Content uploaded by Suzanne E Blatt
Author content
All content in this area was uploaded by Suzanne E Blatt on Sep 13, 2016
Content may be subject to copyright.
J. Acad. Entomol. Soc. : - ()
© A E S
IntroductIon
Pieris rapae (Linnaeus) (Lepidoptera: Pieridae) is the main lepidopteran pest aecting cabbage production on
Prince Edward Island (PEI) and organic producers have limited means to protect their crop. Other pests attacking
cabbage are the cabbage looper (Trichoplusia ni Hübner) (Lepidoptera: Noctuidae), diamondback moth, (Plutella
xylostella L.) (Lepidoptera: Plutellidae), thrips (rips tabaci Lindeman) (ysanoptera: ripidae) and cabbage
Use of Trichogramma and Bt for control of Pieris rapae in cabbage
on Prince Edward Island
Suzanne Blatt, Andrew Ryan, Shelley Adams, Je Franklin, and Joanne Driscoll
Suzanne Blatt and Je Franklin: Agriculture and Agri-Food Canada, 32 Main Street, Kentville, NS, Canada, B4N
1J5.
Andrew Ryan, Shelley Adams, and Joanne Driscoll: PEI Horticultural Association, 420 University Avenue,
Charlottetown, PE, Canada, C1A 7Z5.
Corresponding author (email suzanne.blatt@agr.gc.ca).
Received September . Accepted for publication December . Published on the Acadian Entomological Society website at www.acadianes.
ca/journal.php on December .
AbstrAct
Organic cabbage growers on Prince Edward Island have limited means to control Pieris rapae (Lepidoptera: Pieridae), their
main lepidopteran pest in cabbage. e use of Bt (Bacillus thuringiensis subsp. kurstaki) or spinosad products (e.g., Entrust)
are common practice but repeated applications can become expensive, while parasitoid wasps (Trichogramma brassicae
Bezdenko (Hymenoptera: Trichogrammatidae)) could provide an economically feasible alternative if shown to be eective.
Eight plots were established in an organic cabbage eld and encircled with a fence to reduce dispersion between treatments.
Four treatments (control, Bt, Bt+Trichogramma release, and Trichogramma release) were set up in these plots during the
growing season. e proportion of marketable cabbages was signicantly reduced in control plots compared with other
treatments. Further, Bt+Trichogramma treatment resulted in signicantly higher proportion of marketable cabbages than
Trichogramma alone. No signicant dierences were observed among treatments for cabbage head weight or size. Economics
and potential of Trichogramma in an integrated pest management strategy for organic cabbage production are discussed.
résumé
Les producteurs de choux organiques de l’Île-du-Prince-Édouard ont peu de moyens pour contrôler Pieris rapae
(Lepidoptera : Pieridae), leur principal lépidoptère ravageur du chou. L’utilisation de Bt (Bacillus thuringiensis ssp.
kurstaki) ou de produits à base de spinosad (par exemple, Entrust) est pratique courante, mais des applications répétées
peuvent devenir coûteuses, alors que des guêpes parasitoïdes (Trichogramma brassicae Bezdenko (Hymenoptera :
Trichogrammatidae)) pourraient orir une alternative économiquement viable si leur ecacité pouvait être démontrée.
Huit parcelles ont été établies dans un champ de choux organiques et furent entourées d’une clôture dans le but
de réduire la dispersion entre les traitements. Quatre traitements (contrôle, Bt, Bt + libération de Trichogramma,
et libération de Trichogramma) ont été mis en place dans ces parcelles pendant la saison de croissance . La
proportion des choux commercialisables a été signicativement réduite dans les parcelles témoins par comparaison
aux autres traitements. De plus, le traitement Bt + Trichogramma a produit une proportion signicativement
plus élevée de choux commercialisables que Trichogramma seul. Aucune diérence signicative n’a été observée
entre les traitements en lien avec le poids ou la taille des têtes de chou. Les aspects économiques et le potentiel de
Trichogramma dans une stratégie de gestion intégrée des ravageurs pour la production de chou biologique sont discutés.
