Mortality and Repellent Effects of Coffee Extracts on The Workers of Three Household Ant Species

Abstract

Coffee consists of a variety of chemical compounds that have not been documented to have resistance on insects. Hence, this research was conducted to study the impact of coffee extracts impregnated in gel bait towards survival and feeding behaviour of Tapinoma indicum (ghost ant), Pheidole megacephala (big-headed ant) and Monomorium pharaonis (Pharaoh ant) (Hymenoptera: Formicidae). The three coffee species used were Coffea arabica, Coffea canephora and Coffea liberica. The coffee extracts were obtained using Soxhlet extraction method, diluted to 0.01%, 0.05% and 0.10% concentration, and eventually impregnated into two sets of gel bait at with the first set (Set I) sugar solution and the second set (Set II) with distilled water. The overall results indicated that Coffea arabica gave highest mortality on all three ant species and higher concentration of extracts showed higher ant mortality in most bioassays. The higher mortality in lower concentration bioassays was probably due to their lower repellency percentages. Furthermore, Set I bioassays had higher mortality as the sugar used act as food attractant. T. indicum was the most susceptible species. Owing to the low mortality, the low concentration of coffee used was not effective in killing household ants but it did repel them.

Full text

Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
TROPICAL AGRICULTURAL SCIENCE
Journal homepage: http://www.pertanika.upm.edu.my/
Article history:
Received: 7 August 2017
Accepted: 16 January 2018
Published: 14 November 2018
ARTICLE INFO
E-mail addresses:
abdhaz@usm.my (Abdul Haz Ab Majid)
sallyeoh93@gmail.com (Xue Li Yeoh)
hamachan1@yahoo.com (Hamady Dieng)
* Corresponding author
ISSN: 1511-3701
e-ISSN 2231-8542 © Universiti Putra Malaysia Press
Mortality and Repellent Effects of Coffee Extracts on The Workers
of Three Household Ant Species
Xue Li Yeoh1, Hamady Dieng2 and Abdul Haz Ab Majid1*
1Household and Structural Urban Entomology Laboratory, Vector Control Research Unit, School of Biological
Sciences, Universiti Sains Malaysia, 11800, Malaysia
2Institute of Biodiversity and Environmental Conservation, Universiti Malaysia Sarawak, UNIMAS, 94300,
Kota Samarahan, Sarawak, Malaysia
ABSTRACT
Coffee consists of a variety of chemical compounds that has not been documented to have
resistance on insects. Hence, this research was conducted to study the impact of coffee
extracts impregnated in gel bait towards survival and feeding behaviour of Tapinoma
indicum (ghost ant), Pheidole megacephala (big-headed ant) and Monomorium pharaonis
(Pharaoh ant) (Hymenoptera: Formicidae). The three coffee species used were Coffea
arabica, Coffea canephora and Coffea liberica. The coffee extracts were obtained using
Soxhlet extraction method, diluted to 0.01%, 0.05% and 0.10% concentration, and
eventually impregnated into two sets of gel bait at with the rst set (Set I) sugar solution
and the second set (Set II) with distilled water. The overall results indicated that Coffea
arabica gave highest mortality on all three ant species and higher concentration of extracts
showed higher ant mortality in most bioassays. The higher mortality in lower concentration
bioassays was probably due to their lower repellency percentages. Furthermore, Set I
bioassays had higher mortality as the sugar used act as food attractant. T. indicum was the
most susceptible species. Owing to the low mortality, the low concentration of coffee used
was not effective in killing household ants but it did repel them.
Keywords: Coffee extract, gel bait, household ants,
Monomorium pharaonic, Pheidole megacephala,
soxhlet extraction, Tapinoma indicum

Xue Li Yeoh, Hamady Dieng and Abdul Haz Ab Majid
1558
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
INTRODUCTION
Ants are known to be an ecologically
dominant group and show the highest level
of diversity among eusocial insects (Wilson
& Hölldobler, 2005). They involve in the
interactions with other organisms and also
the functional ecosystem processes (Wilson
& Hölldobler, 2005). With the exception of
Antarctica and Arctic, their dominance is
indicated by their worldwide geographical
distribution. Each ant species possesses its
own particular morphological structures and
behaviours, making it distinguishable from
other ant species (Lucky, 2009).
Furthermore, ants are one of the most
nuisance urban pests when they enter from
outdoors to indoors in searching of food
and water. Due to their properties of large
number appearance, cause contamination of
food and hospital sterile equipments, they
are considered as nuisance pests and disease
organism carriers which make them to be
recognized as potential mechanical vectors
of human diseases (Beatson, 1972). Some
ant species have the ability to cause painful
bites or stings with their pincer-like jaws
or venomous stings (Marer & Flint, 1991).
Household ants can be discovered with a
higher frequency in tropical areas compared
to temperate areas (Campos-Farinha, 2005;
Fowler, Filho, & Bueno, 1993).
In Asia, the pest status of household ants
was less signicant in the 1990s. However,
this situation had soon changed owing to
the rise of its pest status (Lee, 2000). In the
pest control company of Malaysia, around
10% of the business was constituted by
the controlling of ant in 1995 (Na & Lee,
2001). While in United States, the ant
control revenue of ant is so high that the ants
have attained the top household pest status
(Gooch, 1999; Jenkins, 2001; Kaminski,
2000) and ranked the most troublesome pest
(Gooch, 1999). From a residential survey,
which was carried out in 1995, ants have
attained the status of the most important
household pests after mosquitoes and
cockroaches (Na & Lee, 2001). In Malaysia,
there are 23 species of household ants with
a total of 15 genera described (Na & Lee,
2001). However, in this research, only three
common species of household ants are
focused: Tapinoma indicum (Forel) (ghost
ant), Pheidole megacephala (F.) (big-headed
ant) and Monomorium pharaonis (L.)
(Pharaoh ant) (Hymenoptera: Formicidae).
There are several methods to control
the household ants. Baiting and residual
spraying are the common methods for
controlling ants (Lee, 2000), but baiting
has served as a more popular method due
to its usage safety, target-specic and ability
to eliminate or suppress the whole ant
colony without the requirement to locate
the nest (Suiter, Wu, & Bennett, 1997).
Generally, baits are more effective against
household ants as many residual contact
insecticides used act repellent to ants,
especially Pharaoh’s ants (Gooch, 1999).
Residual insecticide treatment just acts as
barriers of preventing ants from entering
the houses instead of eliminating the ant
population (Klotz, Greenberg, Shorey, &
Williams, 1997). Hence, this method is not
effective against some species of household
ants which reside within the house (Lee,
2000).

Coffee extracts on household ant’s mortality
1559
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
Recent studies have shown that plants
such as coffee and tea are used as effective
biological agent in controlling insects (Ab
Majid et al., 2018). Coffee consists of over
1000 chemical compounds (Farah, 2012);
while few Coffea species are resistant
to insect attack naturally (Jaramillo,
Borgemeister, & Baker, 2006). Coffee
has been utilized to study toxicological
effects on several organisms. Caffeine
causes damage to the nervous system in
bullfrog (Higure & Nohmi, 2002), blocks
the fetal development of Rattus norvegicus
(Smith, McElhatton, & Sullivan, 1987)
and inhibits oviposition of shot-hole borer
beetle (Hewavitharanage, Karunaratne,
& Kumar, 1999). Coffee is known to be a
natural repellent to ants at which ants repel
when contact with the coffee grounds. Few
researches have been reported that coffee
is effective in decreasing the mosquitoes’
reproductive capacity (Laranja, Manzatto, &
de Campos Bicudo, 2003), repelling gravid
Aedes albopictus female and inhibiting the
development of their embryos (Satho et al.,
2015).
In this study, three species of coffee:
Coffea arabica (Arabica coffee), Coffea
canephora (Robusta coffee) and Coffea
liberica (Liberian coffee) are extracted by
using Soxhlet extraction and impregnated in
the gel bait to test their impact in controlling
household ants. The coffee species used are
roasted type in which they are not mixed
with sugar to avoid creating bias towards
the attraction of ants. The objectives of this
study are, to investigate the effect of the
extracts of C. arabica, C. canephora and
C. liberica impregnated in gel bait towards
survival of the workers of T. indicum, P.
megacephala and M. pharaonis.
MATERIALS AND METHODS
Coffee Source
The coffee beans of C. arabica, C.
canephora and C. liberica were obtained
from Cap Kuda Coffee Company, Sabah,
Malaysia. The coffee beans were roasted
without adding any sugar compounds. The
temperatures used in the roasting process
vary from 210°C to 240°C and the roasting
time used was about 12 to 30 min. These
roasted coffee beans were then ground,
packaged and shipped to Universiti Sains
Malaysia.
Extract Coffee Using Soxhlet Extraction
A small cotton ball was moistened with
water and placed in the chamber of Soxhlet
extractor. Fifty grams of each of the roasted
coffee (C. arabica, C. canephora and
C. liberica) were weighted and placed
separately into the Soxhlet extractor. A
volume of 250 ml of methanol that used as
the extraction solvent was poured into the
at-bottomed ask. The at-bottomed ask
was then placed on the heating mantle; the
Soxhlet extractor together with the reux
condenser was placed atop of the flat-
bottomed ask. The Soxhlet extractor was
xed and held by retort stand. Both ends of
the reux condenser were connected to pipes
for water in and water out.
When the apparatus was ready, the
extraction solvent (methanol) was heated

