Comparison of release mechanisms for botanical oils to control Varroa destructor and Acarapis woodi in colonies of honey bees


Rice, N.D.; Winston, M.L.; Whittington, R.; Higo, H.A.

Journal of Economic Entomology 95(2): 221-226

2002


Two major parasitic pests threaten honey bee populations, the external mite Varroa destructor and the internal mite Acarapis woodi (Rennie). Varroa are beginning to develop resistance to the main chemical defense fluvalinate, and alternative control methods are being pursued. Previous studies have shown that botanical oils, especially thymol, can be effective. Six release devices for either thymol or a blend of botanical oils known as Magic 3 were tested in beehives. The release devices were as follows: (1) low density polyethylene (LDPE) sleeves filled with Magic 3, (2) Magic 3-infused florist blocks, (3) thymol infused florist blocks, (4) a canola oil and thymol mixture wick release, (5) a plastic strip coated with calcium carbonate and Magic 3, and (6) an untreated control. There were significant decreases in varroa levels with the use of Magic 3 sleeves, but brood levels also decreased. Tracheal mite levels significantly decreased with the Magic 3 sleeve treatment, the Magic 3 florist block treatment, and the thymol canola wick treatment. A second experiment showed that changing the location of Magic 3 sleeves in the colony did not detrimentally effect brood levels, but also did not effectively control varroa mites.

APICULTURE
AND
SOCIAL
INSECTS
Comparison
of
Release
Mechanisms
for
Botanical
Oils
to
Control
Varroa
destructor
(Acari:
Varroidae)
and
Acarapis
woodi
(Acari:
Tarsonemidae)
in
Colonies
of
Honey
Bees
(Hymenoptera:
Apidae)
NATHAN
D.
RICE,
MARK
L.
WINSTON,
ROBIN
WHITTINGTON,
AND
HEATHER
A.
HIGO
Department
of
Biological
Sciences,
Simon
Fraser
University,
Burnaby,
BC,
Canada
V5A
1S6
J.
Econ.
Entomol.
95(2):
221-226
(2002)
ABSTRACT
Two
major
parasitic
pests
threaten
honey
bee
populations,
the
external
mite
Varroa
destructor
and
the
internal
mite
Acarapis
woodi
(Rennie).
Varroa
are
beginning
to
develop
resistance
to
the
main
chemical
defense
fluvalinate,
and
alternative
control
methods
are
being
pursued.
Previous
studies
have
shown
that
botanical
oils,
especially
thymol,
can
be
effective.
Six
release
devices
for
either
thymol
or
a
blend
of
botanical
oils
known
as
Magic
3
were
tested
in
beehives.
The
release
devices
were
as
follows:
(1)
low
density
polyethylene
(LDPE)
sleeves
filled
with
Magic
3,
(2)
Magic
3-infused
florist
blocks,
(3)
thymol
infused
florist
blocks,
(4)
a
canola
oil
and
thymol
mixture
wick
release,
(5)
a
plastic
strip
coated
with
calcium
carbonate
and
Magic
3,
and
(6)
an
untreated
control.
There
were
significant
decreases
in
varroa
levels
with
the
use
of
Magic
3
sleeves,
but
brood
levels
also
decreased.
Tracheal
mite
levels
significantly
decreased
with
the
Magic
3
sleeve
treatment,
the
Magic
3
florist
block
treatment,
and
the
thymol
canola
wick
treatment.
A
second
experiment
showed
that
changing
the
location
of
Magic
3
sleeves
in
the
colony
did
not
detrimentally
effect
brood
levels,
but
also
did
not
effectively
control
varroa
mites.
KEY
WORDS
honey
bee,
Varroa
destructor,
Acarapis
woodi,
botanical
oil,
thymol
HONEY
BEES,
Apis
mellifera
(L.)
are
one
of
the
most
economically
important
insects
in
the
world,
partic-
ularly
for
crop
pollination
(Delaplane
and
Mayer
2000).