© A E S
27 Blatt et al. / Journal of the Acadian Entomological Society 11 (2015): 26-31
ea beetle (Phyllotreta albionica LeConte) (Coleoptera:
Chrysomelidae), with Pieris rapae causing the most
damage. Without a control, of cabbage heads can
be rendered unmarketable due to lepidopteran feeding.
roughout this study, Trichoplusia ni and Plutella
xylostella were only encountered a couple times, therefore,
results reported here will focus on Pieris rapae only.
Application of Bt (Bacillus thuringiensis subsp. kurstaki)
(i.e., Dipel X DF), or products containing spinosad (i.e.,
Entrust) are the two main controls available to organic
producers. Dipel X (Bt) provides good control but does
not last long in the eld while Entrust provides longer
term benet but is more expensive to use. Some growers
use a combination of Entrust and Dipel X to protect their
crop. Eective and economic alternative control methods
are desired to ensure sustainability of production. While
Trichogramma wasps (Hymenoptera: Trichogrammatidae)
are an ecologically sound and sustainable biological control
method in many parts of the world, Trichogramma has not
been used by cabbage producers on PEI. Trichogramma
spp. have been eectively used for control of Helicoverpa
sp. (Lepidoptera: Noctuidae) in tomato, okra or chili
in India (Krishnamoorthy ) and against Erinnyis
ello (L.) (Lepidoptera: Sphingidae) in cassava in Brazil
(Soares et al. ). In Germany, Trichogramma brassicae
Bezdenko is used against European corn borer (Ostrinia
nubilalis Hübner) (Lepidoptera: Crambidae) on ,
hectares of corn; Trichogramma cacoeciae Marchal
and Trichogramma dendrolimi (Matsumura) are used
against codling moth (Cydia pomonella L.) (Lepidoptera:
Tortricidae) and plum moth (Grapholita funebrana Treits.)
(Lepidoptera: Tortricidae) in tree fruits; Trichogramma
evanescens Westwood is used against cabbage pests;
and, other Trichogramma species are used against vine
moths (Lobesia botrana (Den. & Schi.) and Eupoecilia
ambiguella (Hübner) (Lepidoptera: Tortricidae)) on
grapes (Zimmerman ). In China, nearly million
hectares of corn are managed for Asian corn borer
(Ostrinia furnacalis Guen.) (Lepidoptera: Crambidae)
using Trichogramma dendrolimi, Trichogramma chilonis
Ishii and Trichogramma ostriniae Pang and Chen (Wang
et al. ). In cabbage, Lundgren et al. () evaluated
Trichogramma brassicae in the USA and more recently,
Stoleru et al. () evaluated Bt and Trichogramma
evanescens in Romania. Results were promising but showed
variety and planting timing to inuence results. Integration
of Bt with Trichogramma was shown to be eective against
tomato fruitworm (Helicoverpa zea (Boddie) (Lepidoptera:
Noctuidae))and European corn borer, reducing damage
by and , respectively, over control plots (Oatman
et al. and Losey et al. ). Cabbage producers
on PEI were interested in evaluating Trichogramma as
a method to control lepidopteran pests. e objective
of this work was to evaluate the use of Trichogramma
brassicae through application in an organic cabbage eld
as follows: Trichogramma release+Bt, Trichogramma
release, and Bt alone for control of Pieris rapae.
methods
e .-ha eld site was located in Greenvale, Prince
Edward Island (°’.”N, °’.”W). Cabbage
plants, variety Lennox, were transplanted on July .
An application of Entrust W, Dow Chemical, (.
g/ha) was applied to the eld on August to remove any
Pieris rapae larvae prior to the study. Eight treatment plots
were established on August running across the width of
the eld, each plot measuring . m x . m containing
rows of cabbage (Figure ). Rows were spaced ca. cm
apart and cabbages spaced cm apart within the rows.
Cabbage plants, /plot, were monitored for Pieris rapae
throughout the season; they were agged on August and
cleaned of any larvae. e agged cabbages per plot
were selected as follows: cabbages per row from rows
-, with cabbages located between agged cabbages
within the row. Flagged cabbages were surveyed for eggs
and larvae twice per week until harvest on October.
Eggs were counted and circled during each survey, not
removed between surveys and not counted if empty.
