ArticlePDF Available

FAILURE OF THE "MOSQUITO PLANT", PELARGONIUM X CITROSUM'VAN LEENII" TO REPEL ADULT AEDES ALBOPICTU^S AND CULEX QUTNQUEFASCIATUS IN FLORIDA

Authors:
  • Navy Entomology Center of Excellence
Journal of the American Mosquito Control Association, 10(4)14'73-476,
1994
Copyright @ 1994
by the American Mosquito Control Association,
Inc.
FAILURE OF THE "MOSQUITO PLANT",
PELARGONIUM X CITROSUM'VAN LEENII"
TO REPEL ADULT AEDES ALBOPICTU^S AND
CULEX QUTNQUEFASCIATUS IN FLORIDA
J. E. CILEK exp E. T. SCHREIBER
rohnA**i:;;f
r,::rt:,';:i:*"7';;i::#;!:,i.';l;frYUniversitv'
ABSTRACT. The efrcacy of the "mosquito plant", Pelargonium x citrosum
'van Leenii', as an area-
wide repellent against adult host-seeking I edes albopictus and Culex quinquefasciatus
females was eval-
uated. No significant differences
(P > 0.05) were observed in the number of mosquitoes landing on the
forearms
of human subjects in locations where
plants were present
compared
with areas
without plants.
In laboratory cage trials, more Cx. quinErcfasciatus adults rested on excised leaves of this cultivar
compared side-by-side with similar size and shape white paper
leaf models.
INTRODUCTION
A variety of botanical substances have been
evaluated
for their repellency
against adult mos-
quitoes
(Christophers
1947, Dethier 1947,
Suk-
umar et al. 1991). In many
instances
these stud-
ies used
"essential
oils" that were extracts from
whole or parts of plants.
One of the earliest well-
known mosquito repellents (being used as early
as 1882)
is the essential oil ofcitronella
(Dethier
1947). This substance is currently available in
several
commercially formulated repellents.
As
a secondary
plant compound,
citronella was
orig-
inally extracted from the citronella or nardus
grass,
Cymbopogon
nardus Linn., and found to
be composed of a variety of acyclic monoter-
penes
including borneol, geraniol, citronellol, and
citronellal (Christophers 1947, Dethier 1947\.
Some of these compounds are common repel-
lents found in the alarm substance
produced by
some
ant species
(Rodriguez
and kvin 1976).
We have seen a variety of garden
catalogs
and
businesses from the USA and Canada advertis-
ing a "mosquito plant" also sometimes
com-
monly referred
to as "citrosa plant" or "citrosa
geranium". These
common names refet to Pel-
argonium x citrosum'van Leenii' Voigt ex T.
Sprague
(family: Geraniaceae),
a plant that pro-
duces citronella oil and that is claimed to "be
the only effective
mosquito repellent
plant in the
world". This plant is a cross of an African lemon
geranium
(P.
crispum)
:drthan
English fingerbowl
geranium (P. x limoneum) hybid that further
incorporated
tissue cultures
ofa southern
Asiatic
grass
that produced
citronella oil. This resultant
cultivar has
been attributed to the efforts of Dutch
horticulturist, Dirk Van Leenen. Mosquito re-
pellency of the 'van Leenii' cultivar is implied
by vaporization ofplant volatiles (i.e.,
citronella
oil) into the surrounding environment and is
claimed
to provide a barrier ofprotection against
host-seeking
mosquitoes
within a 3-m-diam area.
Our study was initiated to evaluate,
in field and
laboratory trials, the repellency
ofP. x citrosum
'van Leenii', in north Florida against adult host-
seeking females
of Aedes albopictus
(Skuse)
and
Culex quinquefasciatus
Say, 2 common urban
pest mosquitoes.
MATERIALS AND METHODS
Field tests
were conducted
in a 13.9 x 7.2 x
2.6-m screened
enclosure at the John A. Mul-
rennan, Sr. Research
Laboratory, Panama
City,
FL. At l4-d intervals, l0 P. x citrosum'van
Leenii'potted plants
were
placed
0.5 m apart in
one end of the enclosure in a 1.6-m-diam circle
30 min before each study began to allow plant
volatiles to equilibrate in the immediate area
for
trials l-5. At weeks 14 through 18 (trials 6-8),
plants
remained
in the test location for 24 h prior
to mosquito release
and collections. An area
with
no plants,
9 m from the
nearest'van
Leenii'cul-
tivar (i.e.,
at the other end of the enclosure), was
designated as
a control. Fifteen minutes later (af-
ter plants had been set out), approximately 500
Ae. albopictus and 1,000 Cx. quinquefasciatus
laboratory-reared 5-6-day-old nonbloodfed fe-
males
were released
in the middle ofthe screened
enclosure
before collections commenced. After
the 30-min equilibration time 2 people sat in
chairs, each
at opposite
ends ofthe enclosure in
the middle of each
test area and mosquitoes
that
landed
on theirbare forearms were
removed with
an aspirator (Hausherr's
Machine Works, Toms
River, NJ). Four 5-min collection periods ex-
posed
each
individual twice at each
location (i.e.,
plant vs. control). After these collections,
plants
were switched to the control area and the pre-
vious "treatment" area then became a control.
