170
Research Article
Received: 5 April 2010 Revised: 30 July 2010 Accepted: 19 August 2010 Published online in Wiley Online Library: 27 October 2010
(wileyonlinelibrary.com) DOI 10.1002/ps.2047
Lepidopteran larva consumption of soybean
foliage: basis for developing multiple-species
economic thresholds for pest management
decisions
Regiane Cristina Oliveira de Freitas Bueno,
a
Adeney de Freitas Bueno,
b∗
Fl
´
avio Moscardi,
c
Jos
´
e Roberto Postali Parra
d
and Clara Beatriz Hoffmann-Campo
b
Abstract
BACKGROUND: Defoliation by Anticarsia gemmatalis (H
¨
ubner), Pseudoplusia includens (Walker), Spodoptera eridania (Cramer),
S. cosmioides (Walker) and S. frugiperda (JE Smith) (Lepidoptera: Noctuidae) was evaluated in four soybean genotypes. A
multiple-species economic threshold (ET), based upon the species’ feeding capacity, is proposed with the aim of improving
growers’ management decisions on when to initiate control measures for the species complex.
RESULTS: Consumption by A. gemmatalis, S. cosmioides or S. eridania on different genotypes was similar. The highest
consumption of P. includens was 92.7 cm
2
on Codetec 219RR; that of S. frugiperda was 118 cm
2
on Codetec 219RR and 115.1 cm
2
on MSoy 8787RR. The insect injury equivalent for S. cosmoides, calculated on the basis of insect consumption, was double the
standard consumption by A. gemmatalis, and statistically different from the other species tested, which were similar to each
other.
CONCLUSIONS: As S. cosmioides always defoliated nearly twice the leaf area of the other species, the injury equivalent would
be 2 for this lepidopteran species and 1 for the other species. The recommended multiple-species ET to trigger the beginning
of insect control would then be 20 insect equivalents per linear metre.
c
2010 Society of Chemical I ndustry
Keywords: economic injury level; Pseudoplusia includens; Anticarsia gemmatalis; Spodoptera; insect equivalent
1INTRODUCTION
Integrated pest management (IPM) is based on the principle that
insecticides must not be applied before economic damage has
occurred. Consequently, to enhance growers’ decision-making
process, the economic injury level (EIL) concept was developed
as the lowest population density capable of causing economic
damage to plants.
1
Similarly, the economic threshold (ET) was
developed as the population density at which control measures
should be started to prevent an increasing pest population from
reaching the EIL.
2
Accurate EILs and ETs are fundamental to
increase yield and maintain environmental quality by reducing
the unnecessary use of management tactics, especially pesticides.
3
Pursuing this goal has been a challenge to entomologists because
there is a lack of information on this subject regarding different
pest species and crops.
4
Currently, the recommended ETs for lepidopteran larvae on
soybean [Glycine max (L.) Merrill] are slightly different around the
world. In Brazil, pest control measures are initiated either when 20
large (≥1.5 cm) lepidopteran larvae are counted per sample cloth
(1 m soybean line) or when 30% defoliation (in the vegetative
stage) or 15% defoliation (in the reproductive state) is observed.
5
In the United States, it is stated that soybean plants can withstand
as much as 35% defoliation up to the blooming period. However,
during this stage, and when pods begin to form and fill out, any
foliage loss greater than 20% will decrease yield.
6
The tolerance of
soybeans for these defoliation levels depends on the cultivar, and
such differences are very important for IPM.
7
The ET based on the number of larvae was originally proposed
for the velvetbean caterpillar, Anticarsia gemmatalis (H
¨
ubner),
which at the time was the most important defoliator insect
∗
Correspondence to: Adeney de Freitas Bueno, Embrapa Soybean, PO Box 231,
Londrina, Paran
´
a 86001-970, Brazil. E-mail: adeney@cnpso.embrapa.br
a University of Rio Verde (FESURV), Rio Verde, Goi
´
as, Brazil
b Embrapa Soybean, Londrina, Paran
´
a, Brazil
c Centro de Ci
ˆ
encias Agr
´
arias, State University of Londrina, UEL, Londrina,
Paran
´
a, Brazil
d Entomology, Phytopathology and Zoology Department, University of S
˜
ao
Paulo- ESALQ/USP – Piracicaba, S
˜
ao Paulo, Brazil
Pest Manag Sci 2011; 67: 170–174 www.soci.org
c
2010 Society of Chemical Industry
171
Lepidopteran larva consumption of soybean foliage www.soci.org
occurring on soybean crops from Argentina to the south-eastern
United States.