Xue Li Yeoh, Hamady Dieng and Abdul Haz Ab Majid
1560
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
until its boiling point (64.7 °C) was
achieved. Its vapour condensed in the
condenser and the condensed extractant
dripped into the chamber containing the
coffee. When the liquid level in the chamber
had risen to the top of the siphon tube, the
extract-containing solvent of the Soxhlet
chamber were siphoned into the flat-
bottomed ask. The whole apparatus was
heated for 5 h (Mgbemena, Ebe, Nnadozie,
& Ekeanyanwu, 2015).
After 5 h, the entire apparatus was left
to be cooled down. The coffee extracts were
then collected and poured to a glass petri
dish with correct label. The extracts were
then placed into drying oven (Memmert
GmbH + Co, KG, Western Germany) at 80
°C for evaporation for three days to obtain
the coffee extract in solute form. The coffee
extracts were then taken out and scraped
off by using spatula. The scraped coffee
extracts were kept in universal bottle with
labeling and then stored in refrigerator for
further used.
Collection and Identication of the
T. indicum, P. megacephala and M.
pharaonis
Field populations of T. indicum, P.
megacephala and M. pharaonis workers
were collected from the Minden campus
of Univeristi Sains Malaysia, Penang,
Malaysia from 7:00 a.m. to 10:00 a.m.
The traps were set up by using Eppendorf
tubes (with modied holes on the tubes) at
which the inner surface of the Eppendorf
tubes was coated with a thin layer of Fluon,
polytetrauroethylene suspension (BioQuip
Products, Inc., California) to prevent the
trapped ants from escaping (Eow, Chong,
& Lee, 2004). A minute amount of peanut
butter or honey which acts as food attractant
to ants was placed on a small piece of paper
and inserted it into the Eppendorf tubes. The
trapped ants were collected after 1 to 2 h and
transferred into container at which the inner
surface was coated with fluon. The ants
were put in 90% ethanol for identication
according to their distinct characteristics
based on descriptions by Na and Lee (2001)
and Lucky (2009). A brush was used to
separate the ant species if more than one
ant species were trapped in the same tube.
Preparation of Gel Bait by Using
Different Concentration of Coffee
Extracts
The coffee extract solution with desired
concentration was prepared by mixing
coffee extracts and distilled water or 20%
of sugar solution. Preparation of coffee
concentration. i.e. 0.01%, 0.05% and 0.10%
were produced using Arabica, following
method of Ab Majid et al. (2018) with slight
modication. To allow the coffee extracts
to dissolve completely in the solution,
the solution was allowed to stir by using
magnetic stirrer for 30 to 45 min. After all,
the gelling agent, Ferti-plant jelly (Fertiland
Trading Co., Malaysia) was added into the
prepared solution, allowing it to absorb the
solution and expand to its maximum size
for 12 h.
Two sets of gel baits were prepared.
The rst set of gel was the mixture of coffee
extract and 20% of sugar solution (Set I).

Coffee extracts on household ant’s mortality
1561
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
The scraped coffee extract was diluted to
different concentration such that 0.01%
(low), 0.05% (medium) and 0.10% (high)
by using 100 ml of 20% of sugar solution.
The blank bait (control) used for this set
contained only 20% of sugar solution.
The second set of gel was the mixture of
coffee extract and distilled water (Set II).
The scraped coffee extract was diluted to
different concentration such that 0.01%
(low), 0.05% (medium) and 0.10% (high)
by using 100 ml of distilled water. The blank
bait (control) used consists of only distilled
water. All of the gels were made constant
mass of 0.50 g by using weighing machine.
Bioassay
A small hole (5 mm in diameter) was made
at the center of the petri dish lid by using a
hot soldering iron (Williams, 1989). This
was to insert the cotton wool moistened with
distilled water (without touching the base of
petri dish) as moisture for ants (Figure 1)
A 90 mm diameter filter paper was
attached to the outer base of petri dish to
ease the counting process. The perimeter
of the petri dish inner surface was coated
with a thin layer of petroleum jelly (Vasline,
Unilever Thai, Thailand) (Figure 1). Thirty
ant workers were randomly picked and
transferred to a petri dish (90 mm in
diameter). The rst set of gel bait was placed
in the petri dish (Set I). Gel bait with only
20% of sugar solutions (without any coffee
extracts) was used as the control of the
experiment (Figure 2). Paralm was used
to seal up the petri dish to prevent the ants
from escaping.
The ants were observed in 30 min, 1
h, 2 h, 4 h, 8 h, 24 h (1 day), 48 h (2 days)
and 72 h (3 days) under temperature of 25
± 2°C and relative humidity of 76 ± 10%.
The number of ants which cannot move or
respond (ant mortality) were counted and
recorded. Their repellency and behaviours
were also observed and recorded under
dissecting microscope. The number of
ants that were not attracted to the region
with gel bait during the observed time was
considered as repelling.
The above steps were repeated to
complete three sets of replicates for each
ant species with different concentrations
(0.01%, 0.05% and 0.10%) for each coffee
species. The bioassay for the second set
of gel bait was conducted using the same
procedures.
The mean repellency percentage for 30
min, 1 h, 2 h, 4 h, 8 h, 24 h, 48 h and 72
h; and mean mortality were obtained.
The signicant differences of mortality
and repellency were determined using
Kruskal-Wallis (KW) analysis of variance
by SPSS 22.0 software. The repellency
percentage (PR) was calculated using
the formula (Abdullah et al., 2015):
where,
NC = Number of ants on the region
without gel bait
NT = Number of ants on the region with
gel bait

Xue Li Yeoh, Hamady Dieng and Abdul Haz Ab Majid
1562
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
RESULTS
Effect of Coffee Extracts Impregnated
in Gel Bait Towards Survival of T.
indicum, P. megacephala And M.
pharaonis
C. canephora and C. liberica showed
insignicant result (P>0.05) with P value
0.109 and 0.054 respectively against T.
indicum between the three concentrations
(0.01%, 0.0% and 0.10%) and control
in Set I bioassay (Figure 1). C. arabica
showed signicant results (P<0.05) with P
value 0.032 against T. indicum between the
treatments and control in Set I bioassay but
no signicant differences among the three
concentrations (Figure 1). In Set II bioassay,
C. canephora displayed insignicant results
(P>0.05) with P value 0.079 against T.
indicum; while C. arabica and C. liberica
had significant results (P<0.05) with P
values 0.044 and 0.030 respectively against
T. indicum between treatments and control
but no signicant differences among the
three treatments (Figure 2).
In Set I bioassay, P. megacephala
displayed the same results with those
of T. indicum for the similar set. Both
C. canephora and C. liberica showed
insignicant result (P>0.05) with P values
0.144 and 0.114 respectively between the
concentrations and control (Figure 3).
C. arabica indicating significant results
(P<0.05) with P value 0.030 between the
treatments and control but no signicant
difference among the three concentrations
(Figure 3). In Set II bioassay, there were
no signicant results (P>0.05) for all the
three coffee species (C. arabica P=0.067;
C. canephora P=0.127; and C. liberica
P=0.392) against P. megacephala(Figure
4).
For M. pharaonis, all three coffee
species (C. arabica P=0.134; C. canephora
P=0.144; C. liberica P=0.212) had
insignicant results (P>0.05) between the
three concentrations and control in Set I
bioassay (Figure 5). In Set II bioassay, there
were insignicant results (P>0.05) for C.
arabica (P=0.132) and C. liberica (P=0.441)
between the concentrations and control. On
the other hand, C. canephora (P=0.048)
showed a signicant difference (P<0.05).
Nevertheless, it did not display signicant
results among the three concentrations
(Figure 6).
At all concentrations in both sets, C.
arabica, C. canephora and C. liberica
showed insignificant results (P>0.05)
among themselves against T. indicum with
the exception of 0.05% concentration of
Set II bioassay (Figures 1 and 2). Instead,
its P value of 0.047 displayed a signicant
difference of the ant mortality among the
coffee species at 0.05% at which C. arabica
and C. canephora significantly differed
from C. liberica. For P. megacephala,
0.01% showed insignicant results (P>0.05)
among the three coffees with P value 0.141
and 0.100 in Set I and Set II bioassay
respectively. However, 0.05% and 0.10%
in Set I bioassay indicated significant
results (P<0.05) among the coffees with
P value 0.042 and 0.048 respectively, at
which C. arabica differed signicantly to
both C. canephora and C. liberia (Figure

