Content uploaded by Christopher Brown
Author content
All content in this area was uploaded by Christopher Brown on Feb 12, 2019
Content may be subject to copyright.
Article/Artikel
THE EFFICACY OF ULTRASONIC PEST
TICKS
C R BRO$7N* and B D LE$7IS*
CONTROLLERS FOR FLEAS AND
INTRODUCTION
Ultrasound generally refers to high fre-
quency sound inaudible to the human ear
(above approximately 20 kHz). Although
inaudible to humans, some insects are
capable of detecting ultrasound. In par-
ticular, some moths respond, by evasion,
to ultrasound in the 20-40 kHz range, the
range used for prey detection by many in-
sectivorous batst r3. Such observations
provided an early stimulus for in-
vestigating the use of ultrasound for the
control of agricultural insect pests.
Results of field trials, mainly on cotton
bollworffi, tobacco budworm and cabbage
looper moths, are conflicting, some show-
ing promise6 12 and others no effect at allt
2 . In contrast, there is no a priori reason to
suggest that ultrasound will be effective
in repelling other insects, in particular
common household pests (rnainly
cockroaches and fishmoths) and pests of
domestic pets (fleas and ticks). There is
Iittle evidence that domestic insect pests
have receptors capable of detecting ultra-
sound, although fleas may be capable of
detecting ultrasonic frequencies in the
region of 100 to 10 000 kHzi. This is far
*Department of Zoology and Entomology, Rhodes
University, P.O. Box 94, 6140 Grahamstown,
Republic of South Africa
Received: April 1991 Accepted: June 1991
110
above the 20-60 kHz output of commer-
cial ultrasonic pest repellers. Never-
theless, the idea of non-chemical control
of household pests is an attractive one,
and a wide range of ultrasonic pest con-
trollers claiming to repel insect pests in
the domestic environment, is available in
the United States and Europe. The eF
ficacy of some of these devices has been
the subject of several investigations, both
in the laboratory4 5 10 14 and under more
natural conditions8 r0 15. Most of these
studies suggest that ultrasonic devices are
ineffective in controlling domestic pest
populations, although there is still some
controversy on the matterT.
It is only relatively recently that
ultrasonic pest controllers have become
available on the South African market,
mainly through mail order companies
advertising in newspapers and magazines.
Subiective reports from purchasers that
these devices are effective, led us to test
the repellent effects of 2 such devices on
fleas and ticks in a choice chamber.
MATERIALS AND METHODS
Adult cat flea s (Crenocephalides felis) were
collected from domestic cats and dogs.
Fleas were either used on the day they
were collected, or kept overnight in a
glass jar with animal hair. Fleas were not
fed and fresh fleas were used for each
trial.
Adult Rhipicephalus simus ticks were
supplied by the Tick Research Unit at
Rhodes University, (Grahamstown,
Republic of South Africa) and sup-
plemented with adult ticks recovered
from domestic dogs in the Grahamstown
area. Because only limited numbers were
available, some ticks were used in more
than one trial, but none more than 3
times over the entire period of the ex-
periments and not in successive trials.
Two ultrasonic devices, purchased
from mail order companies, were tested.
The smaller unit, a flea and tick collar
unit for pets is made in Taiwan but bears
no brand name. It is designed for attach-
ment around the neck of a cat or dog, or
to be placed in a kennel or pet basket.
The instructions claim that the high fre-
quency sound will work by repelling
pests within a range of 4 feet (lr2 m). It
further claims that fleas within this range
will stop jumping within seconds and so
will not iump onto pets fitted with the
device.
The larger unit also carries no brand
name, nor is there any indication of its
country of origin. It is designed for house-
hold use and is powered by two 9V batte-
ries or supplied mains adapter. The speci-
fications claim that the unit sweeps con-
tinuously over 30 to 6 5 kHz, has a sound
pressure level of 130 dB, and is effective
in an area of 2 000 to 2 500 square feet
(180-225 mr). The rate at which the
device sweeps its frequency range is ad-
justable by the user. Both devices were
tested before and after the experiments to
confirm that they were producing ultra-
sound.