Honey
bee
populations
are
being
diminished
worldwide
by
mites,
in
particular
Varroa
destructor
(formerly
known
as
Varroa
jacobsoni
Oudemans,
re-
cently
reclassified
by
Anderson
and
Trueman
[2000]
)
and
Acarapis
woodi
(Rennie).
Both
mites
have
spread
throughout
the
United
States
and
much
of
Canada
since
their
introductions
to
North
America
in
the
1980s
(De
Jong
1997,
Wilson
et
al.
1997).
Treatments
to
control
varroa
include
fluvalinate,
formic
acid,
and
more
recently
coumaphos,
now
used
in
the
United
States
under
emergency
license
for
mite
control
in
areas
where
resistance
to
Apistan
(fluvali-
nate)
is
high
(Sanford
et
al.
1999).
Formic
acid
cur-
rently
is
registered
as
a
liquid
in
Canada
and
as
a
gel
formulation
in
the
United
States
(Feldlaufer
et
al.
1997,
Kochansky
and
Shimanuki
1999).
Resistance
to
fluvalinate
first
appeared
in
Europe
in
1992
(Milani
1999),
and
in
Florida
and
South
Dakota
in
1997
(Elzen
et
al.
1998).
Varroa
control
with
fluvalinate
has
dropped
from
the
95-99%
level
to
as
little
as
5.1%
(Elzen
et
al.
1998).
Control
methods
that
do
not
leave
residues
in
bee
products,
as
do
coumaphos,
fluvalinate
and
amitraz,
a
miticide
used
in
Europe,
would
be
preferable
(Thrasyvoulou
and
Pappas
1988,
Balayan-
nis
and
Santas
1992,
Lodesani
et
al.
1992,
Wallner
1995,
Bogdanov
et
al.
1998).
Honey
bees
also
are
parasitized
by
the
internal
tracheal
mite
Acarapis
woodi
(Rennie).
This
mite,
first
described
from
the
Isle
of
Wight
in
1921,
became
established
throughout
most
of
Europe
between
1920
and
1930
(Kjer
et
al.
1989).
It
subsequently
was
dis-
covered
in
Mexico
and
the
United
States
in
1980
and
1984,
respectively
(Delfinado-Baker
1984).
Menthol,
formic
acid,
and
vegetable
oil
patties
can
be
used
to
control
tracheal
mites
(Herbert
et
al.
1987,
Smith
et
al.
1991,
Delaplane
1992,
Sammataro
et
al.
1994).
Colony
loss
is
significantly
higher
with
simultaneous
infesta-
tions
of
varroa
and
tracheal
mites
than
with
infesta-
tions
by
either
mite
alone
(Downey
et
al.
2000).
Se-
lection
of
honey
bees
can
quickly
produce
stock
that
is
resistant
to
the
deleterious
effects
of
tracheal
mites
(Sherman
et
al.
1998,
De
Guzman
et
al.
1998).
Experiments
involving
botanical
oils
have
shown
variable
success
in
both
varroa
and
tracheal
mite
con-
trol,
using
diverse
application
methods
(Marchetti
et
al.
1984;
Cohn
1990;
Calderone
et
al.
1991,
1997;
Smith
et
al.
1991;
Delaplane
1992,
Imdorf
et
al.
1994, 1996;
Kraus
et
al.
1994;
Eischen
1996;
Sammataro
et
al.
1998;
Lindberg
et
al.
2000).
Thymol,
in
particular,
has
been
tested
previously
with
promising
although
variable
results
(Calderone
et
al.
1991, 1997;
Imdorf
et
al.
1994,
0022-0493/02/0221-0226$02.00/0
0
2002
Entomological
Society
of
America
222
JOURNAL
OF
ECONOMIC
ENTOMOLOGY
Vol.
95,
no.
2
1996)
and
is
used
as
a
registered
control
in
Europe
as
ApiLife
VAR
(Imdorf
et
al.
1994,
Eischen
1996).