Thirty minutes was allowed to elapse before
473
474 Joururar or rxp Ar"reruclN
MosQuro Covrnor, Assocr,r^nox No.4
Table 1. Total number of Aedes albopictus and Culex quinquefascialus
females collected
from
the forearms of 2 human volunteers in areas with and without Pelargonium x citrosum'van
kenii', moming tests.
Control
Vor-.
Planl
Trial Cx. quinque-
Ae. albopictus fasciatus Ae. albopictus Cx. quinque-
fasciatus l-valuer
I
2
J
4
6
7
8
Total
42
89
I5
22
104
ll
286
627
39
134
105
29
ll0
22
239
678
1.18
2.O4
0.67
o.2l
0.56
t.l2
0.71
4
I
5
l8
44
0
0
72
0
I
l3
t57
0
0
178
I All r-values
not significant,
P > 0.05.
starting
the next set of mosquito collections.
This
allowed any volatiles in the previous treatment
area to dissipate
and volatiles in the new treat-
ment area
with plants to equilibrate.
All mosquitoes
collected were identified and
the number aspirated
from each
person
was re-
corded at each time interval. Because of host-
seeking
differences
between the 2 species,
the
study was conducted during early morning and
late evening crepuscular
periods. No wind was
felt by the investigators during testing.
Between
test
intervals
plants
were removed
from the study
area and placed in a 2nd screened
enclosure,
away
from the study site, similar to the first. The study
was conducted
from August 5 through Novem-
ber 5, I 993. Heights ofeach plant were
recorded
on the day of testing.
Temperature
during morn-
ing tests
averaged
21.3 ! I .0"C and for evening
tests
26.9 + 0.8oC.
Our study used
a cross-over
experimental
de'
sign similar to that used by Schreiber
et al. ( l99l)
utilizing the statistical analysis
of Cochran and
Cox (1957). This design recognized 2 main
sources
ofvariation: l) bloodfeeding
activity of
host-seeking female mosquitoes fluctuates
through time, and 2) rate of attack varied from
human subject to human subject.
Briefly, 2 in-
dividuals (A, B) were subjected
to treatments
C
and P (control [i.e., no plant] and plant, respec-
tively) at 2 locations (R, S). The 4 5-min test
periods exposed each individual to each treat-
ment in each
location using the following order
of rotation: C-R, P-R, C-S,
R-S for subject
A and
P-S, C-S, P-R, C-R for subject
B. Differences
in
the total number ofmosquitoes
collected
for each
species
in control (no plant) vs. plant sites
were
determined
using
a l-test. Differences
were con-
sidered
significant
at P > 0.05.
Laboratory cage
t€sts of excised
P. x citrosum
'van lrenii' leaves
were conducted
to evaluate
short-range
repellency to adult Cx. quinquefas-
ciatus.
On the day before
testing,
approximately
100 5-day old adult females
were
placed
in each
of 4 42 x 42 x 42-cm screened
cages
having 2
opposite sides of clear Plexiglas@.
One mature
leaf
(leaf
area
: 183.9
+ 24.4
cm2)
was
excised
from each
of4 plants and the severed
end ofthe
petiole sealed
with paraffin wax. The upper sur-
face of each leaf was taped to a l5-cm-diam
Whatman #4 filter paper
disc and fastened
to the
screen
surface
at the back ofeach cage.
A similar
shaped
and sized
filter paper model (designated
as a control), to mimic the'van Leenii'leaf, had
been
previously cut out ofthe same
quality filter
paper,
taped
to another l5-cm-diam filter paper
disc and immediately fastened to the back of the
cage.
Excised
leaves and paper models were ap-
proximately I 2 cm from each other on the screen
surface
in the same
cage
and were about 38 cm
from the cage
floor. In previous tests,
no difer-
ence in the number of mosquitoes landing on
green vegetable-dyed filter paper leaves placed
on a l5-cm-diam filter paper
disc
occurred
when
compared with similarly presented uncolored
(white) filter paper leaves.
Therefore,
uncolored
filter paper
served
as
a control. Total number of
mosquitoes
resting on leaves and
controls
in each
cage
were recorded
every half hour (at 3 l-min
intervals) for 8 h. All plants for this study were
obtained locally from a commercial hardware/
garden
shop.