Coffee extracts on household ant’s mortality
1563
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
There were no significant results
(P>0.05) of mean repellency percentage at
all observed time intervals (30 min, 1 h, 2 h,
4 h, 8 h, 24 h, 48 h and 72 h) in T. indicum
between the treatments and control of C.
arabica and C. liberica in both Set I and
Set II bioassays.
C. canephora had signicant differences
at 24 h (P=0.037) and 72 h (P=0.037) in Set
I bioassay; and at 48 h (P=0.031) in Set II
bioassay between concentrations and control
(Tables 1 and 2).
3). For Set II bioassay, the result (P=0.088,
P>0.05) showed no signicant difference at
0.05% among the three Coffea spp.; but it
showed that C. arabica (P=0.034, P<0.05)
experienced significant difference to C.
canephora and C. liberica at 0.10% (Figure
4). The analysis for M. pharaonis showed
that at all concentrations, there were no
signicant differences among the coffees
in both sets of bioassays (Set I: P=0.633 at
0.01%, 0.612 at 0.05% and 0.966 at 0.10%;
Set II: P=0.264 at 0.01%, 0.641 at 0.05%
and 0.396 at 0.10%) (Figure 5).
Figure 1. Mean mortality percentage of three different coffees against T. indicum in Set I bioassay

Xue Li Yeoh, Hamady Dieng and Abdul Haz Ab Majid
1564
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
Figure 2. Mean mortality percentage of three different coffees against T. indicum in Set II bioassay
Figure 3. Mean mortality percentage of three different coffees against P. megacephala in Set I bioassay

Coffee extracts on household ant’s mortality
1565
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
Table 1
Mean mortality percentage and mean repellency percentage of three different coffees against T. indicum in Set I bioassay
Coffee
species
Concentration
(%)
Mean
Mortality
Percentage
(%)1
Mean Repellency Percentage (%)2
30min 1h 2h 4h 8h 24h 48h 72h
Coffea
arabica
0.01% 20.00 ±
5.77Aa
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C 95.55 ± 2.22C 100.00 ±
0.00C
0.05% 46.67 ±
12.62Aa
100.00 ±
0.00C
95.56 ±
4.44C
93.33 ±
6.67C
100.00 ±
0.00C
97.78 ±
2.22C
100.00 ±
0.00C 100.00 ± 0.00C 100.00 ±
0.00C
0.10% 55.55 ±
13.92Aa
97.78 ±
2.22C
100.00 ±
0.00C
100.00 ±
0.00C
97.78 ±
2.22C
100.00 ±
0.00C
100.00 ±
0.00C 97.78 ± 2.22C 100.00 ±
0.00C
Control 4.44 ± 1.11B 86.67 ±
10.18C
91.11 ±
8.89C
95.56 ±
4.44C
95.55 ±
2.22C
93.33 ±
3.85C
88.89 ±
5.88C 91.11 ± 5.88C 95.56 ±
4.44C
Coffea
canephora
0.01% 17.78 ±
6.19Aa
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C 97.78 ± 2.22C 100.00 ±
0.00C
0.05% 20.00 ±
7.70Aa
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C 97.78 ± 2.22C 100.00 ±
0.00C
0.10% 23.33 ±
6.67Aa
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
97.78 ±
2.22C 100.00 ± 0.00C 100.00 ±
0.00C
Control 4.44 ± 1.11A 95.55 ±
2.22C
93.33 ±
3.85C
97.78 ±
2.22C
93.33 ±
3.85C
95.55 ±
2.22C
88.89 ±
4.44D 86.67 ± 3.85C 88.89 ±
4.44D

Xue Li Yeoh, Hamady Dieng and Abdul Haz Ab Majid
1566
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
Coffea
liberica
0.01% 8.89 ±
2.94Aa
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
95.56 ±
4.44C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
0.05% 5.56 ±
2.94Aa
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
91.11 ±
8.89C
97.78 ±
2.22C
0.10% 26.67 ±
7.70Aa
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
97.78 ±
2.22C
95.55 ±
2.22C
97.78 ±
2.22C
Control 2.22 ±
2.22A
100.00 ±
0.00C
95.55 ±
2.22C
97.78 ±
2.22C
91.11 ±
4.44C
91.11 ±
5.88C
86.67 ±
6.67C
84.44 ±
4.44C
86.67 ±
7.70C
1 Mean mortality percentage followed by different letters within the same column are signicant different by subjecting to Kruskal-Wallis H Test at p<0.05: A &
B (comparison of concentration within the same coffee species); a & b (comparison of coffee species within a concentration).
2 Mean repellency percentage followed by different letters within the same column are signicant different by subjecting to Kruskal-Wallis H Test at p<0.05: C,
D & E (comparison of concentration within the same coffee species).
Table 1 (Continue)

Coffee extracts on household ant’s mortality
1567
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
Table 2
Mean mortality percentage and mean repellency percentage of three different coffees against T. indicum in Set II bioassay
Coffee
species
Concentration
(%)
Mean
Mortality
Percentage
(%)1
Mean Repellency Percentage (%)2
30min 1h 2h 4h 8h 24h 48h 72h
Coffea
arabica
0.01% 22.22 ±
2.22Aa
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C 97.78 ± 2.22C
0.05% 25.56 ±
6.76Aa
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
97.78 ±
2.22C
97.78 ±
2.22C
97.78 ±
2.22C
95.55 ±
4.45C
100.00 ±
0.00C
0.10% 33.33 ±
3.85Aa
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
97.78 ±
2.22C 97.78 ± 2.22C
Control 6.67 ± 0.00B 97.78 ±
2.22C
95.56 ±
4.44C
100.00 ±
0.00C
95.55 ±
2.22C
95.56 ±
4.44C
91.11 ±
4.44C
95.56 ±
4.44C
84.44 ±
12.37C
Coffea
canephora
0.01% 18.89 ±
2.22Aa
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
95.55 ±
2.22C
97.78 ±
2.22C
97.78 ±
2.22C
100.00 ±
0.00C
0.05% 16.67 ±
5.09Aa
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
97.78 ±
2.22C
97.78 ±
2.22C
100.00 ±
0.00C
100.00 ±
0.00C 91.11 ± 8.89C
0.10% 22.22 ±
5.88Aa
100.00 ±
0.00C
97.78 ±
2.22C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
Control 3.33 ± 1.93A 97.78 ±
2.22C
95.55 ±
2.22C
97.78 ±
2.22C
95.56 ±
4.44C
93.33 ±
0.00C
93.33 ±
6.67C
88.89 ±
2.22D 88.89 ± 5.88C

Xue Li Yeoh, Hamady Dieng and Abdul Haz Ab Majid
1568
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
Coffea
liberica
0.01% 15.56 ±
2.94Aa 100.00 ± 0.00 100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C 97.78 ± 2.22C
0.05% 1.11 ± 1.11Ab 100.00C ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C 97.78 ± 2.22C
0.10% 21.11 ±
5.56Aa
97.78 ±
2.22C
97.78 ±
2.22C
100.00 ±
0.00C
97.78 ±
2.22C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
Control 4.44 ± 1.11B 97.78 ±
2.22C
95.56 ±
4.44C
100.00 ±
0.00C
97.78 ±
2.22C
95.56 ±
4.44C
97.78 ±
2.22C
95.56 ±
4.44C 93.33 ± 3.85C
1 Mean mortality percentage followed by different letters within the same column are signicant different by subjecting to Kruskal-Wallis H Test at p<0.05: A
& B (comparison of concentration within the same coffee species); a & b (comparison of coffee species within a concentration).
2 Mean repellency percentage followed by different letters within the same column are signicant different by subjecting to Kruskal-Wallis H Test at p<0.05: C,
D & E (comparison of concentration within the same coffee species).
Table 2 (Continue)

Coffee extracts on household ant’s mortality
1569
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
Figure 3. Mean mortality percentage of three different coffees against P. megacephala in Set I bioassay
Figure 4. Mean mortality percentage of three different coffees against P. megacephala in Set II bioassay

Xue Li Yeoh, Hamady Dieng and Abdul Haz Ab Majid
1570
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
Figure 5. Mean mortality percentage of three different coffees against M. pharaonis in Set I bioassay
Figure 6. Mean mortality percentage of three different coffees against M. pharaonis in Set II bioassay

