONT et al.

F I G U R E 7 Rate of tomato damage (%) according to the zoophagy of the Dicyphus hesperus lines (A) and the presence or absence of Encarsia
formosa (B). The boxplots show the median value and standard error (SE). Different letters indicate significant differences between groups (Tukey–
Kramer: p < 0.05).

decline in young nymphs once they reach their first peak CO NCLUSIO N
period. Conversely, lowly zoophagous lines maintain a
high population of young nymphs despite the expected In conclusion, this study highlights the importance of
cycle variations. As a result, populations of lowly zooph- variations in the feeding behavior of zoophytophagous
agous D. hesperus are distinguished by their high density. predators, such as D. hesperus, for the effectiveness of
The level of aggressiveness of lowly zoophagous D. hes- biological control against pests like B. tabaci in tomato
perus, which is expected to be low, would, therefore, en- greenhouses. The results emphasize that artificial
able high population densities to be maintained. selection for high zoophagy improves the rapid control
Artificial selection for zoophagy by zoophytophagous of pest populations in the short term while reducing crop
predators increases benefits and reduces damage by damage. In contrast, in the long term, lowly zoophagous
these organisms (Chinchilla-­Ramírez et al., 2020; Dumont lines, although less efficient individually, maintain higher
et al., 2018). The beneficial effects of highly zoophagous population densities, which can lead to increased plant
lines are obtained in the short term, that is, within the damage. Ultimately, understanding these behavioral
first 4 weeks after the introduction of D. hesperus. During interactions is essential to optimizing the use of D. hesperus
this period, the density of highly and lowly zoophagous in pest management and maximizing economic benefits
D. hesperus populations is similar. The populations de- in agriculture.
rived from these two types of lines are, therefore, dis-
tinguished by their behavior and not by their density. AU T H O R CO N T R I BU T I O N S
Highly zoophagous individuals consume more prey per François Dumont: Conceptualization; data curation;
day than low zoophagous individuals, resulting in bet- methodology; writing – original draft; writing – review
ter regulation of pest populations. Over the long term, and editing; formal analysis. Mireia Solà Cassi:
populations of lowly zoophagous D. hesperus maintain Conceptualization; writing – original draft; writing – review
higher population densities. Strength in numbers com- and editing. Maud Lemay: Investigation; methodology;
pensates for low zoophagy. However, in zoophytopha- writing – review and editing. Caroline Provost: Funding
gous predators, a compromise between zoophagy and acquisition; supervision; validation; writing – review and
phytophagy is observed (Dumont, Lucas, & Réale, 2017). editing; project administration.
Thus, a high density of lowly zoophagous D. hesperus re-
sults in a higher proportion of damage to tomatoes. The ACK N OW LE D G E M E N T S
ratio between the benefits and damage associated with Funding for this project has been provided in part through
high and low zoophagy lines indicates that the former is the AgriScience program-­cluster on behalf of Agriculture
preferable to the latter in the biological control of white- and Agri-­Food Canada (AAC) (grand number: ASP-­ 141,
flies in tomato greenhouses. 2018–2023).
 15707458, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eea.13570 by Cochrane Colombia, Wiley Online Library on [28/04/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ZOOPHAGY'S IMPACT ON BIOCONTROL AND DAMAGE BY DICYPHUS HESPERUS    | 9

CO N FLI C T O F I N T E R E S T S TAT E M E N T De Backer, L., Megido, R.C., Haubruge, É. & Verheggen, F.J. (2014)
Macrolophus pygmaeus (Rambur) as an efficient predator of the to-
The authors declare that they have no known competing
mato leafminer Tuta absoluta (Meyrick) in Europe. A review. Base.
financial interests or personal relationships that could have Dochtermann, N.A. & Dingemanse, N.J. (2013) Behavioral syndromes as
appeared to influence the work reported in this paper. evolutionary constraints. Behavioral Ecology, 24, 806–811.
Dufty, A.M., Jr. (1989) Testosterone and survival: a cost of aggressiveness?