RESULTS AND DISCUSSION
The total number of mosquitoes,
regardless
of
species,
that landed on persons
in the screened
enclosure
was not significantly
lower in locations
where P. x citrosum'van Leenii' was present
DscEMsBR.
1994 Ps
ztneoNtu*t x crrRosuM Anre-Wrpr Rerer-r,sNcy
Table 2. Total number of Aedes
albopictus and Culex quinquefascialas
females collected from
the forearms of 2 human volunteers in areas
with and without Pelargonium x citrosum'varr
Leenii', evening tests.
Control Plant
475
Trial Cx. quinque-
Ae. albopictus fasciatus Cx. quinque-
Ae. albopictus fasciatus /-valuet
I
2
J
4
5
6
7
8
Total
42
22
58
2l
l5
104
l3
0
275
l8
8
139
l0l
105
44
426
r23
964
27
l8
77
l9
ll
ll0
l0
2
274
66
9
ll6
187
232
157
186
294
|,247
0.40
0.13
r.29
l.4l
1.09
0.56
0.18
3.65
'All t-values not sigrificant, P > 0.05.
compared with controls in morning (Table l) or
evening
tests
(Table
2). Literature accompanying
commercial
advertisements
of this cultivar states
that one plant should be able to provide enough
volatile chemicals to repel mosquitoes
within a
3-m2
area. Our study used l0 times the number
ofplants in an area ofthis size with no significant
reduction in the number of host-seeking
mos-
quitoes. In addition, plant height (i.e., size)
did
not afect repellency
during the study. Plant height
at the beginning
of the tests
averaged 28.7 + 1.9
cm and at the end ofthe test averaged
63.7 +
3.2 cm. Individual area
ofmature leaves
sampled
from these
plants
ranged
from 133.9
to 351.6
cm2
(mean
245.7
+ 13.9
cm,). During field
tests
we observed
that several mosquitoes
had landed
on the plants and used
them for resting
sites. In
laboratory cage tests
a significantly
greater
num-
ber of female
mosquitoes rested
on excised leaves
(8
l) compared with paper leaf model controls
(l
3).
Although the'van Leenii' cultivar smelled like
it emitted citronella, no mosquito repellency
was
observed as a result of volatiles emitted from
these plants. Citronellal (one of the repellent
components in oil of citronella) has been re-
ported to be sufficiently volatile to keep mos-
quitoes
at a distance
(Sarkaria
and Brown I 95 I ).
However, only the purified form of this com-
pound was
used for this observation
and not cit-
ronella-producing plants themselves.
As stated
earlier,
oil ofcitronella (as
extracted
from plants)
contains
a complex of components
of which ge-
raniol and citronellal are the chiefactive agents
(Christophers
| 9
47, D ethier I 9
4 7).
Though con-
sidered as an olfactory deterrent (Garson and
Winnike 1968),
contact or gustatory repellency
may exist. Citronellol, another
component of oil
ofcitronella, has
also been reported to deter ovi-
position by the sweetpotato whitefly (Bemisia
tabaci (Gennadius)) via contact (Butler et al.
l 98e).
Evidently in our study the volatiles that are
produced from P. x citrosum 'van kenii'were
either not intrinsically repellent or not released
in sufficient quantity to evoke a repellent re-
sponse
to create a barrier against host-seeking
Ae. albopictu^s
and Cx. quinquefascialzs
females.
It is quite probable
that the leaves
ofthis cultivar
may have to be crushed
or rubbed on the skin
to release
the volatile components of citronella
oil as
a phytophagous
insect
would release
these
bioactive components
upon feeding.
ACKNOWLEDGMENTS
We thank C. H. Hallmon for his assistance in
rearing mosquitoes
and help in field tests and J.
S. Coughlin and M. A. Olson for their help with
the laboratory tests, plant measurements,
and
maintenance.
Lastly, we thank K. D. Perkins,
University of Florida Herbarium, Gainesville,
for providing identification and additional in-
formation about P. x citrosum 'van Leenii'. We
appreciate W. Opp, J. P. Smith, and G. Alex-
ander for bringing this plant cultivar to our at-
tention. Acknowledgment is extended to those
persons
that contributed critical reviews ofpre-
vious
manuscript
drafts.
REF'ERENCES
CITED
Butler,
G. D.,
Jr.,
D. L. Coudriet
and T.
J.
Henneberry.
1989. Sweetpotato whitefly:
host
plant preference
and repellent
effect
ofplant-derived
oils on cotton.
squash, lettuce
and
cantalope. Southwest.
Entomol.
l4:9-16.
476 JounNlr or rrre Ar'cnrclN Moseurro Coxrnor, AssocllnoN Vor. 10. No. 4
Christophers,
S.
R. 1947. Mosquito repellents.
Being
a report of the work of the mosquito repellent in-
quiry, Cambridge 1943-5. J. Hyg. 45:17 6-231.