We
previously
determined
that
thymol
and
Magic
3,
a
proprietary
blend
of
five
botanical
oil
constituents
developed
by
Eco-Smart
Technologies
(Nashville,
TN)
and
Murray
Isman
(University
of
British
Colum-
bia,
Vancouver,
BC,
Canada),
exhibited
high
varroa
mortality
levels
coupled
with
low
bee
mortality
in
laboratory
tests
(Lindberg
et
al.
2000).
Therefore,
these
two
substances
were
selected
for
further
exper-
iments
to
determine
their
effect
on
both
parasitic
mites.
Canola
oil
also
provides
tracheal
mite
control
and
is
a
good
carrier
for
thymol.
We
compared
the
efficacy
of
these
three
oil
blends
in
various
delivery
devices
(experiment
1)
and
locations
(experiment
2)
in
the
colony.
Materials
and
Methods
Experiments
1
and
2:
Assessments.
For
both
exper-
iments,
the
adult
and
sealed
brood
populations
of
all
test
colonies
were
assessed
the
week
before
the
treat-
ments
commenced,
following
the
protocol
outlined
in
Burgett
and
Burikam
(1985).
The
assessment
proce-
dure
was
identical
for
both
experiments.
After
a
24-d
treatment
period,
the
colonies
were
again
assessed
for
adult
and
brood
populations,
and
all
colonies
were
treated
with
Apistan
(Wellmark
International,
Dallas,
TX)
strips
for
6
wk,
following
label
directions.
Sticky
boards
were
placed
on
the
bottom
boards
of
all
col-
onies
to
capture
mites,
and
these
boards
were
re-
moved
and
replaced
once
a
week
for
4
wk
or
until
mite
counts
on
a
board
were
consistently
less
then
10,
at
which
point
mite
measurements
were
stopped.
At
the
end
of
the
6-wk
Apistan
application,
strips
were
re-
moved
from
the
colonies.
Experiment
1.
Comparison
of
Different
Release
Devices.
Colonies.
Forty-eight
colonies
were
placed
in
an
apiary
located
in
Abbotsford,
BC.
The
experimental
treatment
period
began
on
22
Apri11999
and
ended
16
May
1999.
Colonies
of
A.
mellifera
were
created
by
dividing
existing
colonies,
and
each
was
founded
with
a
newly
mated
queen
imported
from
Australia.
Each
experimental
colony
was
housed
in
one
standard
Langstroth
deep
hive
body,
with
three
frames
of
brood
and
bees,
two
frames
of
honey,
one
feeder
frame,
and
three
empty
frames.
Varroa
levels
in
these
48
colonies
were
determined
by
counting
the
number
of
mites
that
dropped
naturally
onto
a
sticky
board
after
24
h.
A
sticky
board
consisted
of
a
thin
cardboard
sheet
coated
with
a
nondrying
latex
glue,
which
was
inserted
onto
the
bottom
board
of
the
colony.
These
sticky
boards
were
overlaid
with
a
wooden
frame
covered
with
wire
mesh
that
allowed
mites
to
fall
through
but
prevented
bees
from
contacting
the
sticky-board.
To
equalize
varroa
mite
levels,
the
18
colonies
with
the
highest
mite
levels
and
the
18
colonies
with
the
lowest
mite
levels
had
200
g
of
bees
taken
from
the
most
highly
infested
colony
and
placed
in
the
least
infested
colony,
and
vice
versa
continuing
up
the
rank:
second
highest
with
second
lowest,
third
highest
with
third
lowest,
and
so
on.
Colonies
with
middle
levels
of
var-
roa
were
not
manipulated.
Mite
Sampling.
Varroa
mites
were
collected
using
sticky
boards,
which
were
removed
and
replaced
once
every
4
d,
and
the
mite
numbers
counted.
Tracheal
mites
were
sampled
by
collecting
approximately
30
bees
in
ethanol
from
the
inner
cover
of
each
colony,
immediately
before
the
treatments
were
applied
and
shortly
after
they
were
removed
for
the
final
time.
Bee
dissections
were
carried
out
as
outlined
by
Shimanuki
and
Knox
(1991),
and
mites
were
scored
as
either
present
or
absent
in
the
tracheae.
Treatments.