Larvae were counted but not removed between survey
dates. ere were replicates of each treatment: control
(no Bt, no Trichogramma), Bt only, Bt + Trichogramma,
and Trichogramma only. Treatments were arranged
to reduce interference from other treatments, reduce
contamination of the control and Bt plots, and for ease
of Bt application for the grower. As such, the control plots
were on either edge of the eld, with treatment plots in
the following order moving into the eld: Bt only, Bt +
Trichogramma, and Trichogramma only (centre of the eld,
see Figure ). e enclosures around each plot were ca.
cm high and consisted of Tyvek house wrapping stapled to
cm x cm posts spaced ca. cm apart. Each plot was
fully enclosed and separated from the others using this
fencing. e fences were installed to reduce movement
of Trichogramma between plots as PEI can be windy and
Trichogramma are known to be aected by strong winds
(Yu et al. ; Fournier and Boivin ; Lundgren et al.
). Trichogramma cards (Tricho-Gard) were obtained
from Anatis BioProtection ( rang Saint-André, Saint-
© A E S
Blatt et al. / Journal of the Acadian Entomological Society 11 (2015): 26-31 28
Jacques-le-Mineur, Quebec), placed into plots starting
August and replaced every - days until September.
Each Trichogramma and Trichogramma + Bt plot received
card based on the recommendation of Anatis BioProtection
where card is used for m and each card contained
Trichogramma brassicae eggs. Dipel X DF, Valent
Canada, was applied to Bt and Bt + Trichogramma plots
at a rate of g/ha on August, September, and
September. At harvest, all cabbage heads were visually
inspected for damage and qualitatively categorized as
either marketable or non-marketable. Cabbages evaluated
as non-marketable displayed obvious feeding damage from
Pieris rapae. Marketable heads were weighed and measured.
All analyses were performed using R statistical soware
(R Core Team, ). Given the non-randomized layout of
the treatment plots within the eld, analysis was conducted
to determine if there was a pre-treatment bias in egg
counts across the plots. We wanted to establish whether
adult butteries had equal access to all cabbages within the
study, or if the enclosures aected their access to parts of
the eld. As the same cabbages were surveyed repeatedly
over time, egg counts were treated as a repeated measures,
Poisson-distributed time series variable analysed using a
generalized linear mixed model function (GLMM) in R
(glmer) with sampling date and plot as random variables
and with treatment as a xed variable. e treatment
chosen to be the comparative or base case was the control.
e model was restricted to the rst sampling dates (
– August) before the rst application of Bt on August.
Pairwise comparisons of the coecients were carried out
using Tukey’s HSD (multcomp package) with α=..
Coecient estimates from the analysis were exponentiated
to give estimates of the mean densities per treatment.
Larval counts were treated the same as egg counts but
with no restriction on sampling dates. Results of the egg
count analysis showed a statistically signicant dierence
between the control plot egg counts compared to the other
treatments (Table ). Closer inspection of the data revealed
that of the outliers occurred in the rst rows of plot
‘control ’. To oset the eects of the bias in the control
plots, the larval analysis was carried out using egg counts as
a covariate. Dispersion values for the egg and larval models
were . and ., respectively, so we considered that no
correction for over dispersion was required. Marketable
and unmarketable cabbage heads were pooled across plots
and analysed using a x contingency analysis on the
proportions followed by Fisher’s pairwise comparison
with Bonferroni adjustment. For the marketable cabbages,
head weight and size were compared between treatments
using an ANOVA. e economics of using Trichogramma
was evaluated by comparing the cost per hectare when
using Trichogramma cards compared with the current
controls of Entrust or Dipel X. Due to the bias in the
pre-treatment egg counts, the authors recognize that
the following results should be interpreted with caution.
results
Adult Pieris rapae were observed ying throughout all plots
over the course of the study. Pieris rapae eggs and larvae
were surveyed from August through October inclusive
(Figures and ). e pattern of oviposition during the
rst survey dates (prior to Dipel X application) showed
a bias towards the edge plots (controls) (Figure and Table
). Eggs were signicantly more abundant in the control
plot than in either the Bt+Trichogramma or Trichogramma
alone plots (GLMM, Table ). e number of eggs in the Bt
alone plot was not signicantly dierent from the control
plot or either of the other two treatments. Similar results
were obtained for larval counts; signicantly higher counts
were observed in the control plot than the other treatment
plots. e lowest number of larvae were counted in the
Bt+Trichogramma treatment (GLMM, Table ). For both
egg and larval counts, numbers tapered o starting in
early September with minimal eggs and larvae found on
the cabbages by early October. At harvest, percentage of
Figure 1: Layout of the plot showing treatments and placement of
the sampled cabbages (1-20). Although shown only in the Bt 2 plot,
the same cabbage placement was used in each treatment plot.