Coffee extracts on household ant’s mortality
1571
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
Effect of Coffee Extracts Impregnated
in Gel Bait Towards the Feeding
Behaviour of T. indicum, P. megacephala
and M. pharaonis
In Set I bioassay, P. megacephala showed
no signicant difference (P>0.05) of their
repellency behaviour at 30 min (P=0.416), 1
h (P=0.164), 2 h (P=0.382), 24 h (P=0.587),
48 h (P=0.056) and 72 h (P=0.229) between
all concentrations of C. arabica and
the control. At 4 h, the P values 0.038
displayed significant difference among
the concentrations and control, at which
0.05% differed to 0.01% (P=0.043), 0.10%
(P=0.043) and the control (P=0.046). At 8 h,
the P value 0.028 (P<0.05) was a signicant
result; with 0.01% and 0.05% differed
from both 0.10% and control (Table 3).
For C. canephora, P. megacephala showed
insignicant results for all time intervals
except at 48 h (P=0.030, P<0.05); while
there were no signicant differences in C.
liberica at all time intervals between the
concentrations and control (Table 3).
In Set II bioassay, C. arabica indicated
no signicant differences of P. megacephala
feeding behaviour at all time intervals
with the exception of 48 h (P=0.023,
P<0.05) and 72 h (P=0.025, P<0.05)
between concentrations and control. At
48 h, 0.01% (P=0.05), 0.05% (P=0.05)
and 0.10% (P=0.046) differed from the
control. At 72 h, 0.05% and 0.10% had
signicant differences with 0.01% and the
control (Table 4). C. canephora showed
signicant results at 30 min (P=0.043), 8 h
(P=0.025), 24 h (P=0.013), 48 h (P=0.017)
and 72 h (P=0.016) between concentrations
and control. At 30 min, 8 h and 24 h,
the results were significantly differed
from the control but not among the three
concentrations; while at 48 h and 72 h, the
results displayed significant differences
among concentrations and also with the
control (Table 3). For C. liberica, only 48
h (P=0.024) and 72 h (P=0.024) showed
signicant results, at which they differed
among the concentrations and with the
control (Table 4).
For M. pharaonis, there were no
significant results (P>0.05) of mean
repellency percentage at all observed
time intervals between the concentrations
and control of all Coffea spp. in Set II
bioassay (Table 6). On the other hand,
in Set I bioassay, C. liberica displayed
significant differences at 8 h (P=0.012),
24 h (P=0.013), 48 h (P=0.032) and 72 h
(P=0.012); but there were no differences
among the concentrations (Table 6).
In Set I bioassay, the mortality of T.
indicum and P. megacephala increased
with the increasing concentration (0.01%,
0.05% and 0.10%) of C. arabica and C.
canephora but this trend was not shown in
C. liberica. Instead, the results showed the
lowest mortality at 0.05% of concentration
for both the ant species, but still there
were no signicant differences (T. indicum
P=0.054; P. megacephala P=0.114, P>0.05)
among the three concentrations in C.
liberica (Figures 1 and 5). At 0.05% of C.
liberica, the mean repellency percentage for
P. megacephala at 1 h, 2 h, 4 h, 8 h and 24 h
are relatively higher as compared to that of
0.01% and 0.10%. These higher percentages
indicated P. megacephala repelled more

Xue Li Yeoh, Hamady Dieng and Abdul Haz Ab Majid
1572
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
Table 3
Mean mortality percentage and mean repellency percentage of three different coffees against P. megacephala in Set I bioassay
Coffee
species
Concentration
(%)
Mean Mortality
Percentage
(%)1
Mean Repellency Percentage (%)2
30min 1h 2h 4h 8h 24h 48h 72h
Coffea
arabica
0.01% 11.11 ± 4.84Aa 84.45 ±
9.69C
88.89 ±
8.01C
84.44 ±
8.89C
-8.89 ±
4.44C
97.78 ±
2.22C
91.11 ±
5.88C
80.00 ±
7.70C
71.11 ±
8.01C
0.05% 36.67 ± 6.67Aa 91.11 ±
2.22C
93.33 ±
3.85C
82.22 ±
8.01C
95.56 ±
4.44D
100.00 ±
0.00C
77.8 ±
9.69C
80.00 ±
3.85C
40.00 ±
34.21C
0.10% 42.22 ± 4.01Aa 84.44 ±
15.56C
86.67 ±
6.67C
55.56 ±
14.57C
15.56 ±
11.11C
-53.33 ±
23.09D
68.89 ±
16.02C
42.22 ±
12.37C
77.78 ±
8.89C
Control 4.45 ± 2.22B 48.89 ±
22.55C
48.89 ±
15.56C
53.33 ±
25.24C
-26.67 ±
17.64C
-60.00 ±
13.88D
20.00 ±
7.70C
31.11 ±
29.15C
24.22 ±
18.19C
Coffea
canephora
0.01% 4.44 ± 1.11Aa 100.00 ±
0.00C
97.78 ±
2.22C
95.56 ±
4.44C
88.89 ±
8.01C
97.78 ±
2.22C
93.33 ±
6.67C
91.11 ±
8.89C
88.89 ±
4.44C
0.05% 5.56 ± 2.94Ab 88.89 ±
8.01C
82.22 ±
11.11C
100.00 ±
0.00C
97.78 ±
2.22C
100.00 ±
0.00C
100.00 ±
0.00C
-11.11 ±
5.88D
75.56 ±
8.01C
0.10% 6.66 ± 3.33Ab 100.00 ±
0.00C
93.33 ±
3.85C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
93.33 ±
3.85C
82.22 ±
9.69C
Control 0.00 ± 0.00A 88.89 ±
5.88C
82.22 ±
9.69C
93.33 ±
6.67C
82.22 ±
2.22C
100.00 ±
0.00C
82.22 ±
11.11C
-26.66 ±
6.67D
42.22 ±
15.56C

Coffee extracts on household ant’s mortality
1573
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
Coffea
liberica
0.01% 25.56 ± 9.87Aa 97.78 ±
2.22C
97.78 ±
2.22C
100.00 ±
0.00C
93.33 ±
6.67C
100.00 ±
0.00C
93.33 ±
3.85C
33.33 ±
26.94C
35.55 ±
18.99C
0.05% 11.11 ± 2.22Ab 97.78 ±
2.22C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
77.78 ±
8.89C
95.55 ±
2.22D
0.10% 13.33 ± 3.33Ab 95.55 ±
2.22C
88.89 ±
11.11C
95.56 ±
4.44C
93.33 ±
3.85C
93.33 ±
3.85C
86.67 ±
10.18C
97.78 ±
2.22D
97.78 ±
2.22D
Control 3.33 ± 1.93A 82.22 ±
11.11C
86.67 ±
10.18C
93.33 ±
3.85C
93.33 ±
0.00C
82.22 ±
5.88D
55.56 ±
19.75C
24.45 ±
13.52C
26.67 ±
30.06C
1 Mean mortality percentage followed by different letters within the same column are signicant different by subjecting to Kruskal-Wallis H Test at p<0.05: A &
B (comparison of concentration within the same coffee species); a & b (comparison of coffee species within a concentration).
2 Mean repellency percentage followed by different letters within the same column are signicant different by subjecting to Kruskal-Wallis H Test at p<0.05:
C, D & E (comparison of concentration within the same coffee species).
Table 3 (Continue)
Table 4
Mean mortality percentage and mean repellency percentage of three different coffees against P. megacephala in Set II bioassay
Coffee
species
Concentration
(%)
Mean Mortality
Percentage (%)1
Mean Repellency Percentage (%)2
30min 1h 2h 4h 8h 24h 48h 72h
Coffea
arabica
0.01% 3.33 ± 1.93Aa 80.07 ±
13.81C
71.11 ±
14.57C
53.33 ±
3.85C
-31.11 ±
9.69C
-35.56 ±
27.84C
53.33 ±
11.55C
42.22 ±
25.04C
-42.22 ±
31.35D
0.05% 12.22 ± 7.29Aa 77.78 ±
5.88C
80.00 ±
11.55C
62.22 ±
9.69C
-24.44 ±
15.55C
-46.66 ±
13.33C
86.67 ±
7.70C
77.78 ±
12.37C
42.22 ±
8.89C
0.10% 13.33 ± 1.93Aa 84.45 ±
9.69C
68.89 ±
17.78C
57.78 ±
14.57C
73.33 ±
10.18C
42.22 ±
29.14C
68.89 ±
31.11C
95.55 ±
2.22C
80.00 ±
3.85C
Control 1.11 ± 1.11A 57.78 ±
32.28C
60.00 ±
23.09C
48.89 ±
13.52C
-53.33 ±
37.12C
-62.22 ±
31.35C
46.67 ±
7.70C
-57.78 ±
21.20D
-57.78 ±
9.69D

