Bagworms (Lepidoptera: Psychidae) are leaf eating caterpillars characterized
by the possession of bag, which they build out of tough silk embedded with pieces
of dried plant materials such as leaves and small twigs (Barlow,
1982). In Malaysia, bagworms are common on many orchard, landscape and ornamental
trees (Ahmad and Ho, 1980). The common species found
in oil palm plantation are Metisa plana (Walker), Pteroma pendula
(Joannis) and Mahasena corbetti (Tams) (Wood, 1968;
Sakaran, 1970). In Peninsular Malaysia, M. plana
is the most serious and dominant pest of oil palm (Norman
et al., 1994; Norman and Basri, 2007) while
P. pendula is the second most economical important bagworm (Basri
et al., 1988). Mahasena corbetti is rarely reported as a serious
pest in Peninsular Malaysia, but it is the major defoliator and widely distributed
in Eastern Sabah (Wood and Nesbit, 1969; Young,
1971) and some locations in Peninsular Malaysia, particularly in Johor (Basri
et al., 1988) and Perak (Norman and Basri, 2007).
Before 1956, P. pendula had been reported to occur in small populations
with moderate damages in small areas, while M. plana was never found
(Corley and Tinker, 2003). Pteroma pendula, like
the other bagworms has a wide range host plants, amounting to about 31 species
of shade trees and agricultural crops (Ahmad and Ho, 1980;
Norman et al., 1994). The status of this species
as an important pest of oil palm has been taken over by M. plana since
1955, after repeated applications of broad spectrum persistent insecticides
in oil palm plantations (Wood, 2002). A recent survey
conducted by Norman and Basri (2007) maintains that
M. plana was most widely distributed species and followed by P. pendula
in oil palm plantations in Peninsular Malaysia. Even though, bagworms have been
recurring problems in oil palm plantations, factors contributing to the outbreaks
have yet to be fully understood. Planters often claimed that bagworms outbreaks
were associated with the dry season (Chung and Sim, 1991).
They believed that bagworms feed more actively and spread faster in a dry and
hot weather. Therefore, no information was available to substantiate this phenomenon.
In this study effect of rainfall on the outbreak, the abundance of the bagworm
species and their natural enemies were examined.
MATERIALS AND METHODS
Site Description and Establishment
The study was conducted at Langkap Plantation, in Hutan Melintang, Perak.
The plantation cover an area of 366.66 ha, with 20 divisional plots of approximately
18.33 ha plot-1 The oil palms were planted in 1991.
Bagworm Outbreaks from 2002-2005
Data on the occurrence of bagworms outbreaks in the plantation during the
years 2002 to 2005, recorded by a team two field assistants, were studied and
analyzed. The field assistants marked and estimated the extent of damages of
each palm. This information was then used to estimate volume of insecticide
to be applied to control the bagworms within the plots. Monthly yield production
of Fresh Fruit Bunches (FFB) from the infested plots were obtained from the
management office. The rainfall data were obtained from the whether station
located at the plantation.
Infestation Study from 2005 to 2006
One plot was randomly selected for the study. Application of chemical pesticide
was stopped at least two weeks before commencement of the study. Two hundreds
of palms, about 10% of oil palms in the plot, were selected for bagworm sampling.
The plot was divided into two subplots with 100 palms in each subplot. One subplot
was located near the road side, while the other subplot was in the fifth to
seventh row off the road side. Two palms were randomly selected from each row
and these palms were marked with plastic tape.
The selected palm was observed for the presence bagworms. One frond of the
palm with fresh damage symptoms was cut and about 25 bagworms were sampled each
time. In total, eight fronds of the infested palm were cut and about 200 bagworms
per palm were picked by using a pair of fine forceps. The specimen was placed
inside a small plastic bag and marked. At any time, the forceps used for picking
the bagworms were soaked in alcohol before they were used to collect the next
specimen. The specimens were sorted out in the field office. They were separated
into species. The cases were cut with a pair of fine scissors and examined under
a microscope for any microbial infection and parasitism. The diseased larvae
were placed individually in a plastic vial, marked and brought back to entomology
laboratory in Faculty of Forestry, Universiti Putra Malaysia, for diagnosis.