Control 1
Bt 1
Bt + Release 1
Release 1
Release 2
Bt + Release 2
Bt 2
Control 2
1
4
3
2
9
12
11
10
8
5
6
7
17
20
19
18
16
13
14
15
N
© A E S
29 Blatt et al. / Journal of the Acadian Entomological Society 11 (2015): 26-31
marketable heads was lowest and signicantly so for the
control plots compared with all other treatment plots
(Figure ). e highest percentage of marketable heads was
obtained from the Bt+Trichogramma plots, signicantly
greater than the control and the Trichogramma alone
plots. Use of Trichogramma alone was not signicantly
dierent from using Bt alone. ere was no signicant
impact of any treatment on head weight or size (Table
). It is worth noting that there was only marketable
cabbage from the control plots and it was approximately
of the weight of cabbages from the other plots.
dIscussIon
Our results are consistent with an earlier study by
Lundgren et al. () who found the use of Trichogramma
to be less eective than Bt in controlling lepidopterans.
Trichogramma was released into those plots when, on
average, Pieris rapae larvae per cabbage head were
present. Our study introduced Trichogramma when
there were less than larvae per cabbage head present.
Figure : Mean (± SE) number of Pieris rapae eggs found on
cabbage heads over the growing season in . Arrows denote
application of Dipel X (Bt) on Aug, Sept, and Sept. *
denotes rst introduction of Trichogramma cards.
0
2
4
6
Mean (±SE) number of eggs
Survey date
Control
Bt
Bt+Trichogramma
Trichogramma
*
Figure : Mean (± SE) number of Pieris rapae larvae found
on cabbage heads over the growing in . Arrows indicate
application of Dipel X (Bt) on Aug, Sep and Sep. * denotes
rst introduction of Trichogramma cards.
0
2
4
6
Mean (±SE) number of larvae
Survey date
Control
Bt
Bt+Trichogramma
Trichogramma
*
Table 1: Regression table of xed eect coecients from generalized linear mixed
model analysis of Pieris rapae eggs on cabbages grown on PEI within plots for the
rst six survey dates. Random eects variance (standard deviation): Date – 0.16
(0.39), Location – 0.71 (0.84).
Table 2: Regression table of xed eect coecients from general linear mixed
model analysis of Pieris rapae larval counts with egg as a covariate on cabbages
grown on PEI. Random eects variance (standard deviation): Date – 1.28 (1.13),
Location – 0.11 (0.35), Eggs – 0.002 (0.013).
Figure : Percentage of unmarketable and marketable cabbage
heads in plots treated with Trichogramma or Bt or both during
in an organic cabbage eld. The proportion of marketable
heads was signicantly dierent among treatments (χ=.,
P<.). Bars with the same letter not signicantly dierent
(Fishers’s pairwise test using Bonferroni adjustment).
0
20
40
60
80
100
Control Release Bt Bt+Release
Percentage unmarketable and marketable cabbage heads
Treatment
Unmarketable
Marketable
a
b
bc
c
© A E S
Blatt et al. / Journal of the Acadian Entomological Society 11 (2015): 26-31 30
Lundgren used cards containing ~ , parasitized
eggs in their m x m plots, while this study had cards
containing ~, eggs per m x m plot. While
these dierences may be subtle, they may have been
signicant enough to cause the dierence in results
obtained. Lundgren et al. () found larval levels in
Trichogramma plots to equal those found in the control,
which is counter to our results where Bt+Trichogramma
did reduce larval populations when compared with
the control. Lundgren also showed no increase in
marketable cabbage heads with the use of Trichogramma.