DATA AVAI L AB I LIT Y S TAT E M E N T Hormones and Behavior, 23(2), 185–193. Available from: https://​doi.​
org/​10.​1016/​0018-­​506x(89)​90059​-­​7
Datasets analyzed for the current study can be ob-
Dumont, F., Aubry, O. & Lucas, E. (2018) From evolutionary aspects of zoo-
tained from the corresponding author upon request phytophagy to biological control. Frontiers in Ecology and Evolution,
or via the open repository Zenodo at https://​doi.​org/​ 6, 221.
10.​5281/​zenodo.​14547931 (reference number: 10.5281/ Dumont, F., Lucas, E. & Réale, D. (2016) Evidence of genetic basis of zoo-
zenodo.14547931). phagy and nymphal developmental time in isogroup lines of the
zoophytophagous mullein bug, Campylomma verbasci. BioControl,
61, 425–435.
O R CI D Dumont, F., Lucas, E. & Réale, D. (2017) Coexistence of zoophytophagous
François Dumont https://orcid.org/0000-0001-7587-072X and phytozoophagous strategies linked to genotypic diet special-
ization in plant bug. PLoS One, 12, e0176369.
Dumont, F., Maisonhaute, J.-­É., Labrie, G. & Provost, C. (2021) Biological
REFERENCES control of silverleaf whitefly Bemisia tabaci (Hemiptera: Aleyrodidae)
Arnó, J., Castañé, C., Riudavets, J. & Gabarra, R. (2010) Risk of dam- using predatory bugs, Dicyphus hesperus (Hemiptera: Miridae) and
age to tomato crops by the generalist zoophytophagous pred- Orius insidiosus (Hemiptera: Anthocoridae). Phytoprotection, 101,
ator Nesidiocoris tenuis (Reuter) (Hemiptera: Miridae). Bulletin of 38–45.
Entomological Research, 100, 105–115. Dumont, F., Réale, D. & Lucas, E. (2017) Isogroup selection to optimize bio-
Belgrad, B.A. & Griffen, B.D. (2016) Predator–prey interactions mediated control increases cannibalism in omnivorous (Zoophytophagous)
by prey personality and predator hunting mode. Proceedings of the bugs. Insects, 8, 74.
Royal Society B: Biological Sciences, 283, 20160408. Dumont, F., Réale, D. & Lucas, É. (2019) Can Isogroup selection of highly
Bennett, J.A., Gillespie, D.R., Shipp, J.L. & Vanlaerhoven, S.L. (2009) zoophagous lines of a Zoophytophagous bug improve biocontrol
Foraging strategies and patch distributions: intraguild interac- of spider mites in apple orchards? Insects, 10, 303.
tions between Dicyphus hesperus and Encarsia formosa. Ecological Dumont, F., Solà Cassi, M., Lemay, M. & Provost, C. (2024) Artificial selec-
Entomology, 34(1), 58–65. Available from: https://​doi.​org/​10.​1111/j.​ tion of zoophagous lines of the biological control agent Dicyphus
1365-­​2311.​2008.​01043.​x hesperus. Entomologia Experimentalis et Applicata, 172, 874–882.
Bielza, P., Balanza, V., Cifuentes, D. & Mendoza, J.E. (2020) Challenges Fox, L.R. (1975) Cannibalism in natural populations. Annual Review of
facing arthropod biological control: identifying traits for genetic Ecology and Systematics, 6, 87–106.
improvement of predators in protected crops. Pest Management Gerling, D., Alomar, Ò. & Arnò, J. (2001) Biological control of Bemisia ta-
Science, 76, 3517–3526. baci using predators and parasitoids. Crop Protection, 20, 779–799.
Briffa, M., Sneddon, L.U. & Wilson, A.J. (2015) Animal personality as a Available from: https://​doi.​org/​10.​1016/​S0261​-­​2194(01)​00111​-­​9
cause and consequence of contest behaviour. Biology Letters, 11, Gillespie, D.R. & McGregor, R.R. (2000) The functions of plant feeding in
20141007. the omnivorous predator Dicyphus hesperus: water places limits on
Calvo, F.J., Torres, A., González, E.J. & Velázquez, M.B. (2018) The potential predation. Ecological Entomology, 25, 380–386.
of Dicyphus hesperus as a biological control agent of potato psyl- Gillespie, D.R., VanLaerhoven, S.L., McGregor, R.R., Chan, S. & Roitberg,
lid and sweetpotato whitefly in tomato. Bulletin of Entomological B.D. (2012) Plant feeding in an omnivorous mirid, Dicyphus hespe-
Research, 108, 765–772. rus: why plant context matters. Psyche: A Journal of Entomology,
Calvo, F.J., Torres-­Ruiz, A., Velázquez-­González, J.C., Rodríguez-­Leyva, 2015, 805.