Samplings were carried out in two phases, twice per month from October to November
(raining season) and from March to April (dry season).
All data were analyzed by using an Analysis of Variance (ANOVA) with further
analysis of Tukey HSD test (p<0.05 and p<0.01).
Rainfall, Yield and Bagworm Outbreak
Serious outbreaks occurred in mid 2002 and 2003. This was followed by almost
uninterrupted outbreaks throughout 2004 and 2005. In 2002, the outbreak was
recorded throughout June to October; with about 25,703 of palms infested. A
record of 6, 598.89 tons crude palm oil was harvested from the plantation during
2002. Figure 1-3 show the yield of the plantation,
the amount of rainfall recorded in the area and the outbreaks of the bagworms
starting from January 2002 to December 2005. During the outbreaks, a total of
34,182 palms were treated. The yield remained high i.e., 6, 356.63 tons, but
decreased by 242.26 tons or 3.67% in the following year. More serious outbreaks
were recorded in 2004. The outbreaks that occurred from January to April 2004
came after treatment. However, outbreaks occurred again during July, September,
October and December in the same year with 87,444 infested palms, an increase
of 54, 262 or 220% from the previous year. Starting from 2004 the yield dropped
sharply to 5, 663.82 tons compared to the yield recorded in 2003. This led to
a decrease of 692.81 tons or 10.90% in fruits harvested as compared to the previous
year. In 2005, serious bagworm outbreaks occurred again in the plantation. The
infestation occurred continuously every month throughout the year, except in
August. A total of 90,980 palms were treated. This was an increase of 3, 536
or 4.04% of infested palms and recorded a yield of 3,953.09 tons of fruits harvested,
a decrease of 1710.29 tons or 30.19% as compared to 2004.
volume and frequencies per month, 2002 to 2005
and treated oil palms in study site during 2002 to 2005
production and bagworm outbreaks, 2002 to 2005
Bagworms Infesting the Oil Palm
Five species of bagworms were recorded from 268 infested oil palms or 67%
of the sampled palms from October to November 2005. Six species of bagworms
were recorded from 391 oil palms from March to April 2006, 97.8% of the sampled
palms. A higher rate of infestation invariably occurred in plots along the roadside
than that of interior side of the plantation (Fig. 4). About
6,166 and 9,775 bagworms collected from October to November 2005 and March to
April 2006, respectively. Out the total, 15,941 individuals, P. pendula
was the most dominant species recorded from the palms. It represented 98.38%
(6, 066 individuals) from October to November 2005 but decreased to 95.2% (9,
306 individuals) from March to April 2006. About 37 individuals (0.60%) of M.
plana was recorded from October to November 2005 and 326 individual (3.34%)
was recorded from March to April 2006.
of infested oil palms from exterior and interior plots
of bagworm species in study site
Seventeen individuals of M. corbetti (0.28%) were recorded from October
to November 2005 and 56 individual (0.57%) from March to April 2006. Bagworms
of Eumeta sp. recorded 21 individuals (0.34%) from October to November
2005 and 28 individuals (0.29%) from March to April 2006. Twenty five individuals
(0.40%) of Amatissa sp. were recorded from October to November 2005 and
29 individuals (0.30%) from March to April 2006. An additional species, Brachycytterus
sp. with total of 30 individuals (0.31%) was recorded from March to April
2006 (Fig. 5).