Stoleru et al. () evaluated Bt and Trichogramma
evanescens in organic cabbage elds in Romania. e mid-
season varieties yielded high harvest weight when treated
with Trichogramma evanescens at , individuals/
ha while the late season varieties showed no benet of
Trichogramma evanescens when compared with the other
treatments. Given that Lennox is a late season variety,
and PEI has a short growing season, it is possible that
Trichogramma experienced less optimal conditions for
foraging in the eld toward the end of the eld season.
Trichogramma spp. require appropriate temperature
and plant structure to provide optimal control. Romário
de Carvalho et al. () found temperature to play an
important role in the foraging behavior of Trichogramma
pretiosum Riley and Pak and vanHeiningen () found
eld temperature to inuence behavior in Trichogramma
spp. with Trichogramma pretiosum performing best at °C
- °C. At °C or °C, parasitism did occur, but at lower
rates than parasitism occurring at higher temperatures. In
the PEI eld during , mean daily temperatures were
above °C up to September. Aer this point, the average
daily temperature dropped to between °C and °C with
only days reaching an average temperature of °C. is
could have inuenced the hatching rate and subsequent
foraging behavior of Trichogramma during the latter part
of this study. During the study done by Lundgren et al.
(), the mean daily temperature was approximately
°C, and similarly in Romania, Stoleru et al. () had
temperatures above °C for the duration of the mid-
season crop but this dropped toward the end of their study.
As our study utilized a late season cabbage variety, further
study could utilize short season varieties. ese may benet
from using Trichogramma as exposure time would be
reduced and daily temperatures would be more suitable for
Trichogramma foraging throughout the growing period.
ere was some concern that the architecture and
spacing of the cabbage plants would negatively inuence
Trichogramma distribution within the plot. Cabbages
were planted with a spacing of cm in early July. By
early August, when Trichogramma was released, the plants
were considerably larger and had leaves well into the gap
between plants. Although the canopy was not dense, it
is expected that Trichogramma would have been able
to move between plants within a plot. As other studies
using Trichogramma in cabbage (Stoleru et al. )
reported success, it is more likely that length of season and
temperature contributed to the results obtained in our study.
Economically, the use of Trichogramma alone is a
moderately expensive management option (Table ).
Dipel X is the least expensive option costing ./ha
when used three times during the season. Addition of
Trichogramma, as used in this study (three applications
of Dipel X and ve sets of cards), increases the cost to
./ha, with no signicant increase in marketable
cabbage heads. ese options are less expensive than
using Entrust, a spinosad product, on its own (./ha).
Another option used by growers is to alternate Entrust and
Dipel X with two applications each, costing ./ha.
Clearly Dipel X is the least expensive option, however,
there is denitely room to improve upon the percentage of
marketable heads at harvest. Should growers wish to use
Trichogramma for control of Pieris rapae, further study
of Trichogramma on PEI, incorporating shorter season
varieties, is required before its use becomes economically
feasible. A short season variety, e.g., one that needs
only - days to reach maturity, would mean fewer
cards to be deployed and, if planted early in the season,
would ensure that hatching Trichogramma are exposed
to adequate temperatures for foraging. Reducing the
number of sets of cards from to or would make use of
Trichogramma either alone or in combination with Dipel
X more cost eective method for control of Pieris rapae.
is study found the best treatment for control of
Pieris rapae in organic cabbage to be Bt+Trichogramma,
however these results must be taken with caution due
to the lack of randomization in the plot layout. Further
study using a randomized plot design could conrm the
Table 3: Mean head weight and dimensions (polar and equator) of marketable
cabbage heads. There were no signicant dierences between treatments
(F
3,47
=0.79, P=0.50, F
3,48
=0.79, P=0.50, F
3,48
=0.45, P=0.72 for head weight,
polar and equator circumference, respectively).
© A E S
31 Blatt et al. / Journal of the Acadian Entomological Society 11 (2015): 26-31
results obtained here. Plot layout may have inuenced
oviposition, and further study is warranted to conrm
the results obtained from this study. is study combined
Bt sprays (Dipel X) with Trichogramma release, as
growers were interested in knowing if combining the two
treatments would yield an additive eect and increase
percentage of marketable heads at harvest. Results from
this study show that a combination of Bt+Trichogramma
will lead to a signicant increase in marketable cabbage
heads over the controls and Trichogramma alone, however,
the increase is not large enough to oset the cost of both
products. Until further trials have conrmed the potential
of Trichogramma for control of Pieris rapae, organic
cabbage growers on PEI should continue to use Dipel X.