Xue Li Yeoh, Hamady Dieng and Abdul Haz Ab Majid
1574
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
Coffea
canephora
0.01% 4.44 ± 1.11Aa 91.11 ±
8.89C
60.00 ±
19.24C
91.11 ±
8.89C
97.78 ±
2.22C
100.00 ±
0.00C
100.00 ±
0.00C
97.78 ±
2.22C
62.22 ±
8.01C
0.05% 4.44 ± 2.94Aa 64.45 ±
9.69C
73.33 ±
17.64C
82.22 ±
8.89C
82.22 ±
11.76C
100.00 ±
0.00C
100.00 ±
0.00C
77.78 ±
5.88D
33.33 ±
7.70C
0.10% 4.44 ± 1.11Ab 100.00 ±
0.00C
95.56 ±
4.44C
93.33 ±
6.67C
100.00 ±
0.00C
97.78 ±
2.22C
100.00 ±
0.00C
100.00 ±
0.00C
97.78 ±
2.22D
Control 0.00 ± 0.00A 48.89 ±
13.52D
37.78 ±
19.37C
68.89 ±
19.37C
55.56 ±
37.97C
71.11 ±
5.88D
57.78 ±
13.52D
-33.33 ±
16.78E
-51.11 ±
8.89E
Coffea
liberica
0.01% 0.00 ± 0.00Aa 66.67 ±
11.55C
73.33 ±
10.19C
82.22 ±
8.01C
68.89 ±
27/84C
91.11 ±
5.88C
51.11 ±
19.75C
-53.33 ±
20.00C
-77.78 ±
5.88C
0.05% 0.00 ± 0.00Aa 82.22 ±
8.01C
60.00 ±
30.55C
80.00 ±
10.18C
73.33 ±
23.41C
97.78 ±
2.22C
80.00 ±
20.00C
62.22 ±
5.88D
53.33 ±
3.85D
0.10% 1.11 ± 1.11Ab 95.55 ±
2.22C
95.55 ±
2.22C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
97.78 ±
2.22E
88.89 ±
11.11E
Control 0.00 ± 0.00A 57.78 ±
35.55C
71.11 ±
22.22C
71.11 ±
25.63C
51.11 ±
36.38C
84.45 ±
9.69C
26.67 ±
10.19C
-55.55 ±
13.52C
-80.00 ±
10.18C
Table 4 (Continue)
1 Mean mortality percentage followed by different letters within the same column are signicant different by subjecting to Kruskal-Wallis H Test at p<0.05: A
& B (comparison of concentration within the same coffee species); a & b (comparison of coffee species within a concentration).
2 Mean repellency percentage followed by different letters within the same column are signicant different by subjecting to Kruskal-Wallis H Test at p<0.05:
C, D & E (comparison of concentration within the same coffee species).

Coffee extracts on household ant’s mortality
1575
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
to the gel and resulted in lowest mortality.
The overall results showed that the highest
concentration, 0.10% of all three coffee
species had the highest mortality against
T. indicum (Figure 1). However, 0.01% of
C. liberica showed the highest mortality
against P. megacephala in the same set of gel
bait as P. megacephala displayed the lowest
mean repellency percentage at 48 h and 72
h (Table 3). For M. pharaonis, the mortality
also showed a merely increasing trend from
0.01% to 0.10% but the differences were
not much noticeable. For instances, 0.05%
and 0.10% of C. arabica had the same
mortality values; 0.01% and 0.05% of both
C. canephora and C. liberica shared the
similar mean mortality (Figure 5). This is
because the mean repellency percentages
for the three concentrations at all hours
range from 95.55% to 100.00% (Table 5),
indicating a very high repellency behaviour
of M. pharaonis.
In Set II bioassay, there were increasing
trends of C. arabica against T. indicum,
P. megacephala and M. pharaonis with
increasing concentrations (Figures 2 and
6). C. canephora and C. liberica showed
that the lowest mortality on T. indicum was
at 0.05% (Figure 2). On the other hand, C.
canephora had the same mean mortality on
P. megacephala at all concentrations (Figure
4) as the mean repellency percentage had
no significant results from 30 min to 24
h (Table 4). The concentration of 0.01%
C. liberica had a slightly higher mortality
on M. pharaonis as the mean repellency
percentages at 30 min, 1 h, 24 h and 72
h were lower than those of 0.05% and
0.10% (Figure 6 and Table 6). The lower
repellency percentage indicated the more
ants attracted to the gel and thus fed on
the gel.
DISCUSSION
Residual spraying and baiting are common
methods in controlling the pest ants. Baiting
is considered a more effective measure as it
is able to eliminate the entire colony through
trophallaxis among the ants (Lee, 2000;
Suiter et al., 1997). The uses of commercial
and synthetic products have known to
create certain issues such as environmental
problem. Recent studies have revealed that
plants act as potential insecticides, such as
essential oil of Pogostemon cablin possess
the insecticidal and repellence properties
against the urban ants (Albuquerque et al.,
2013). Plant secondary metabolites such as
caffeine (1, 3, 7-trimethylxanthine) have
pesticidal activity, anti-feeding properties
and potential to be natural pesticide
(Magalhães, Fernandes, Demuner, Picanco,
& Guedes, 2010).
The overall results indicated T. indicum
was the most susceptible species as it had
the highest mortality among the three ant
species. According to Lee, Lim and Yap
(1996), the erratic movement of crazy ant
allowed it to pick up more insecticidal
materials and thus causing a higher
mortality. This nature behaviour could
also be observed in T. indicum at which
they move rapidly and erratically, in turn
leading to increased foraging activity and
higher chance of picking up the toxicant,
resulting in higher mortality (Lee et al.,
1996). In addition, the very frequent

Xue Li Yeoh, Hamady Dieng and Abdul Haz Ab Majid
1576
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
Table 5
Mean mortality percentage and mean repellency percentage of three different coffees against M. pharaonis in Set I bioassay
Coffee
species
Concentration
(%)
Mean
Mortality
Percentage
(%)1
Mean Repellency Percentage (%)2
30min 1h 2h 4h 8h 24h 48h 72h
Coffea
arabica
0.01% 6.67 ±
1.93Aa
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
0.05% 7.78 ±
2.94Aa
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
97.78 ±
2.22C
0.10% 7.78 ±
6.19Aa
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
Control 0.00 ± 0.00A 97.78 ±
2.22C
95.56 ±
4.44C
95.55 ±
2.22C
93.33 ±
6.67C
97.78 ±
2.22C
91.11 ±
4.44C
86.67 ±
6.67C
91.11 ±
5.88C
Coffea
canephora
0.01% 4.44 ±
1.11Aa
97.78 ±
2.22C
97.78 ±
2.22C
97.78 ±
2.22C
100.00 ±
0.00C
100.00 ±
0.00C
95.55 ±
2.22C
95.56 ±
4.44C
100.00 ±
0.00C
0.05% 4.44 ±
1.11Aa
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
95.56 ±
4.44C
100.00 ±
0.00C
100.00 ±
0.00C
0.10% 5.56 ±
5.09Aa
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
Control 0.00 ± 0.00A 93.33 ±
6.67C
95.55 ±
2.22C
95.56 ±
4.44C
97.78 ±
2.22C
97.78 ±
2.22C
93.33 ±
3.85C
88.89 ±
5.88C
88.89 ±
5.88C

Coffee extracts on household ant’s mortality
1577
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
Coffea
liberica
0.01% 4.44 ±
2.94Aa
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
97.78 ±
2.22C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
0.05% 4.45 ±
2.22Aa
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
0.10% 5.56 ±
1.11Aa
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
97.78 ±
2.22C
100.00 ±
0.00C
Control 0.00 ± 0.00A 100.00 ±
0.00C
97.78 ±
2.22C
93.33 ±
6.67C
93.33 ±
6.67C
93.33 ±
0.00D
86.67 ±
3.85D
86.67 ±
3.85D
93.33 ±
0.00D
Table 5 (Continue)
1 Mean mortality percentage followed by different letters within the same column are signicant different by subjecting to Kruskal-Wallis H Test at p<0.05: A
& B (comparison of concentration within the same coffee species); a & b (comparison of coffee species within a concentration).
2 Mean repellency percentage followed by different letters within the same column are signicant different by subjecting to Kruskal-Wallis H Test at p<0.05: C,
D & E (comparison of concentration within the same coffee species).
Table 6
Mean mortality percentage and mean repellency percentage of three different coffees against M. pharaonis in Set II bioassay
Coffee
species
Concentration
(%)
Mean Mortality
Percentage (%)1
Mean Repellency Percentage (%)2
30min 1h 2h 4h 8h 24h 48h 72h
Coffea
arabica
0.01% 1.11 ± 1.11Aa 100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
0.05% 2.22 ± 2.22Aa 100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
0.10% 5.55 ± 2.22Aa 100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
91.11 ±
8.89C
Control 0.00 ± 0.00A 93.33 ±
6.67C
97.78 ±
2.22C
100.00 ±
0.00C
93.33 ±
6.67C
86.67 ±
13.33C
97.78 ±
2.22C
95.56 ±
4.44C
91.11 ±
8.89C