Status of Bagworms
Data from October to November 2005 show that 71.0% (4, 377 individuals)
of the individuals collected were alive, 7.1% (438 individuals) of the emerged
individuals, 14.8% (912 individuals) empty bags and 7.1% (439 individuals) dead
bagworms. The second bagworm sampling from March to April 2006 recorded 70.1%
(6, 852 individuals) of the individuals collected alive, 17.1% (1, 646 individuals)
emerged individuals, 10.2% (993 individuals) empty bags and 2.9 % (284 individuals)
dead bagworms (Fig. 6).
of bagworms from the study site
factors affecting bagworms population
Mortality Factors Affecting Bagworms
The result shows that predators, parasitoids and fungi were responsible
for causing mortality to the bagworms. Of these natural enemies, predators caused
the highest percentage (37%) of mortality affecting to bagworm population in
the plantation. About 35.9 and 27.2% of mortality were contributed by parasitoids
and fungal infection, respectively. The first sampling recorded 31.1% of the
mortality was contributed by predators, 36.7% by parasitoids and 32.2% by fungal
infection. The second sampling recorded 46% bagworms were attacked by predators,
34.5% of killed by parasitoids and 19.5% were infected by fungi (Fig.
During the first phase of bagworm sampling about 68.8% of the preyed bagworms were attacked by the hemipteran predators. One of the predators, Cosmelestes picticeps, (Hemiptera: Reduviidae), was observed attacking P. pendula during the bagworm sampling. This predator used its piercing-sucking mouthparts to pierce into the bagworm case and attack the bagworm. The inflicted bagworm had a small opening on its case and the case was normally empty. From this study, about 76.6% of the preyed bagworms was caused by this hemipteran. However, the mortality rate caused by predators decreased to 60.6% during second phase of bagworm sampling to from March to April 2006. During the first phase of the sampling, the hemipteran predators caused 75.4% P. pendula and 100% M. plana mortalities, whilst in the second phase of the bagworm sampling the predation caused by hemipterans dropped to 59.8% on P. pendula and 80% on M. plana.
The other predator recorded attacking the bagworms was Callimerus arcufer, (Coleoptera: Cleridae). This species contributed about 31.2% mortality of the bagworm. During the first phase of sampling from October to November 2005, 23.4% of the preyed bagworms were attacked by the C. arcufer and during the second phase, the predation increased to 39.4%. The predation rate of C. arcufer on P. pendula was about 24.6 and 40.2% during the first and second phase of sampling, respectively. Only one case of C. arcufer was recorded from M. plana during the second sampling.
Four species of hyperparasitoids and parasitoids were recorded in early sampling conducted from October to November 2005. The parasitoids were Pediobius imbrues (Hymenoptera: Eulophidae), Pediobius elasmi (Hymenoptera: Eulophidae), Dolichodenidea metasae (Hymenoptera: Braconidae) and Aulosaphes psychidivorus (Hymenoptera: Braconidae). In the second phase of bagworm sampling conducted from March to April 2006, six additional of parasitoids were recorded from the bagworms. The additional species were Aphanogmus thylax (Hymenoptera: Ceraphronidae), Eupelmus catoxanthae (Hymenoptera: Eupelmidae), Eurytoma sp. (Hymenoptera: Eurytomidae), Temelucha sp. (Hymenoptera: Ichneumonidae), Goryphus sp. (Hymenoptera: Ichneumonidae) and Friona sp. (Hymenoptera: Ichneumonidae). The result shows that both P. pendula and M. plana were parasitized by P. imbrues. This species recorded 64% of total parasitoids, 61.7% of total parasitoids from October to November 2005 and 67.7% of total parasitoids from March to April 2006. A total of 62.4 and 67.4% of P. pendula was parasitized by P. imbrues during the October to November 2005 and March to April 2006 samplings. The parasitoid also caused 100 and 75% mortality on M. plana from first and second phase of sampling, respectively.