8,9
However, nowadays this is not necessarily
true in either South America or the south-eastern United
States.
10,11
The occurrence of another insect, the soybean looper
[Pseudoplusia includens (Walker)] has increased in soybean fields
as a consequence of indiscriminate use of insecticides.
11
Similarly,
species of the genus Spodoptera, such as the southern armyworm,
S. eridania (Cramer), as well as S. cosmioides (Walker), are now
considered key pests by some Brazilian soybean growers, for
example in the states of Goi
´
as and Mato Grosso.
12
Considering
that the capacity for damage of Spodoptera spp. and P. includens
might differ from that of A. gemmatalis,
13
an ET based on 20
lepidopteran larvae per sample cloth might not be accurate.
A possible refinement to the EIL concept is to consider injury
in standard units, termed injury equivalents.
14
An insect injury
equivalent would be the total injury produced by a single pest
over its lifespan. It has been proposed to use injury equivalents for
making management decisions for a pest guide, such as a complex
of defoliators.
14
These authors considered the consumption of five
soybean-defoliating insects and established injury equivalents to
build a multispecies EIL.
The present study compared the foliage consumption of the
important soybean pests P. includens, S. cosmiodes, S. eridania
and S. frugiperda, in an attempt to establish injury equivalents
and multispecies ETs to be used for management decisions. The
defoliation caused by different lepidopterans was also compared
in foursoybeancultivars, to evaluatepossibledifferences incultivar
tolerance.
2 MATERIALS AND M ETHODS
The experiment was carried out in a factorial randomised block
design 4×5 (four soybean genotypes ×five lepidopteran species)
with four replicates, each containing ten lepidopteran larvae. The
insects used in the trial were obtained from laboratory colonies
where they had been reared for 4–6 generations following their
collection in the soybean fields. Lepidopteran larvae to begin
laboratory insect colonies were manually collected during the
crop season of 2006/2007 (from December 2006 to January 2007)
at Rio Verde, State of Goias, Brazil. They were transferred to
the laboratory and reared on an artificial diet.
15,16
The soybean
leaflets used in the trials were excised from plants grown in field
conditions.
2.1 Insect colonies
Colonies of P. includens, A. gemmatalis, S. comioides, S. frugiperda
and Seridaniawere kept in the laboratory under controlled
environmental conditions (25 ± 2
◦
C temperature, 70 ± 10% RH,
14 : 10 h light : dark photoperiod). The insect larvae were reared on
an artificial diet,
15,16
and, after emergence, the adults were fed on
a 10% honey/water solution in cages (10 cm diameter × 21.5 cm
high). The walls of the cages were covered with A4 paper, used as
a substrate for eggs. These eggs were removed daily and used for
maintaining the insect colonies.
2.2 Soybean plots
The soybean genotypes studied, Conquista (maturity group 8.2),
Codetec 219RR (maturity group 8.2), Monsoy 6101 (maturity
group 6.1) and Monsoy 8787RR (maturity group 8.7), are the
most important cultivars sown in the different Brazilian regions.
These genotypes were sowed at Embrapa, Goi
ˆ
ania, Brazil. Four
plots (5 × 5 m) of each soybean cultivar were manually sowed
in rows spaced 0.5 m apart. The soil is a clayey latosol Typic
Haplustox.
17
Plots with different soybean cultivar were spaced
2 m from each other and fertilised with 20 kg P
2
O
5
ha
−1
and
20 kg K
2
Oha
−1
.
18
During the entire plant cycle, no pesticides were
used, to avoid influencing the insects’ biological features. Weeds
were manually removed from the area, and insects and diseases
occurred homogeneously in all four cultivars and at low levels.
2.3 Consumption experiment
The experiment started with first-instar larvae and soybean
cultivars at minimum in the R
2
growing stage,
19
with all the
cultivars evaluated at the same time. Individual leaflets were
collected daily from the sunny sides of the four soybean
cultivar plots, and were used to feed the lepidopteran larvae.
To standardise the age of the leaflets, only the newest fully
expanded leaves were collected at the top of each plant canopy.
Leaflets were kept fresh in closed plastic bags and used within
2 h after excision. Single leaflets were placed on double sheets of
moist filter paper in 2.5 × 11 cm diameter plastic petri dishes.