E. & Lomeli-­Flores, J.R. (2016) Evaluation of Dicyphus hesperus for Hamdi, F., Chadoeuf, J. & Bonato, O. (2013) Functional relationships be-
biological control of sweet potato whitefly and potato psyllid on tween plant feeding and prey feeding for a zoophytophagous bug.
greenhouse tomato. BioControl, 61, 415–424. Physiological Entomology, 38, 241–245.
Calvo, J., Bolckmans, K., Stansly, P.A. & Urbaneja, A. (2009) Predation Hothorn, T., Bretz, F., Westfall, P., Heiberger, R.M., Schuetzenmeister, A.,
by Nesidiocoris tenuis on Bemisia tabaci and injury to tomato. Scheibe, S. et al. (2016) Package ‘multcomp.’ Simultaneous inference
BioControl, 54, 237–246. in general parametric models. Vienna, Austria: Project for Statistical
Careau, V., Réale, D., Humphries, M.M. & Thomas, D.W. (2010) The pace of Computing.
life under artificial selection: personality, energy expenditure, and Laine, V.N. & van Oers, K. (2017) The quantitative and molecular genetics
longevity are correlated in domestic dogs. The American Naturalist, of individual differences in animal personality. In: Vonk, J., Weiss,
175, 753–758. A. & Kuczaj, S.A. (Eds.) Personality in Nonhuman Animals. Germany:
Castañé, C., Arnó, J., Gabarra, R. & Alomar, O. (2011) Plant damage to Springer, pp. 55–72.
vegetable crops by zoophytophagous mirid predators. Biological Martins, E.P. & Bhat, A. (2014) Population-­level personalities in zebrafish:
Control, 59, 22–29. aggression-­boldness across but not within populations. Behavioral
Chang, C., Teo, H.Y., Norma-­Rashid, Y. & Li, D. (2017) Predator personality Ecology, 25, 368–373.
and prey behavioural predictability jointly determine foraging per- McGregor, R.R. & Gillespie, D.R. (2005) Intraguild predation by the gener-
formance. Scientific Reports, 7, 1–8. alist predator Dicyphus hesperus on the parasitoid Encarsia formosa.
Chang, C.-­C., Ng, P.J. & Li, D. (2017) Aggressive jumping spiders make Biocontrol Science and Technology, 15, 219–227.
quicker decisions for preferred prey but not at the cost of accuracy. McGregor, R.R., Gillespie, D.R., Quiring, D.M. & Foisy, M.R. (1999) Potential
Behavioral Ecology, 28, 479–484. use of Dicyphus hesperus knight (Heteroptera: Miridae) for biologi-
Chinchilla-­Ramírez, M., Pérez-­Hedo, M., Pannebakker, B.A. & Urbaneja, A. cal control of pests of greenhouse tomatoes. Biological Control, 16,
(2020) Genetic variation in the feeding behavior of Isofemale lines 104–110.
of Nesidiocoris tenuis. Insects, 11, 513. Melotto, A., Ficetola, G.F. & Manenti, R. (2019) Safe as a cave? Intraspecific
Dall, S.R., Bell, A.M., Bolnick, D.I. & Ratnieks, F.L. (2012) An evolutionary aggressiveness rises in predator-­devoid and resource-­depleted en-
ecology of individual differences. Ecology Letters, 15, 1189–1198. vironments. Behavioral Ecology and Sociobiology, 73, 1–14.
 15707458, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/eea.13570 by Cochrane Colombia, Wiley Online Library on [28/04/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
10 |    DUMONT et al.