Dolichodenidea metasae was the second highest parasitoid parasitizing only P. pendula. This species achieved an average 10.7% parasitizing rate, with 17.8% from the first phases of sampling. Pediobius elasmi was the third highest parasitoid recorded 8.8% parasitizing rate, causing mortality only P. pendula with 10.2 and 7.4% from the first and second phases of sampling, respectively. About 8.4% of the bagworms were parasitized by A. psychidivorus. There was 11.1 and 4.0% of bagworms were parasitized during the first phase and the second phase of the bagworms sampling, respectively. The rates of parasitization by A. psychidivorus on P. pendula were 9.6 and 4.2% in the first and second phases of sampling, respectively. Aulesaphes psychidivorus was the only species recorded parasitizing on M. corbetti during first phase of sampling. No parasitoid was recorded in the second phase of sampling on M. corbetti. Another six parasitoids recorded in the second phase of sampling causing mortality on P. pendula were A. thylax (7.4%), E. catoxanthae (3.2%), Eurytoma sp. (2.1%), Temelucha sp. (1.1%), Friona sp. (1.1%) and Goryphus sp. (1.1%). Goryphus sp. also parasitized M. plana (25%) from March to April 2006.
A total of 140 and 53 cases from the first and the second phases of samplings, respectively, were infected with entomopathogenic fungi, Paecilomyces fumosoroseus and Metarhizium anisopliae. They contributed about 23.9 and 6.6% of diseased bagworms, respectively. The other isolated five fungal species were Stachybotrys sp. (7.5%), Pennicillium sp. (15.2%), Aspergillus sp. (13.4%), Trichoderma sp. (12.0%) and Pestalotia sp. (5.9%). About 15.5% of the fungi had yet to be identified. Highest infection occurred in P. pendula, 97.9%, M. plana and M. corbetti recorded 0.7 and 1.4% from October to November 2005, respectively. The mortality caused by fungi during March to April 2006 dropped to about 94.3% on P. pendula and 5.7% on M. plana.
Even though rainfall has been claimed to influence bagworm outbreaks, the relationship
between amount of rainfall and the latter was not significant (R2
= 0.009) (Fig. 8). This phenomenon was shown by the frequency
of bagworm outbreaks in 2005 where outbreaks occurred throughout the year irrespective
of the amount of rainfall. The result of this study confirmed the finding obtained
by Chung and Sim (1991). They recorded building up of
bagworm infestations occurred in a relatively high rainfall period and not necessarily
in dry months, contrary to a belief popular among many planters. Previous study
on the population of the bagworm in oil palm plantation showed that M. plana
is the major bagworm pest rather than P. pendula in Peninsular Malaysia.
The resurgence of P. pendula as the dominant bagworm attacking oil palms
in this part of the country could be due to several factors. As a polyphagous
insect, P. pendula lives on wide range of host plants. They continue
to survive on other host plants and oil palms in villages within the vicinity
when the palms are being treated with insecticides. They move back to the palms
when the palms are relatively free from the insecticide residue. In many instances,
the small holders do not regularly manage the bagworm populations. They believe
that the cost of the bagworm control using trunk injection is not economical
for a small farm. Thus, the mixture of large plantations and the small holdings
in villages, common scene in this region, can be a hindrance in managing P.
Pteroma pendula is the dominant species and responsible for the major defoliation in the oil palm plantation at the study site. The result of this study shows the infestation rates increased to 22% in one month and reached 100% in three months without management on the bagworms. Some other factors may also contribute to the outbreak of the bagworm.
One of the factors that regulate bagworm population is natural enemies. Natural
enemies normally are able to maintain their prey population in balance at low
level under a natural condition. The low population of natural enemies in the
study site might be due to the monoculture in oil palm plantation that provides
large food resource for insect pests (Speight and Wylie, 2001).