20
Leaflets were replaced daily to avoid excessive dehydration. The
insects were maintained in an environmental chamber (25 ± 2
◦
C
temperature, 70 ±10% RH, 14 : 10 h light : dark photoperiod) up to
pupation, which took different times according to the life cycle of
each tested species. The sexes of the larvae used in the trial were
not taken into consideration.
The soybean foliage area (cm
2
) was determined in a leaf area
meter (Model LI-3100; Li-Cor, Lincoln, NE) before and after larval
feeding. The daily foliage consumption by each species was then
calculated by subtraction. A control leaflet was used for each
genotype to estimate leaf dehydration and consequent reduction
in leaflet size. This was performed during the entire period of
evaluation, by measuring the leaf area (cm
2
) of the control leaflets,
and thenusing these changes toadjustthe daily larvaconsumption
results.
The total consumption by each individual larva was recorded for
each species, and the average consumption of each replicate was
used in the analysis. The insect injury equivalent was calculated by
dividing the foliage consumption of each replicate by the average
consumption of A. gemmatalis on each soybean genotype, which
was taken as the standard consumption.
2.4 Data analysis
Data were analysed for the assumption of normality
21
and ho-
mogeneity of treatment variance in all parameters.
5
Lepidopteran
larva consumption (cm
2
) and insect injury equivalent data did not
follow normality or homogeneity of variance, and needed to be
transformed into
√
X to perform the analysis of variance (ANOVA).
Then, the means were compared using Tukey’s Studentised range
test at 5% probability for statistical significance.
22
3 RESULTS AND DISCUSSION
Factorial analysis revealed an interaction between the pest species
and soybean cultivars for insect foliage consumption (cm
2
)
(P
pest×soybean
= 0.0158, df
residue
= 172, F = 33.46) (Table 1).
The defoliating capacity of P. includens and S. frugiperda varied
among thedifferentsoybeangenotypes tested (Table 1). Anticarsia
gemmatalis, S. cosmioides and S. eridania consumed all the
genotypes similarly. However, P. includens foliage consumption
was higher for Codetec 219RR, and S. frugiperda consumed larger
Pest Manag Sci 2011; 67: 170–174
c
2010 Society of Chemical Industry wileyonlinelibrary.com/journal/ps
172
www.soci.org RCO de Freitas Bueno et al.
Table 1. Consumption (cm
2
) by lepidopteran larvae on different soybean cultivars
a
Soybean genotype
Pest species Codetec 219RR MSoy 6101 MSoy 8787RR Conquista
Anticarsia gemmatalis 92.6 (±4.5) bA 74.2 (±4.2) bcA 94.9 (±6.3) cA 90.0 (±3.4) bA
Pseudoplusia includens 92.7 (±4.5) bA 63.9 (±7.4) cB 63.9 (±7.5) aB 64.0 (±4.5) cB
Spodoptera cosmioides 183.6 (±14.9) aA 184.8 (±8.9) aA 185.4 (±5.4) aA 175.1 (±5.7) aA
Spodoptera eridania 107.2 (±6.9) bA 98.3 (±11.0) bA 101.9 (±5.0) bA 86.9 (±7.2) bA
Spodoptera frugiperda 118.0 (±6.4) bA 90.0 (±9.3) bB 115.1 (±5.7) bA 95.4 (±8.2) bAB
CV (%) 10.4
a
Means followed by similar upper-case letters in the row and similar lower-case letters in the column are not statistically different using Tukey’s
Studentised range test at 5 per cent probability. Original data followed by statistics performed on data transformed in
√
X.
leaf areas of Codetec 219RR and MSoy 8787RR compared with
MSoy 6101. This indicates that the soybean genotypes have
different levels of resistance to outbreaks of lepidopteran larvae.
MSoy 6161 may be more resistant to both P. includens and S.
frugiperda, as it was consumed less by both species (Table 1).
However, types and mechanisms of resistance need to be further
studied in field and laboratory trials. In addition to being attacked
by insects to a greater or lesser degree, soybean cultivars can
also show different tolerances to various defoliation levels, and
this information is crucial for crop protection,
7
considering that
plant resistance or tolerance is environmentally safe and generally
compatible with other IPM strategies.