The test chamber comprised a
Y-shaped plywood box with a broad base
(16 x 10 x 8 cm) and two narrow arms (30
x 8 x 8 cm). The broad base of the Y was
partly divided by a cardboard baffle, ef-
fectively dividing the chamber into a left
and right side. A 3-piece perspex lid
allowed for observation and easy access to
the chamber. Linen-covered rectangular
holes (6 cmz) cut in the ends of the arms
allowed the ultrasonic devices to be plac-
ed immediately outside the chamber with
their transponders facing into the
chamber. The chamber was lined with
dressmaker's batting, covered with white
linen, to absorb the ultrasound and
restrict it, as far as possible, to one arm of
the chamber. Tests with an ultrasonic bat
detector (QMC Mini Bat Detector)
0038-22809 Tydskr.S.Afr.aet.Ver. (1991) 62(3): 1 10-1 1 3
Table l: Effect of the
chamber. NS pet-collar ultrasound
= not significant unit on the distribution of Ctenocephalides felis in a choice
Trial
No. No. of
fleas Initial distribution
quiet arm unit arm Final distribution
quiet arm unit arm x2 Significance
I
2
3
4
5
6
7
8
l9
2l
28
L7
33
29
20
30
11
T2
5
8
l6
ll6
l6
8
9
23
9
l7
l8
t4
L4
8
9
5
8
l7
5
7
23
l1
l2
23
9
16
24
t3
7
1,349 NS
1,215 NS
0,061 NS
0,059 NS
0,030 NS
4,431 P < 0,05
0,060 NS
5,659 P < 0105
Pooled 197 85 tt2 82 ll5 0,129 NS
established that no ultrasound from the
pet-collar unit penetrated to the end of
ih. "quiet" arrn-of the chamber. With the
large unit in position, ultrasound in the
"quiet" arm of the chamber was still de-
tectable, but was substantially less than in
the arm with the unit, attenuation being
estimated at 80 to 90V0. Trials were car-
ried out in a constant environment room
at 24oC with a l2L l2D light cycle.
Fleas or ticks were introduced into the
chamber at the base of the Y. The
number of insects used varied depending
on availability, but was never fewer than
I I fleas and l0 ticks. Fleas and ticks were
left for at least 60 min to distribute them-
selves in the chamber and their distri-
bution (left or right arm) noted (initial
distribution). The ultrasonic unit being
tested was then placed at the end of one
of the arms of the chamber and switched
on. Initially, the unit was placed at the
left and right arms at random based on
odd (left) and even (right) numbers
generated by a random number
generator. In a second trial of the large
unit with fleas, the unit was sometimes
deliberately placed in the arm containing
the most fleas.
After the unit was switched oo: the
chamber was left undisturbed for 24 h. At
the end of the trial, the number of fleas or
ticks in each side was again counted (final
distribution). Overall, 8 trials were car-
ried out on fleas with the pet-collar unit
and 7 on ticks. Fourteen and 6 trials were
carried out with the large unit on fleas
and ticks, respectively.
Chi-square (X,) tests, corrected for
continuity,u, were carried out for each in-
dividual trial to establish any significant
differences between initial and final dis-
tributions of fleas and ticks in the
chamber and, where X2 values were
homogeneous, pooled X2 were obtained
by summing the initial and final distribu-
tions.
RESULTS
Electronic analysis showed that the pet-
collar unit produced pulsed ultrasound at
a frequency of 35 kHz, giving a 2 mrllise-
cond (ms) tone burst every 40 ms. The
sound pressure level (SPL) of the unit
could not be measured, but output from
the unit was virtually undetectable with
the bat detector at ranges > 30 cm.
The larger unit produced modulating
sound which cycled between 20 and 37
kHz with no break in modulations. SPL
was not measured, but the unit was detec-
table with a bat detector for at least 10 m.
Of 8 trials with the pet-collar unit
against fleas, 6 trials showed no signifi-
cant difference in the distribution of fleas
before and after the unit had been switch-
ed otr: one trial showed a significant
change in distribution towards the unit
and one trial away from the unit (Table
l). Overall, there was no significant
change in the distribution of fleas after 24
h exposure to ultrasound (pooled Xz =
01129, P > 0,50).
Ticks also showed no response to ultra-
sound generated by the pet-collar unit, all
7 trials showing no significant differences
in their initial and final distributions
(pooled Xr=0,006; P > 0,75) (Table 2).
Four out of 6 trials on fleas using the
large unit, showed a significant change in
distribution after 24 h exposure.
However, the movement was towards the
ultrasound (Table 3). Xz values for in-
dividual trials were not homogeneous and
were therefore not pooled. A further
series of 8 trials with substantially more
Table 2: Effect of the pet-collar ultrasound unit on the distribution of Rhipicephalus sitnus in a choice
chamber
Trial
No. No. of
fleas Initial
quiet arm distribution
unit arm Final distribution
quiet arm unit arm Xz Significance
I
2
3
4
5
6
7
20
24
20
20
39
24
20
8
t3
8
9
22
t4
ll
L2
ll
t2
1l
l7
l0
9
10
t2
9
10
2l
t2
10
l0
l2
ll
10
l8
l2
l0
0,469 NS
0,040 NS
0,052 NS
0,273 NS
0,026 NS
0,386 NS
0,05I NS
0038-2809 Jl S.Afr.uet.Ass. (1991) 62(3): 1 10-1 13
82 84 83 0,006 NS
111
Pooled 167 85
fleas showed no significant difference
between initial and final distributions
after 24 h exposure to ultraso,rnd in any
individual trial (Table 4), or overall(Xz =
0,082; P > 0,75).