Michalko, R., Gibbons, A.T., Goodacre, S.L. & Pekár, S. (2021) Foraging Schausberger, P. & Croft, B.A. (2000) Cannibalism and intraguild preda-
aggressiveness determines trophic niche in a generalist biological tion among phytoseiid mites: are aggressiveness and prey pref-
control species. Behavioral Ecology, 32, 257–264. erence related to diet specialization? Experimental and Applied
Michalko, R. & Řežucha, R. (2018) Top predator's aggressiveness and me- Acarology, 24, 709–725.
sopredator's risk-­aversion additively determine probability of pre- Sih, A. & Bell, A.M. (2008) Insights for behavioral ecology from behavioral
dation. Behavioral Ecology and Sociobiology, 72, 1–8. syndromes. Advances in the Study of Behavior, 38, 227–281.
Modlmeier, A.P., Keiser, C.N., Wright, C.M., Lichtenstein, J.L. & Pruitt, Sih, A., Bell, A.M., Johnson, J.C. & Ziemba, R.E. (2004) Behavioral syn-
J.N. (2015) Integrating animal personality into insect popula- dromes: an integrative overview. The Quarterly Review of Biology, 79,
tion and community ecology. Current Opinion in Insect Science, 9, 241–277.
77–85. Sih, A., Cote, J., Evans, M., Fogarty, S. & Pruitt, J. (2012) Ecological implica-
Mollá, Ó., Montón, H., Vanaclocha, P., Beitia, F. & Urbaneja, A. (2009) tions of behavioural syndromes. Ecology Letters, 15, 278–289.
Predation by the mirids Nesidiocoris tenuis and Macrolophus pyg- Sinia, A., Roitberg, B., McGregor, R.R. & Gillespie, D.R. (2004) Prey feeding
maeus on the tomato borer Tuta absoluta. Predation by the Mirids increases water stress in the omnivorous predator Dicyphus hespe-
Nesidiocoris Tenuis and Macrolophus Pygmaeus on the Tomato Borer rus. Entomologia Experimentalis et Applicata, 110, 243–248.
Tuta Absoluta, 49, 209–214. Toscano, B.J., Gownaris, N.J., Heerhartz, S.M. & Monaco, C.J. (2016)
Nachappa, P., Margolies, D.C., Nechols, J.R. & Campbell, J.F. (2011) Personality, foraging behavior and specialization: integrating behav-
Variation in predator foraging behaviour changes predator–prey ioral and food web ecology at the individual level. Oecologia, 182,
spatio-­temporal dynamics. Functional Ecology, 25, 1309–1317. 55–69.
Nachappa, P., Margolies, D.C., Nechols, J.R. & Morgan, T.J. (2010) Response Velasco-­Hernández, M.C., Ramirez-­Romero, R., Cicero, L., Michel-­Rios, C.,
of a complex foraging phenotype to artificial selection on its com- Desneux N. (2013) Intraguild predation on the whitefly parasitoid
ponent traits. Evolutionary Ecology, 24, 631–655. Eretmocerus eremicus by the generalist predator Geocoris punctipes:
Perdikis, D., Lucas, E., Garantonakis, N., Giatropoulos, A., Kitsis, P., a behavioral approach. PLoS One, 8, e80679.
Maselou, D. et al. (2014) Intraguild predation and sublethal inter-
actions between two zoophytophagous mirids, Macrolophus pyg-
maeus and Nesidiocoris tenuis. Biological Control, 70, 35–41.
Réale, D., Reader, S.M., Sol, D., McDougall, P.T. & Dingemanse, N.J. (2007)
How to cite this article: Dumont F, Solà Cassi M,
Integrating animal temperament within ecology and evolution.
Biological Reviews, 82, 291–318. Lemay M & Provost C (2025) Selection for zoophagy
Richardson, M.L., Mitchell, R.F., Reagel, P.F. & Hanks, L.M. (2010) Causes influences biocontrol efficacy and fruit damage by
and consequences of cannibalism in noncarnivorous insects. Dicyphus hesperus in greenhouses. Entomologia
Annual Review of Entomology, 55, 39–53. Experimentalis et Applicata 00: 1–10. https://doi.
Royer, P., Dumont, F., Provost, C. & Lucas, E. (2022) Selecting aggressive-
org/10.1111/eea.13570
ness to improve biological control agents efficiency. Journal of Pest
Science, 95, 1589–1596.