23
Analysis of soybean leaf consumption data for the different
pest species clearly shows the damage potential of Spodoptera
spp. The feeding capacity of S. cosmioides was nearly double
the leaf area consumed by the other lepidopteran species
(P
pest
< 0.0001, df
pest
= 4, df
residue
= 172, F = 33.46). With the
results obtainedhere, showing that lepidopteranlarvae of different
species consume different levels of foliage, it can be assumed that
soybean plants cannot withstand the same level of infestation
by all lepidopteran species. Variable defoliating capacity among
larvae was previously reported for soybean lepidopterans:
13
Spodopteraexigua (H
¨
ubner) (Lepidoptera: Noctuidae) was theleast
damaging species, consuming 52 cm
2
leaf larva
−1
compared with
Trichoplusia ni (H
¨
ubner) (Lepidoptera: Noctuidae), P. includens,
Heliothis virescens F. (Lepidoptera: Noctuidae), Helicoverpa zea
(Boddie) (Lepidoptera: Noctuidae) and A. gemmatalis, which
consumed 119, 114, 155, 336 and 84 cm
2
soybean leaf larva
−1
respectively. This differs from the present results, where the
studied Spodoptera species (S.frugiperda,S.cosmioidesand S.
eridania) showed the same or even higher defoliating capacity
than A. gemmatalis and P. includens (Table 1).
Safe ETs should be below EILs, and are calculated using the
formula
2
EIL = C/VID, where C is the cost of management activity
per production unit, V is the market value per production unit,
I is injury units per insect per production unit and D is damage
per injury unit.
2
As this formula includes the necessary injury
and damage information to calculate accurate EILs, the ET can be
adjustedformultiple-species outbreaksofsoybean defoliators.The
central requirement for developing a multiple-species EIL is that
injuries by different species of interest produce a homogeneous
physiological response in the plant. Species producing a common
injury such as defoliation, for example, constitute an injury guild.
14
When proposing this adjustment for this guild, it was assumed
that C would be the same for all lepidopteran larva species, as
the same insecticides could be recommended for their control.
Given that V is the market value, this was also considered to be the
same. The ID, usually calculated together,
2
was the only variable
suggested to be changed among the lepidopteran species. In this
way, injury by any individual species can be related to injury by
another species in the same guild.
7,24
Because the ID of S. cosmioides was double that of the other
species, it is appropriate to assume that the EIL, and consequently
the related ET, will be half that of the other species studied.
Therefore, because S. cosmioides caused a higher injury level
to soybean plants (P
pest
< 0.0001, df
pest
= 4, df
residue
= 172,
F = 33.46) compared with the other species, this demonstrates
that the ET for multiple species is feasible and would represent an
improvement in precision for the grower’s management decisions.
A.gemmatalis,P.includens,S.eridaniaand S. frugiperda consumed
roughly the same leaf area, and therefore showed a similar
potential for injury to soybean plants (Table 1).
Considering that these lepidopteran species usually occur
together in soybean fields, an insect injury equivalent
24
based
on the different feeding capacity of each pest species might
be developed. In this context, the insect injury equivalent was
calculated. Taking into accountitsconsumption,A. gemmatalis was
chosen as the standard equivalent species, and the S. cosmioides
insect injury equivalent was statistically different from the others
and close to double that of A. gemmatalis (P
pest
< 0.0001,
df
pest
= 4, df
residue
= 172, F = 33.96) (Table 2). This species was
used to calculate the economic threshold (ET) presently in use of
20 lepidopteran larvae m
−1
.
18
In conclusion, the injury equivalent
must be 2 for S. cosmioides and 1 for all other species tested.
Furthermore, the recommended ET to trigger insect control would
be 20 insect equivalents per sample cloth (1 m soybean line),
similar to the level proposed for other soybean defoliators.
7,24
An injury equivalency system as proposed here can sometimes
be erratic, mainly because, if insect densities are high, competition
can reduce injury rates per individual. This issue should be
examined in future studies on ET of soybean pests, focusing
on the necessary changes in injury equivalence. This might be
achieved by an appropriate adjustment based on the density
injury per individual function, easily developed as part of an
interactive computer implementation of the multiple-species ET
model.
25
Because the present trials used field-grown leaves, the
calculated multiple-species ET presented here is more precise than
greenhouse trials.
25
It is alsoimportantto consider that defoliation-
induced yield losses correspond more closely to the percentage of
light interception reductions than to the percentage of leaf area
reductions.
26
Thus, ET might vary not only among lepidopteran
species but also among soybean cultivars.