Xue Li Yeoh, Hamady Dieng and Abdul Haz Ab Majid
1578
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
Coffea
canephora
0.01% 4.44 ± 1.93Aa 100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00
± 0.00C
100.00 ±
0.00C
100.00 ±
0.00C
97.78 ±
2.22C
100.00 ±
0.00C
0.05% 4.44 ± 1.93Aa 100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00
± 0.00C
97.78 ±
2.22C
93.33 ±
3.85C
97.78 ±
2.22C
100.00 ±
0.00C
0.10% 7.78 ± 2.22Aa 100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00
± 0.00C
100.00 ±
0.00C
95.55 ±
2.22C
100.00
± 0.00C
100.00 ±
0.00C
Control 0.00 ± 0.00B 100.00 ±
0.00C
95.56 ±
4.44C
100.00 ±
0.00C
97.78 ±
2.22C
95.56 ±
4.44C
88.89 ±
2.22C
82.22 ±
11.76C
84.45 ±
9.69C
Coffea
liberica
0.01% 4.45 ± 2.22Aa 97.78 ±
2.22C
95.56 ±
4.44C
100.00 ±
0.00C
100.00
± 0.00C
100.00 ±
0.00C
97.78 ±
2.22C
100.00
± 0.00C
95.55 ±
2.22C
0.05% 3.33 ± 1.93Aa 100.00 ±
0.00C
100.00 ±
0.00C
100.00 ±
0.00C
97.78 ±
2.22C
97.78 ±
2.22C
100.00 ±
0.00C
100.00
± 0.00C
100.00 ±
0.00C
0.10% 3.33 ± 3.33Aa 100.00 ±
0.00C
97.78 ±
2.22C
97.78 ±
2.22C
100.00
± 0.00C
100.00 ±
0.00C
100.00 ±
0.00C
100.00
± 0.00C
100.00 ±
0.00C
Control 0.00 ± 0.00A 100.00 ±
0.00C
97.78 ±
2.22C
93.33 ±
6.67C
100.00
± 0.00C
95.56 ±
4.44C
95.56 ±
4.44C
93.33 ±
6.67C
95.55 ±
2.22C
1 Mean mortality percentage followed by different letters within the same column are signicant different by subjecting to Kruskal-Wallis H Test at p<0.05: A &
B (comparison of concentration within the same coffee species); a & b (comparison of coffee species within a concentration).
2 Mean repellency percentage followed by different letters within the same column are signicant different by subjecting to Kruskal-Wallis H Test at p<0.05: C,
D & E (comparison of concentration within the same coffee species.
Table 6 (Continue)

Coffee extracts on household ant’s mortality
1579
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
self-grooming of T. indicum also contributed
to its mortality. Ants perform self-grooming
when they detect microbes or materials that
endanger themselves (Hughes, Eilenberg, &
Boomsma, 2002). They also tend to perform
self-grooming for a longer time when
encounter with more harmful microbes or
materials (Morelos-Juárez, Walker, Lopes,
& Hughes, 2010). During self-grooming, the
ants may ingest the insecticidal-containing
materials. Ants do perform allogrooming,
a process of grooming towards other
individuals by using their shovel-like
mouthparts to remove potential harmful
matters from the body surface (Wilson &
Hölldobler, 2005). From the observation,
T. indicum displayed a high frequency of
allogrooming among them. This behaviour
may cause them to accidentally ingest
the toxicant on the body surfaces of other
individuals, resulting in the highest mean
mortality among the three ant species. On
the contrary to T. indicum, M. pharaonis
performed self-grooming and allogrooming
less frequently, hence, they experienced the
lowest mean mortality.
Trophallaxis is a process of exchanging
regurgitated food among the colony members
and it is very common to be observed among
ants. It also allows the donors and recipients
to gain information upon interaction.
During trophallaxis, the ants share the food
and even the insecticide-impregnated bait
among themselves (Lee, 2000). The high
frequency of trophallaxis displayed by T.
indicum might lead to a higher chance for
the other members to ingest the toxicant
and result in a higher mortality value.
Moreover, high performance of antennation
among T. indicum might also contribute to
its highest mortality. Based on the study
by Hölldobler (1985) on ponerine ants,
antennation functions in social greeting,
recruitment and food solicitation. The
touching the antennae with the others are
known as tactile communication. Both
trophallaxis and antennation play important
role in food distribution and transmission
(Hölldobler, 1985).
The results showed that the mortality
increased with the increasing concentration
(0.01%, 0.05% and 0.10%) for most but
not all bioassays; hence, they were not
concentration-dependent. This was in
contrast with the previous study of effect
of caffeine on tobacco hornworm larvae.
The study indicated a dose-dependent effect
at which higher concentration of caffeine
lead to higher feeding and development
inhibition of the larvae (Nathanson, 1984).
In the present study, for instance, 0.01% of
C. liberica showed the highest mortality
against P. megacephala in Set I bioassay.
This might be due to its lowest concentration
and lowest mean repellency percentage
at 48 h and 72 h indicating the ants were
attracting the most to the bait and fed on
it, thus causing the highest mortality. Other
similar observations had also implied the
higher repellency percentages result in
lowest mortality of P. megacephala in Set I
bioassay and T. indicum in Set II bioassay
at 0.05% of C. liberica. Hence, it can be
concluded that the repellency percentage
is associated with the ant mortality. In
addition, Set II C. canephora bioassay of the
big-headed ants showed almost same or the
same mean mortality at all concentrations
(Figure 4) although there were uctuations

Xue Li Yeoh, Hamady Dieng and Abdul Haz Ab Majid
1580
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
addition, the slow and inactive movement
of Pharaoh ant may also explain their lowest
mortality.
Almost all Set I bioassays showed higher
mortality as compared to Set II bioassay for
all three ant species. T. indicum displayed
feeding preference towards carbohydrate
foods but there was no specic preference
of carbohydrate foods (Chong & Lee, 2006).
According to Lee (2000), most household
ants were attracted most to 20% to 30% of
sucrose solution. As a proven, T. indicum
preferred Set I gel bait that consisted of 20%
sugar solution and assumed to feed more on
the bait, thus resulting in a higher mortality
as compared to that of Set II gel bait. Albeit
there are food preferences in different ant
species, ants do display higher preference
towards the sugar solution when compared
to water as the carbohydrate providing
energy to them (David & Venkatesha, 2013).
This explained that most of the results
showing higher mortality in Set I bioassay.
Among the three coffee species, C.
arabica showed the highest ant mortality on
T. indicum, P. megacephala and M. pharaonis
at almost all concentrations in both Set I and
Set II bioassays. According to Itoyama and
Bicudo (1992), caffeine reduces the mating
frequency, egg-laying capacity, fertility
and longevity of Drosophila prosaltans
(Diptera: Drosophilidae). Further research
showed that caffeine suppressed the feeding
activity of flies and beetles (Pedronel,
Casanova, Ortiz, Henao, & Pelaez, 2007).
However, the data obtained was in contrast
with those previous studies. From our record
on GC-MS analysis (unpublished data), C.
of the mean repellency percentage at 48 h
and 72 h. This occurrence can be explained
by the delay action of the toxicant in the
gel bait, which is one of the important
characteristics of the gel (Knight & Rust,
1991).
The big-headed ants, P. megacephala
are known as one of the worst and highly
invasive ant species. They have a better
ability to discover and exploit food resources
than other native ant species (Callan &
Majer, 2009). This observation was similar
to the results in this present study; at which
P. megacephala showed higher attraction
behaviour towards the bait as they are able
to locate the food attractant better. Though
they were attracted mostly to the bait when
compared to the other two ant species, their
ant mortality is not the highest. According to
Cokendolpher and Francke, (1985), the ant
body size affects their desiccation rate. Ants
with smaller body size possess larger surface
area to volume ratios tend to desiccate faster
(Cokendolpher & Francke, 1985). Both
minor and major workers of P. megacephala
that are larger in size experience a lower
desiccation rate and hence lower mortality.
However, the smaller size of Pharaoh ant
did not show a higher mortality value even
though their sizes are much smaller than
the big-headed ants. The high repellency
behaviour observed in M. pharaonis caused
the lowest mean mortality. This might be
due to the Pharaoh ants showing higher
degree of repellency towards the water
and sugar solution incorporated in the
bait; as according to Fowler et al. (1993),
M. pharaonis prefer high protein food. In