Acknowledgements
e authors are indebted to Red Soil Organics, for allowing
this trial in their commercial eld and Christine Noronha
(AAFC-Charlottetown), Deb Moreau (AAFC-Kentville)
and two anonymous reviewers for thoughtful comments
on the manuscript. Funding for this work was awarded to
JD from a Growing Forward : Agriculture Research and
Innovation Program, Applied Research Subprogram grant.
references
Fournier, F., and Boivin, G. . Comparative dispersal of
Trichogramma evanescens and Trichogramma pretiosum
(Hymenoptera: Trichogrammatidae) in relation to
environmental conditions. Environmental Entomology
29: -.
Krishnamoorthy, A. . Exploitation of egg parasitoids
for control of potential pests in vegetable ecosystems in
India. Comunicata Scientiae 3: -.
Losey, J.E., Fleischer, S.J., Calvin, D.D., Harkness, W.L., and
Leahy, T. . Evaluation of Trichogramma nubialis and
Bacillus thuringiensis in management of Ostrinia nubialis
(Lepidoptera: Pyralidae) in sweet corn. Environmental
Entomology 24: -.
Lundgren, J.G., Heimpel, G.E., and Bomgren, S.A. .
Comparison of Trichogramma brassicae (Hymenoptera:
Trichogrammatidae) augmentation with organic
and synthetic pesticides for control of cruciferous
Lepidoptera. Environmental Entomology 31: -
.
Oatman, E.R., Wyman, J.A., VanSteenwyk, R.A., and
Johnson, M.W. . Integrated control of the tomato
fruitworm (Lepidoptera: Noctuidae) and other
lepidopterous pests on fresh-market tomatoes in
southern California. Journal of Economic Entomology
76: -.
Pak, G.A., and vanHeiningen, T.G. . Behavioral
variations among strains of Trichogramma spp.:
adaptability to field temperature conditions.
Entomological Experimentalis et Applicata 38: -.
R Core Team (). R: A language and environment
for statistical computing. R Foundation for Statistical
Computing, Vienna, Austria. ISBN ---, URL
http://www.R-project.org/.
Romário de Carvalho, J, Pratissoli, D, Dalvi, L.P., Silva,
M.A., Bueno, R.C.O.dF, and Bueno, A.dF. .
Parasitism capacity of Trichogramma pretiosum on
eggs of Trichoplusia ni at dierent temperatures. Acta
Scientiarum Agronomy 36: -.
Soares, M.A., Leite, G.L.D., Zanuncio, J.C, Ferreira,
C.S., Rocha, S.L. and Mendes de Sá, V.G. .
Assessmment of Trichogramma species (Hymenoptera:
Trichogrammatidae) for biological control in cassava
(Manihot esculenta Crantz). Acta Scientiarum Agronomy
36: -.
Stoleru, V.V., Munteanu, N.C., Stoleru, C.M.V. and Rotaru,
L.G. . Cultivar selection and pest control techniques
on organic white cabbage yield. Notulae Botanicae Horti
Agrobotanici 40: -.
Wang, Z-Y, He, K-L, Zhang, F, Lu, X and Babendreier,
D. . Mass rearing and release of Trichogramma
for biological control of insect pests of corn in China.
Biological Control 68: -.
Yu, D.S.K., Laing, J.E. and Hagley, E.A.C. . Dispersal of
Trichogramma spp. (Hymenoptera: Trichogrammatidae)
in an apple orchard aer inundative releases.
Environmental Entomology 13: -.
Zimmerman, O. . Der einsatz von Trichogramma-
schlupfwespen in Deutschland. Gesunde Panzen 56:
-.
Table 4: Economic comparison of control methods for Pieris rapae in organic
cabbage production on PEI.
© A E S
Blatt et al. / Journal of the Acadian Entomological Society 11 (2015): 26-31 32