Cochran,
W. G. and G. M. Cox. 1957. Experimental
designs.
John Wiley & Sons, Inc., New York.
Dethier, V.D. 1947. Chemical
insect attractants
and
repellents.
The Blakiston Co., Philadelphia.
Garson, L. R. and M. E. Winnike. 1968. Relation-
ships between insect repellency and chemical and
physical parameters-a review. J. Med. Entomol.
5:339-352.
Rodriguez,
E. and D. A. kvin. 1976. Biochemical
parallelisms
of repellents
and attractants
in higher
plants
and arthropods.
Recent Adv. Phytochem.
I 0:
2t4-270.
Sarkaria,
D. S. and A. W. A. Brown. 1951. Studies
on the responses of the femzle Aedes mosquito. Part
II. The action of liquid repellent compounds. Bull.
Entomol.
Res. 42:l 15-122.
Schreiber,
E. T., T. G. Floore and J. P. Ruff. 1991.
Evaluation of an electronic mosquito repelling de-
vice with notes on the statistical
test.
J. Fla. Mosq.
Control
Assoc. 62:37-40.
Sukumar, K., M. J. Perich and L. R. Boobar.
Botanical derivatives
in mosquito control: a review.
J. Am. Mosq.
Control Assoc.
T:2lO-237.
ResearchGate has not been able to resolve any citations for this publication.
Chapter
In reviewing the distribution of micromolecular constituents (secondary compounds) that are known to function as defensive repellents* in arthropods, it becomes evident that a large number of chemical communicatory substances are also common metabolites of higher plants. This remarkable parallelism in inter and intra-specific chemical communication has been noted by numerous biologists, with Eisner (1970) commenting that “the very fact that plants possess the same materials that in other organisms are known to be defensive, may in itself be considered to be circumstantial evidence in support of the latter view.”
Article
A number of liquid mosquito repellents were assessed for vapour repellency power in an olfactometer mounted in a very large cage filled with females of Aëdes aegypti . They were also tested for their knockdown power in fumigation bottles. Their vapour pressures were determined by the Ramsay-Young method. All the liquids showed vapour repellency, and in 39 out of the 42 tested this effect was highly significant. The highest vapour repellency ratings were shown by compounds already known to be the most effective repellents. Although the more volatile compounds such as citronellal tend to show the highest repellency ratings, nevertheless compounds of low vapour pressure such as indalone, DMP and isobornyl morpholinoacetate may also show high vapour repellency. It is concluded that vapour repellency, although in the first instance dependent upon volatility, can vary independently of vapour pressure, so that compounds may be found which afford not only a long protection period due to their nonvolatility, but also a high vapour repellency due to the potency of the comparatively few molecules that are volatilised. The vapours of most of the repellents were found to induce knockdown of mosquitos, but there was no correlation between the speed of this process and the vapour repellency of the compounds.
Article
A review on the reported uses of chemicals derived from botanical sources is presented, along with the part of the plant used for extraction, the mosquito species studied and the bioactivity observed for 344 plant species. Examples of phytochemicals evaluated against mosquitoes as general toxicants, growth and reproduction inhibitors, repellents and ovipositional deterrents are given. The effects of mosquito species and life stage specificity, solvents used for extraction, phototoxic activity and the geographical source from where the plant compounds are derived are discussed.
Article
The pertinent literature has been reviewed from January 1940 through October 1967. The relationships between molecular constitution and insect repellency are described as well as the role of such physical parameters as boiling point and vapor pressure, molecular weight, partition coefficient, and concentration effect. The desirable properties of an ideal repellent substance are summarized, and definitions of repellency relative to this discussion are presented.
Evaluation of an electronic mosquito repelling device with notes on the statistical test
  • E T Schreiber
  • T G Floore
  • J P Ruff
Schreiber, E. T., T. G. Floore and J. P. Ruff. 1991. Evaluation of an electronic mosquito repelling device with notes on the statistical test. J. Fla. Mosq. Control Assoc. 62:37-40.
Chemical insect attractants and repellents. The Blakiston Co
  • V D Dethier
Dethier, V.D. 1947. Chemical insect attractants and repellents. The Blakiston Co., Philadelphia.
Sweetpotato whitefly: host plant preference and repellent effect ofplant-derived oils on cotton. squash, lettuce and cantalope
  • Ref'erences Cited
  • G D Butler
  • D L Jr
  • T J Coudriet
  • Henneberry
REF'ERENCES CITED Butler, G. D., Jr., D. L. Coudriet and T. J. Henneberry. 1989. Sweetpotato whitefly: host plant preference and repellent effect ofplant-derived oils on cotton. squash, lettuce and cantalope. Southwest. Entomol. l4:9-16.