This situation adversely affects the stability of natural insect population
but favours the development of insect species that feed on the oil palm.
between rainfall and oil palms infested by bagworms
Natural enemies such as predators, parasitoids and pathogens was found to be
associated with mortality on bagworms. The predatory beetle, Callimerus arcufer
(Cleridae) and hemipteran predators such as Cosmelestes picticeps (Reduvidae),
Sycanus dichotomus (Reduvidae), Systropus roepkei (Reduvidae),
Sycanus macracanthus (Reduvidae) and Isyndrus heros (Reduvidae)
has been recorded to naturally controlling the bagworm population in oil palm
plantation (Basri et al., 1995). However, being
generalist predators, they may not totally prey on the bagworms and may feed
on other insects available in the ecosystem.
Among the parasitoids, Pediobius imbrues was the dominant parasitoids
found with parasitized bagworms. This result corroborates with that of Basri
et al. (1995). Pediobius imbrues has broad range of 18 hosts
and has been recorded as an obligate hyperparasitoids of D. metesae
larvae and as facultative hyperparasitoids of the G. bunoh pupae
(Basri et al., 1995). As the dominant parasitoids
in the study site, P. imbrues might indicating high parasitizing
activities of D. metesae and G. bunoh. Thus, its
effectiveness as primary parasitoids of bagworm might not be affected by its
The mortality rate of bagworm in exterior part was lower than interior part,
probably because the dust billowing from passing vehicles interfere with the
predators and parasitoids in search of hosts as it hinders flight and damages
their wings. This situation was worst during the dry season in the plantation.
The high predation and parasitizing activity on bagworms in the interior part
than exterior part might be due to the different in environmental factors such
as temperature. The Ichneumonid parasitoids for example were preferred the cooler
environment, with high host specific and feeding on the food or plants (Wahl
and Sharkey, 1993; La Salle and Gauld, 1993). The
temperature of the exterior part of the plantation were probably higher than
interior part, with less crown protected to the direct sunlight. Yet, plenty
of flower weed and vegetation were observed in the interior part of the plantation,
indicated the interior part provide more shelter, food and mating site that
important to the parasitoids (Idris et al., 2001)
and predators, indirectly result the mortality rate of the bagworm pest being
preyed and parasitized were centralize in interior part of the study site.
Two of the entomopathogenic fungi, Paecilomyces fumosoroseus and Metarhizium
anisopliae were isolated from fungal infected bagworms. Paecilomyces
fumosoroseus contributed a higher infection rate compared to M. anisopliae
on P. pendula during both dry and wet seasons. Sajap
and Siburat (1992) stated the pupal population of P. pendula had
a higher percentage of infected individuals compared to larval population and
the epizootic of the fungal infection appeared to fluctuate widely over time.
Sajap and Siburat (1992) noted a high proportion of
infected bagworm was found during the wet season. This situation may due to
the nature behaviour of the bagworm pupae which hung from branches and underside
of the leaves. The pupae trapped raindrops with the conidia cause more pupae
to be infected by the fungus (Sajap and Siburat, 1992).
Paecilomyces fumosoroseus is more tolerant in higher temperature up to
32°C, than M. anisopliae (30°C) (Vidal et
al., 1997). This factor probability contributes to the high prevalence
of P. fumosoroseus in the field.
The risk of bagworms outbreaks of could be reduced if damage can be detected
prior to firm establishment of the bagworm population. It is important for a
plantation to have infestation data from its regular monitoring program for
decision making for estimation of potential crop loss and to determine the need
for control, especially when the natural enemies have failed to regulate the
pest population below its economic threshold (Wood, 2002).
Thus, a regular census should be conducted from time to time for effective bagworm
control thereby minimizing yield loss. The natural enemies, predators, parasitoids
and pathogens, were ineffective in keeping the bagworm populations below the
economic threshold level, with less than 8% of sampled bagworms. Thus, an effort
to augment these natural enemies is warranted.
The authors wish to thank the management of MHC Plantations Berhad for providing the permission to conduct this study in their premise. We are also indebted to all kinds of assistance rendered including field work, transportation, lodging and necessary information needed to undertake the study.