7
wileyonlinelibrary.com/journal/ps
c
2010 Society of Chemical Industry Pest Manag Sci 2011; 67: 170–174
173
Lepidopteran larva consumption of soybean foliage www.soci.org
Table 2. Insect injury equivalent calculated for five species of lepidopteran larvae feeding on four soybean cultivars
a
Soybean genotype
Pest species Codetec 219RR MSoy 6101 MSoy 8787RR Conquista
Anticarsia gemmatalis 1.00 (±0.05) bA 1.00 (±0.06) bA 1.00 (±0.07) bA 1.00 (±0.04) bA
Pseudoplusia includens 1.00 (±0.05) bA 0.86 (±0.10) bA 0.86 (±0.11) bA 0.71 (±0.05) bA
Spodoptera cosmioides 1.98 (±0.16) aB 2.49 (±0.12) aA 1.95 (±0.06) aB 1.94 (±0.06) aB
Spodoptera eridania 1.16 (±0.07) bA 1.32 (±0.15) bA 1.07 (±0.05) bA 0.97 (±0.08) bA
Spodoptera frugiperda 1.27 (±0.07) bA 1.21 (±0.13) bA 1.21 (±0.06) bA 1.06 (±0.09) bA
CV (%) 10.6
a
Means followed by similar upper-case letters in the row and similar lower-case letters in the column are not statistically different using Tukey’s
Studentised range test at 5 per cent probability. Original data followed by statistics performed on data transformed to
√
X.
Spodoptera spp. can also damage pods. In this situation, the
usual ET of 10% damaged pods
18
must remain unchanged, also to
be used to initiate the control measure to prevent an increasing
pest population
3
either at 10% damaged pods or when 20 insect
equivalents are found on the sample cloth (1 m soybean line).
The importance of the improvement in the ET, as proposed
here, needs to be tested in field conditions to evaluate how much
this might increase soybean economic profits. Crop losses due
to arthropods, even if other management tactics are adopted,
would be massive without insecticides.
27
Moreover, elimination
of pesticide use would increase the necessary land area to
produce food, leading to destruction of natural habitat and
potentially depleting soil, water and energy resources.
23
The
economic sustainability of the agricultural production system
would also be affected.
28
However, pesticides mustonlybe applied
when necessary, to avoid problems related to indiscriminate
use of chemicals.
12
In conclusion, a precise ET provided by the
refinement proposed here is of theoretical and practical interest
for soybean growers, allowing them to combine economic profits
with ecological sustainability, as required by modern agricultural
practice.
ACKNOWLEDGEMENTS
The authors would like to thank Embrapa Soja and Embrapa Arroz
eFeij
˜
ao for the support provided. Thanks are also extended to
Coordenac¸
˜
ao de Aperfeic¸oamento de Pessoal de Nível Superior
(CAPES) for the supporting grant (process 23038.035744/2008-89),
and to Janet W Reid, JWR Associates, for the English revision. This
paper was approved for publication by the Editorial Board of
Embrapa Soja as manuscript 10/2010.
REFERENCES
1 Stern VM, Smith RF, Van Den Bosch R and Hagen KS, The integrated
control concept. Hilgardia 29:81–101 (1959).
2 Higley LG and Pedigo LP, The EIL concept, in Economic Thresholds
for Integrated Pest Management, ed. by Higley LG and Pedigo LP.
University of Nebraska Press, Lincoln, NE, pp. 9–21 (1996).
3 Pedigo LP and Higley LG, Introduction to pest management and
thresholds, in Economic Thresholds for Integrated Pest Management,
ed. by Higley LG and Pedigo LP. University of Nebraska Press,
Lincoln, NE, pp. 1–8 (1996).
4 Buntin GD, Economic thresholds for insect management, in Economic
Thresholds for Integrated Pest Management,ed.byHigleyLGand
Pedigo LP. University of Nebraska Press, Lincoln, NE, pp. 128–150
(1996).
5 Burr IW and Foster LA, A test for equality of variances. Mimeo Series
No. 282, University of Purdue, West Lafayette, IN, 26 pp. (1972).
6 Andrews G, Daves C, Koger T, Reed J, Burdine B, Dodds D, et al., Insect
control guides for cotton, soybeans, corn, grain sorghum, wheat,
sweet potatoes and pastures. Mississippi State University Extension
Service, Publication 2471, 64 pp. (2009).
7 Haile FJ, Higley LG and Specht JE, Soybean cultivars and insect
defoliation: yield loss and economic injury levels. Agron J
90:344–352 (1998).
8Hoffmann-CampoCB,OliveiraLJ,MoscardiF,GazzoniDL,Corr
ˆ
ea-
Ferreira BS, Lorini IA, et al., Integrated pest management in Brazil, in
Integrated Pest Management in the Global Arena, ed. by Maredia KM,
Dakouo D and Mota-Sanches D. CABI Publishing, Wallingford and
Cambridge, UK, pp. 285–299 (2003).