Coffee extracts on household ant’s mortality
1581
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
arabica contained the lowest composition
percentage of caffeine (25.31%) when
compared to C. canephora (44.70%) and C.
liberica (47.30%). C. arabica occupied the
lowest composition percentage of caffeine
but it had the highest mortality against
the ants. This has suggested caffeine may
not be the main compound of Coffea spp.
to cause ant mortality. From the study
on bumblebees, nectar toxins such as
caffeine, quinine, nicotine, amygdalin and
grayanotoxin do not impede the pollination
activity of bumblebees (Tiedeken, Stout,
Stevenson, & Wright, 2014). Same to the
research on honeybees, low concentrations
of the caffeine tend to enhance their
visitation frequency to the solution (Hagler
& Buchmann, 1993). Moreover, bees show
poor acuity and weak sensitivity of detecting
plant toxins in sucrose solution (Tiedeken
et al., 2014). As both ants and bees are
eusocial insects and from the same order
of Hymenoptera, it is possible to assume
that ants possess the same nature with bees,
indicating that caffeine is not the cause of
causing mortality.
C. arabica contained undecane,
hexadecanamide and tetradecanamide which
could not be found in the other two coffee
species. Undecane is a volatile hydrocarbon
compound and also an alarm pheromone that
can be found in the ants (Lenz, Krasnec, &
Breed, 2013). Regnier and Wilson (1969)
reported that a minute amount of undecane
caused some ant species move rapidly.
Undecane had also shown attraction and
excitement in the workers of the crazy
ants (Witte, Attygalle, & Meinwald, 2007).
Another outcome had demonstrated this
alarm pheromone allow recruitment of
workers to the disturbance region. The
high volatility of undecane improves the
rate of spreading of this compound to
the surrounding, increasing the activity
of the ants (Lenz et al., 2013). Based on
these evidences, undecane is the possible
compound in coffee that attracts ants,
increases their movement and activities.
This can also be proven that the repellency
percentages in most C. arabica bioassays
are lower. However, the possibility of
undecane to be the compound in coffee that
causes ant mortality is yet to be known.
Many studies revealed the effectiveness
of coffee in controlling insects. Caffeine was
known to block the larval development of
Aedes aegypti (Diptera: Culicidae) and cause
lethal effect. The effect is dose-dependent as
the higher the concentration of the caffeine,
the faster the blockade of larval development
(Laranja et al., 2003). Caffeine also impedes
the oviposition activity and drags the
appearance of developmental stages in
the life cycle of the tea shot-hole borer
beetle, Euwallacea (= Xyleborus) fornicates
(Coleoptera: Scolytidae). Nevertheless,
there was no observed lethal outcome on
the beetle (Hewavitharanage et al., 1999).
From a recent study on leaf-cutting ants
Atta sexdens rubropilosa (Hymenoptera:
Formicidae) by Miyashira, Tanigushi,
Gugliotta and Santos (2012), caffeine had
no signicant effect on their survival but
decreased the growth rate of the mutualistic
fungus of leaf-cutting ants. The fungal
growth rate decreased with the increasing

Xue Li Yeoh, Hamady Dieng and Abdul Haz Ab Majid
1582
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
concentration of caffeine. The mutualistic
fungus acts as the only food source for
the immature stages of leaf-cutting ants at
which they require glycogen-rich diet for
development. The adults obtained nutrients
from the decomposition process of plant
tissue by the fungus while the fungus
gained benet from the competition free
circumstances with other microorganisms.
The symbiotic relationship between the leaf-
cutting ants and fungus had demonstrated
the ants might be associated with caffeine
toxicity to the fungus but not affected
directly by the caffeine (Miyashira et al.,
2012). According to this study, again, it can
be concluded caffeine has little or no direct
lethal effect on the ants, therefore matching
the data obtained in this research.
Nevertheless, the chemical composition
of the coffee compounds may vary
dependind on the roasting temperature
and time. For instance, the coffee roasted
at higher temperatures for a shorter period
display higher acidity, more soluble solids
and a different volatile profile while
comparing with coffee that roasted at a
lower temperature with longer period of
time (Farah, 2012). Therefore, it is very
crucial to have standardized and constant
roasting temperature and time to obtain a
reliable chemical composition percentage
of the compounds while comparing the
coffee species.
Basically, the higher concentration
of caffeine had led to a higher repellency.
Honey bees were less likely to consume the
sucrose solution with high dose of caffeine
(Mustard, Dews, Brugato, Dey, & Wright,
2012). Nonetheless, this phenomenon
was not observed in this study. The
concentrations, 0.01%, 0.05% and 0.10%
used were based on the study of Miyashira
et al. (2012). The repellency behaviours of
T. indicum and M. pharaonis towards the
three concentrations were similar. While
P. megacephala showed some degree of
signicant results at only 8, 24, 48 and 72 h,
at which higher repellency was observed at
higher concentration. This is probably due
to the highest concentration, 0.10% used in
this research was considered low to deter
the feeding of ants. It was shown that 0.30%
to 10% of caffeine suppressed the feeding
activity and growth of tobacco hornworm
larvae, Manduca sexta (Lepidoptera:
Sphingidae) (Nathanson, 1984). From the
results obtained and according to these
previous studies, coffee that consists of
various volatile and non-volatile compounds
has the potential to act as a repellent for
ants. Minor compounds may act as potent
synergists to increase the impact of major
compounds. Therefore, future research
can be performed to reinforce the caffeine
repellency effect on ants by using a higher
concentration.
CONCLUSION
In conclusion, all the ant species displayed
slightly higher mortality in the bioassay
with bait containing sugar attractant. Coffee
with low concentration was not effective
in killing the household ants because the
mortalities obtained after three days did not
exceed 50%. Therefore, future researches
to test on the lethal effect of coffee can
be studied by using higher concentration.
Moreover, the potential of coffee to be
formulated as ant repellent cannot be ruled
out as the overall results showed a great

Coffee extracts on household ant’s mortality
1583
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
extent of repellency towards the baits.
The natural behaviour of the ant had also
contributed to their mortalities. T. indicum
with higher frequency of trophallaxis, self-
grooming, allogrooming and antennation
had the highest mortality among the three
ant species tested. A further study could be
conducted to test the coffee effect on the ant
colony instead of only on the ant workers.
Furthermore, C. arabica had the best impact
on the ant mortality. Hence, a focus study
on the effect of this coffee species on ants
should be carried out in the future.
ACKNOWLEDGMENTS
The authors would like to acknowledge
Ministry of Higher Education (MOHE) for
funding the research under Fundamental
Research Grant (FRGS) (FRGS: 203 /
PBIOLOGI / 6711360).
REFERENCES
Ab Majid, A. H., Dieng, H., Ellias, S. S., Sabtu,
F. S., Abd Rahim, A. H., & Satho, T. (2018).
Olfactory behavior and response of household
ants (Hymenoptera) to different types of coffee
odor: A coffee-based bait development prospect.
Journal of Asia-Pacific Entomology, 21(1),
46–51.
Abdullah, F., Subramanian, P., Ibrahim, H., Malek,
S. N. A., Lee, G. S., & Hong, S. L. (2015).
Chemical composition, antifeedant, repellent,
and toxicity activities of the rhizomes of galangal,
Alpinia galanga against Asian subterranean
termites, Coptotermes gestroi and coptotermes
curvignathus (Isoptera: Rhinotermitidae).
Journal of Insect Science, 15(1), 7.
Albuquerque, E. L. D., Lima, J. K. A., Souza, F. H. O.,
Silva, I. M. A., Santos, A. A., Araújo, A. P. A.,
… Bacci, L. (2013). Insecticidal and repellence
activity of the essential oil of Pogostemon cablin
against urban ants species. Acta Tropica, 127(3),
181–186.
Beatson, S. (1972). Pharaoh’s ants as pathogen vectors
in hospitals. The Lancet, 299(7747), 425–427.
Callan, S. K., & Majer, J. D. (2009). Impacts of an
incursion of African big-headed ants, Pheidole
megacephala (Fabricius), in urban bushland in
Perth, Western Australia. Pacic Conservation
Biology, 15(2), 102–115.
Campos-Farinha, A. E. (2005). Urban Pest Ants of
Brazil (Hymenoptera: Formicidae). In C.Y. Lee
& W. H. Robinson (Eds.), Proceedings of the
Fifth International Conference on Urban Pests
(pp. 1–4). Suntec, Singapore: Perniagaan Ph'ng
@ P&Y Design Network, Malaysia.
Chong, K. F., & Lee, C. Y. (2006). Food preferences
and foraging activity of field populations of
a pest ant, Tapinoma indicum (Hymenoptera:
Formicidae). Sociobiology, 48(3), 875–883.
Cokendolpher, J. C., & Francke, O. F. (1985).
Temperature preferences of four species of re
ants (Hymenoptera: Formicidae: Solenopsis).
Psyche (New York), 92(1), 91–101.
David, A. I., & Venkatesha, M. G. (2013). Attraction
of household ants (Hymenoptera: Formicidae)
to various food sources in different seasons.
Journal of Entomology, 10(2), 66–75.
Eow, A. G. H., Chong, A. S. C., & Lee, C. Y. (2004).
Colonial growth dynamics of tropical urban
pest ants, Monomorium pharaonis, M. oricola
and M. destructor (Hymenoptera: Formicidae).
Sociobiology, 44(2), 365–377.
Farah, A. (2012). Coffee constituents. Coffee:
Emerging Health Effects and Disease Prevention,
1, 22–58.