There was no significant diflerence in
the initial and final distributions of ticks
in any of the 6 individual trials (Table 5)
or overall (72 = 01219; P ) 0150).
strongly suggesting that the devices are
ineffective for repelling fleas and ticks.
These results are consistent with previous
studies on ultrasonic pest controllers. For
example, several studies have shown that
cockroaches are unaffected by a wide
range of ultrasonic frequenciesa 5 e 10.
More specifically, Rust & Parkerl4 found
no movement of fleas away from an
evidence that they are not adversely af-
fected by ultrasound. Similarly cockroach
nymphs have been found inside
ultrasonic pest repellers after trials in
apartment builditrBS, showing that
cockroaches were even using the devices
for harbouragelo.
The claim in the instructions accompa-
nying the pet-collar unit used in the pre-
Table 3: Effect of
chamber the large ultrasound unit on the
during the first series of trials distribution of Ctenocephalides felis in a choice
Trial
No. No. of
fleas Initial
quiet,arm distribution
unit arm Final distribution
quiet arm unit arm X2 Significance
I
2
3
4
5
6
24
34
L7
18
42
11
2l
18
3
9
2l
6
3
16
t4
9
2L
5
8
8
2
I
18
I
16
26
15
l7
24
10
59,520 P < 0,001
10,655 P < 0,001
0,010 NS
12,500 P < 0,001
0,595 NS
7,425 P<0,001
Table 4: Effect of the large
chamber during the ultrasound unit on the
second series of trials distribution of Ctenocephalides felis in a choice
Trial
No. No. of
fleas Initial distribution
quiet arm unit arm Final distribution
quiet arm unit arm x2 Significance
I
2
3
4
5
6
7
8
43
42
39
2l
33
45
37
46
3l
30
30
11
6
6
5
34
t2
L2
9
10
27
39
32
t2
32
33
3l
t6
3
5
2
34
1l
9
8
5
30
40
35
t2
0,029 NS
0,729 NS
0,036 NS
3,866 P < 0102
1,273 NS
o,o4g NS
1,445 NS
0,028 NS
Pooled 306 t53 r53 t56 150 0,082 NS
DISCUSSION
Ultrasonic sound is rapidly attenuated by
distance and is diffracted by solid obf ects.
In the present study, absorption and at-
tenuation was such that either no or very
little ultrasound was present in the
"quiet" arm of the choice chamber. If
ultrasound generated by the devices
repelled insects as claimed, one would ex-
pect a significant movement of fleas and
ticks away from the ultrasonic units into
the sound shadow of the "quiet" arm.
Such movement was not observed,
112
ultrasonic device in a cardboard tube.
Furthermore, Dryden et ai8 and Schein et
alts showed that pet-collar devices were
ineffective in reducing flea numbers on
cats and Schein et alt5 found no difference
between numbers of fleas and ticks initial-
ly placed on dogs with ultrasonic pet-
collars and on control dogs, even after 14
d exposure.
In the present study, fleas and ticks
were observed on the linen at the end of
an arm of the choice chamber within one
cm of the transponder, supporting
sent study that fleas will cease fumping
within seconds of exposure to the collar,
is also unfounded. Fleas in the chamber
were regularly observed to iump and
previous studies have demonstrated that
ultrasound has no eflect on fleas' jumping
or on their normal circadian rhythm of ac-
tivityt I 14. Rust & Parkerla, however,
showed that bursts of CO2 did elicit in-
creased activity, os might be expected of
insects that rely on CO, concentration
and thermal and visual cues to locate
hosts. Ticks in the present study, on the
0038-22809 Tydskr.S.Afr.aet.Ver. (1991) 62(3):1 10-1 13
Table 5: Effect of the large ultrasound unit on the distribution of Rhipicephalus simus in a choice
chamber
Trial
No. No. of
fleas Initial distribution
quiet arm unit arm Final distribution
quiet arm unit arm x2 Significance
I
2
3
4
5
6
20
20
r4
10
30
20
9
10
4
4
18
L2
1l
10
10
6
l2
8
11
10
6
6
T7
10
0,455 NS
o,o5o NS
0,788 NS
0,938 NS
0,035 NS
0,469 NS
Pooled tt4 57 57 60 54 0,219 NS
other hand, showed little movement after
initially distributing themselves in the
chamber, even when they had settled
within one cm of the ultrasonic devices.
Gently exhaling in their vicinity,
however, did elicit movement showing
that they were not immobilised by the
ultrasound.
The leaflet accompanying the large unit
used in the present study also claims that
the unit will stun larger insects such as
moths, bees and cockroaches, rendering
them immobile and allowing them "to be
swept away at leisure". To test this claim,
a single trial with 7 cockroaches
(Periplaneta americana) was carried out.