9 Panizzi AR and Corr
ˆ
ea-Ferreira BS, Dynamics in the insect fauna
adaptation to soybean in the tropics. Trends Entomol 1:71–88
(1997).
10 Hamadain EI and Pitre HN, Oviposition and larval behavior of
soybean looper, Pseudoplusia includens (Lepidoptera: Noctuidae),
on soybean with different row spacings and plant growth stages.
J A gric Urban Entomol 19:29–44 (2002).
11 Bueno RCOF, Parra JRP and Bueno AF, Biological characteristics and
thermal requirements of a Brazilian strain of the parasitoid
Trichogramma pretiosum reared on eggs of Pseudoplusia includens
and Anticarsia gemmatalis. Biol Cont 51:355–361 (2009).
12 Bueno RCOF, Parra JRP, Bueno AF, Moscardi F, Oliveira JRG and
Camillo MF, Sem barreira. Revista Cultivar, Pelotas 55:12–15 (2007).
13 Boldt PE, Biever KD and Ignoffo CM, Lepidopteran pest of soybean:
consumption of soybean foliage and pods and development time.
J Econ Entomol 68:480–482 (1975).
14 Hutchins SH, Higley LG and Pedigo LP, Injury equivalency as a basis
for developing multiple-species economical injury levels. JEcon
Entomol 81:1–8 (1988).
15 Greene GL, Leppla NC and Dickerson WA, Velvetbean caterpillar: a
rearing procedureand artificialmedium.JEconEntomol 69:487–488
(1976).
16 Parra JRP, T
´
ecnicas de Criac¸
˜
ao de Insetos para Programas de Controle
Biol
´
ogico. FEALQ, Piracicaba, S
˜
ao Paulo, Brazil, 134 pp. (2001).
17 Keys to Soil Taxonomy, 7th edition. USDA, Washington, DC, 644 pp.
(1996).
18 Tecnologia de produc¸
˜
ao desoja: regi
˜
ao central do Brasil, Sistemas de
produc¸
˜
ao. Embrapa Soja, Vol. 12. Empresa Brasileira de Pesquisa
Agropecu
´
aria (Embrapa), Brazil, 280 pp. (2009).
19 Fehr WR, Caviness RE, Burmood DT and Penninton JS, Stage of de-
velopment descriptions for soybeans, Glycine max L. Merrill. Crop
Sci 11:929–931 (1971).
20 Reid JC and Greene GL, The soybean looper: pupal weight, devel-
opment time, and consumption of soybean foliage. Fla Entomol
56:203–206 (1973).
21 Shapiro SS and Wilk MB, An analysis of variance test for normality.
Biometrika 52:591–611 (1965).
22 SAS User’s Guide: Statistics, Version 8e. SAS Institute, Cary, NC (2001).
23 Funderburk J, Higley LG and Buntin GD, Concepts and directions in
arthropods pest management. Adv Agron 51:125–172 (1993).
24 Hunt TE, Higley LG and Haile FJ, Imported longhorned weevil
(Coleoptera: Curculionidae) injury to soybean: physiological
response and injury guild-level economic injury levels. JEcon
Entomol 96:1168–1173 (2003).
Pest Manag Sci 2011; 67: 170–174
c
2010 Society of Chemical Industry wileyonlinelibrary.com/journal/ps
174
www.soci.org RCO de Freitas Bueno et al.
25 Hammond RB, Pedigo LP and Poston FL, Green cloverworm leaf
consumption on greenhouse and field soybean leaves and
development of a leaf consumption model. J Econ Entomol
72:714–717 (1979).
26 Board JE, Kumudini S, Omielan J, Piror E and Kahlon CS, Yield response
of soybean to partial and total defoliation during the seed-filling
period. Crop Sci 50:703–712 (2010).
27 Pimentel D, Agroecology and economics, in Ecological Theory and
Integrated Pest Management Practice, ed. by Kogan M. John Wiley &
Sons, Inc., New York, NY, pp. 299–319 (1986).
28 Zilberman D, Schmitz A, Casterline G, Lichtenberg E and Siebert JB,
The economics of pesticides usage and regulation. Science
253:518–522 (1991).
wileyonlinelibrary.com/journal/ps
c
2010 Society of Chemical Industry Pest Manag Sci 2011; 67: 170–174