Xue Li Yeoh, Hamady Dieng and Abdul Haz Ab Majid
1584
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
Fowler, H. G., Filho, F. A., & Bueno, O. C. (1993).
Seasonal space usage by the introduced
Pharaoh’s ant, Monomorium pharaonis (L.)
(Hym., Formicidae), in institutional settings
in Brazil and its relation to other structural
ant species. Journal of Applied Entomology,
115 (1–5), 416–419.
Gooch, H. (1999). Baiting remains the treatment of
choice. Pest Control, 67(9), 40–43.
Hagler, J. R., & Buchmann, S. L. (1993). Honey bee
(Hymenoptera: Apidae) foraging responses to
phenolic-rich nectars. Journal of the Kansas
Entomological Society, 66(2), 223–230.
Hewavitharanage, P., Karunaratne, S., & Kumar, N.
S. (1999). Effect of caffeine on shot-hole borer
beetle (Xyleborus fornicatus) of tea (Camellia
sinensis). Phytochemistry, 51(1), 35–41.
Higure, Y., & Nohmi, M. (2002). Repetitive application
of caffeine sensitizes caffeine-induced Ca2+
release in bullfrog sympathetic ganglion neurons.
Brain Research, 954(1), 141-150.
Hölldobler, B. (1985). Liquid food transmission and
antennation signals in ponerine ants. Israel
Journal of Entomology, 19, 89-99.
Hughes, W. O., Eilenberg, J., & Boomsma, J. J.
(2002). Trade-offs in group living: transmission
and disease resistance in leaf-cutting ants.
Proceedings of the Royal Society of London B:
Biological Sciences, 269(1502), 1811–1819.
Itoyama, M. M., & Bicudo, H. E. M. C. (1992). Effects
of caffeine on fecundity, egg laying capacity,
development time and longevity in Drosophila
prosaltans. Revista Brasilliensis de Genetica,
15(2), 303–321.
Jaramillo, J., Borgemeister, C., & Baker, P. (2006).
Coffee berry borer Hypothenemus hampei
(Coleoptera: Curculionidae): searching for
sustainable control strategies. Bulletin of
Entomological Research, 96(3), 223–233.
Jenkins, M. (2001). Battling a common enemy. Pest
Control, 69(9), S10-S12, S16, S27.
Kaminski, L. (2000). Ants are still tops, for now. Pest
Control, 68(9), S9-S11.
Klotz, J. H., Greenberg, L., Shorey, H. H., & Williams,
D. F. (1997). Alternative control strategies for
ants around homes. Journal of Agricultural
Entomology, 14(3), 249–257.
Knight, R. L., & Rust, M. K. (1991). Efcacy of
formulated baits for control of Argentine
ant (Hymenoptera: Formicidae). Journal of
Economic Entomology, 84(2), 510–514.
Laranja, A. T., Manzatto, A. J., & de Campos
Bicudo, H. E. M. (2003). Effects of caffeine
and used coffee grounds on biological features
of Aedes aegypti (Diptera, Culicidae) and their
possible use in alternative control. Genetics and
Molecular Biology, 26(4), 419–429.
Lee, C. Y. (2000). Performance of hydramethylnon-
and pronil-based containerized baits against
household ants in residential premises. Tropical
Biomedicine, 17(1), 45–48.
Lee, C. Y., Lim, C. Y., & Yap, H. H. (1996). Contact
toxicity of bendiocarb against three species
of Malaysian household ants (Hymenoptera:
Formicidae). Journal of Biosciences, 7(1),
79–82.
Lenz, E. L., Krasnec, M. O., & Breed, M. D.
(2013). Identication of undecane as an alarm
pheromone of the ant Formica argentea. Journal
of Insect Behavior, 26(1), 101–108.
Lucky, A. (2009). Urban ants of North America
and Europe: Identification, biology, and
management-edited by John Klotz, Lauren
Hansen, Reiner Pospischil and Michael Rust.
Systematic Entomology, 34(2), 406–407.
Magalhaes, S. T. V., Fernandes, F. L., Demuner, A. J.,
Picanço, M. C., & Guedes, R. N. C. (2010). Leaf
Alkaloids, phenolics, and coffee resistance to the

Coffee extracts on household ant’s mortality
1585
Pertanika J. Trop. Agric. Sci. 41 (4): 1557-1586 (2018)
leaf miner Leucoptera coffeella (Lepidoptera:
Lyonetiidae). Journal of Economic Entomology,
103(4), 1438-1443.
Marer, P., & Flint, M. (1991). Residential, industrial
and institutional pest control (2nd ed.).
Oakland, California: University of California,
Statewide Integrated Pest Management Program,
Agriculture and Natural Resources.
Mgbemena, I. C., Ebe, T., Nnadozie, A.I., &
Ekeanyanwu, K.K. (2015). Repellent activities of
the methanolic leaf extracts of Moringa oleifera
and Stachytarpheta indica against Aedes aegypti
mosquito. Journal of Pharmacy and Biological
Sciences, 10(4), 77-81.
Miyashira, C. H., Tanigushi, D. G., Gugliotta,
A. M., & Santos, D. Y. (2012). Influence of
caffeine on the survival of leaf-cutting ants Atta
sexdens rubropilosa and in vitro growth of their
mutualistic fungus. Pest Management Science,
68(6), 935–940.
Morelos-Juárez, C., Walker, T. N., Lopes, J. F. S., &
Hughes, W. O. H. (2010). Ant farmers practice
proactive personal hygiene to protect their fungus
crop. Current Biology, 20(13), R553-R554.
Mustard, J. A., Dews, L., Brugato, A., Dey, K., &
Wright, G. A. (2012). Consumption of an acute
dose of caffeine reduces acquisition but not
memory in the honey bee. Behavioural Brain
Research, 232(1), 217–224.
Na, J. P. S., & Lee, C.Y. (2001). Identication key to
common urban pest ants in Malaysia. Tropical
Biomedicine, 18(1), 1-17.
Nathanson, J. (1984). Caffeine and related
methylxanthines: possible naturally occurring
pesticides. Science, 226(4671), 184–187.
Pedronel, A., Casanova, H., Ortiz, C., Henao, B.,
& Pelaez, C. (2007). Insecticidal activicity of
caffeine aqueous solutions and caffeine oleate
emulsions against Drosophila melanogaster and
Hypothenemus hampei. Journal of Agricultural
and Food Chemistry, 55(17), 6918-6922.
Regnier, F. E., & Wilson, E. O. (1969). The alarm-
defence system of the ant Lasius alienus. Journal
of Insect Physiology, 15(5), 893–898.
Satho, T., Dieng, H., Ahmad, M. H. I., Ellias, S.
B., Hassan, A. A., Abang, F., … Nolasco-
Hipolito, C. (2015). Coffee and its waste repel
gravid Aedes albopictus females and inhibit the
development of their embryos. Parasites and
Vectors, 8(1), 272.
Smith, S. E., McElhatton, P. R., & Sullivan, F. M.
(1987). Effects of administering caffeine to
pregnant rats either as a single daily dose or
as divided doses four times a day. Food and
Chemical Toxicology, 25(2), 125–133.
Suiter, D., Wu, D., & Bennett, G. (1997). The
evolution of ant control. Pest Control, 65, 46–51.
Tiedeken, E. J., Stout, J. C., Stevenson, P. C., &
Wright, G. A. (2014). Bumblebees are not
deterred by ecologically relevant concentrations
of nectar toxins. Journal of Experimental
Biology, 217(9), 1620–1625.
Williams, D. F. (1989). An improved articial nest
for laboratory rearing of the imported re ant,
Solenopsis invicta (Hymenoptera: Formicidae).
Florida Entomologist, 72(4), 705–707.
Wilson, E. O., & Holldobler, B. (2005). The rise
of the ants: A phylogenetic and ecological
explanation. Proceedings of the National
Academy of Sciences, 102(21), 7411–7414.
Witte, V., Attygalle, A. B., & Meinwald, J. (2007).
Complex chemical communication in the
crazy ant Paratrechina longicornis Latreille
(Hymenoptera: Formicidae). Chemoecology,
17(1), 57–62.