The trial was conducted as described for
fleas and ticks, but cockroaches were pro-
vided with food and water and a card-
board tube was placed at the end of each
arm of the chamber as harbourage. Ultra-
sound from the large unit had no notice-
able effect on cockroach activity, cock-
roaches at night being especially active
with no signs of immobility. Although
there were too few cockroaches for
statistical purposes, there was also no
change in their distribution after 24 h ex-
posure to ultrasound, but after 48 h all 7
cockroaches were clustered in the tube
immediately in front of the ultrasound
unit, but immediately moved when dis-
turbed.
In addition-to activity, ultrasound has
also been shown to have no effect on
reproduction in either cockroachesto or
fleass to, the latter despite claims that the
use of ultrasonic pet collars inhibit flea
population growthz.
The present study demonstrates that
the 2 ultrasonic devices tested fall short
of claims in their specification and
instruction leaflets with regard to their
performance. Furthermore, the study has
failed to substantiate that these ultrasonic
devices have any efficacy in repelling
common household pests. On the con-
trary, this and other studies have shown
such devices to be ineffective for control-
ling fleas, ticks or cockroaches.
ACKNOWLEDGEMENTS
We thank Prof. B Fwaz of the Tick
Research Unit at Rhodes University for
kindly supplying us with ticks and P E
Hulley and B Fivaz for commenting on
the manuscript. Mr J McKinnel carried
out the electronic tests on the units and
Mrs S Radloff provided statistical advice.
Financial assistance of the Joint Research
Committee of Rhodes University is grate-
fully acknowledged.
REFERENCES
1. Agee H R 1967 Response of acoustic sense cells
of the bollworm and tobacco budworm to ultra-
sound. Journal of Economic Entomology 60:
366-369
2. Agee H R, Webb J C 1969 Ultrasound for con-
trol of bollworms on cotton. Journal of
Economic Entomology 62: 1322-1326
3. Amrine J W, Jarabek M A 1983 Possible
ultrasonic receptors on fleas. Annals of the En-
tomological Society of America 76: 395-399
4. Ballard J B, Gold R E 1983 The response of
male German cockroaches to sonic and ultra-
sonic sound. Journal of the Kansas Entomo-
logical Society 56: 93-96
5. Ballard J B, Gold R E, Decker T N 1984
Response of German cockroach (Orthoptera:
Blattellidae) populations to a frequency sweep-
ing ultrasound-emitting device. Journal of
Economic Entomology 77: 976-979
6. Belton J B, Kempster R H 1962 A field test on
the use of sound to repel the European corn
borer. Entomologia Experimentalis et Applicata
5: 281-288
7. Bonge N J 1990 Opposing viewpoints on the ef-
ficacy of ultrasonic flea collars. Journal of the
American Veterinary and Medical Association
196: 1354-1355
8. Dryden M W, Long G R, Gaafar S M 1989 EF
fects of ultrasonic flea collars on Ctenocephalides
felis on cats. Journal of the American Veterinary
and Medical Association 195: l717-1718
9. Gold T N, Decker T N, Vance A D 1984
Acoustical characterization and efficacy evalua-
tion of ultrasonic pest control devices marketed
for control of German cockroaches (Orthoptera:
Blattellidae). Journal of Economic Entomology
77: 1507-1512
10. Koehler P G, Patterson R S, Webb J C 1986 EF
ficacy of ultrasound for German cockroach (Or-
thoptera: Blattellidae) and oriental rat flea
(Siphonaptera: Pulicidae) control. Journal of
Economic Entomology 79: 1027-1031
ll. Koehler P G, Leppla N C, Patterson R S 1989
Circadian rhythms of cat flea (Siphonaptera:
Pulicidae) locomotion unaffected by ultrasound.
Journal of Economic Entomology 82: 516-518
l2.Payne T L, Shorey H H 1968 Pulsed ultrasonic
sound for control of oviposition by cabbage
looper moths. Journal of Economic Entomology
6l: 3-7
13. Roeder K D 1966 Acoustic sensitivity of the
noctuid tympanic organ and its range for the
cries of bats. Journal of Insect Physiology 12:
843-859
14. Rust M K, Parker R W 1988 Lack of behavioral
response of the cat flea, Ctenocephalides felis
(Siphonaptera: Pulicidae), to a broad spectrum
of ultrasound. Journal of Medical Entomology
25: 144-146
15. Schein E, Gothe R, Hauschild S 1988 Ultra-
schallgerlte gegen Fltlhe und Zecken bei
Hunden und Katzen - nur umweltfreundlich?
Kleintierpraxis 33: 147-149
l6.Zar J H 1974 Biostatistical analysis. Prentis
Hall Inc., New Yersey
0038-2809 Jl S.Afr.aet.Ass. (1991) 62(3): 1 10-1 13 113