The
six
treatments
examined
were
as
follows:
(1)
low
density
polyethylene
(LDPE)
sleeves
(Pherotech,
Delta,
BC,
Canada)
filled
with
Magic
3
prepared
by
Rod
Bradbury
(Eco-Safe
Technologies,
Saanichton,
BC,
Canada);
(2)
plastic
strip
coated
with
calcium
carbonate
and
Magic
3;
(3)
Magic
3-infused
florist
blocks;
(4)
thymol
(Sigma-Aldrich
Canada,
Oakville,
ON)
infused
florist
blocks
(considered
a
positive
control);
(5)
canola
oil
and
thymol
mixture
wick
release;
and
(6)
untreated
control.
Eight
colonies
per
treatment
were
used,
and
treatments
were
as-
signed
randomly.
The
release
devices
were
replaced
at
4-d
intervals
for
24
d.
Low
Density
Polyethylene
(LDPE)
Sleeve.
The
LDPE
sleeve
measured
18.5
by
6.5
cm
and
the
cotton
batting
placed
inside
the
strip
measure
16
by
5
cm.
Doses
of
Magic
3
(16.4
ml)
were
poured
onto
strips
of
cotton
batting
placed
in
LDPE
sleeves
to
prevent
pooling
in
the
bottom.
These
sleeves
had
a
release
rate
of
0.2
ml
of
Magic
3
per
day
(unpublished
data).
Therefore,
three
sleeves
were
used
to
obtain
the
desired
release
rate
of
0.6
ml
per
day.
These
sleeves
were
hung
be-
tween
the
brood
frames
with
a
piece
of
wire
looped
through
a
hole
in
the
sleeve.
The
hole
was
not
in
a
section
of
the
sleeve
containing
the
treatment.
The
sleeves
were
placed
in
the
center
of
the
brood
rearing
frames
so
that
they
contacted
as
much
of
the
brood
as
possible.
Calcium
Carbonate
Paste.
Eco-Safe
Technologies
(Saanichton,
BC,
Canada)
provided
a
mixture
of
10%
Magic
3,
10%
Linseed
oil,
and
80%
calcium
carbonate
coated
on
a
15
cm
by
2
cm
by
2-mm
thick
strip.
Each
strip
had
approximately
2
mm
of
paste
on
each
side,
which
gave
an
approximate
dosage
of
1.3
ml
of
Magic
3
per
strip.
These
strips
had
a
release
rate
of
0.05
ml
per
day.
Three
strips
per
colony
were
hung
between
the
brood
frames
so
that
the
strips
contacted
as
much
of
the
brood
as
possible.
The
strips
had
an
uncoated
section
at
the
top,
which
was
bent
at
a
90°
angle
to
the
rest
of
the
strip
and
allowed
the
strips
to
hang
without
falling
through
the
frames.
Thymol
Florist
Blocks.
Whittington
et
al.
(2000)
showed
efficacy
against
varroa
using
vermiculite
blocks
dosed
with
3.8
g
of
thymol,
for
a
two-hive
body
colony.
Because
our
experiment
used
colonies
one
hive
body
in
size,
the
dose
was
halved.
To
achieve
a
release
rate
of
0.15
g
of
thymol
per
hive
body
per
day
from
the
florist
blocks
at
33°C,
1.9
g
of
thymol
was
applied
to
the
blocks.
Pieces
of
vermiculite
block
40
by
55
by
7
mm
were
treated
with
19
ml
of
a
10%
thymol
April
2002
RICE
ET
AL.:
BOTANICAL
OILS
FOR
HONEY
BEE
PARASITE
CONTROL
223
in
hexane
solution.
The
thymol-infused
hexane
was
poured
over
the
florist
block
and
the
hexane
allowed
to
volatilize,
leaving
the
thymol
behind.
Florist
blocks
were
weighed
before
the
application
of
the
thymol/
hexane
and
after
the
hexane
had
evaporated.
Loss
of
thymol
during
the
transfer
from
hexane
to
florist
block,
and
the
subsequent
volatilization
of
the
hexane,
did
not
significantly
reduce
the
amount
of
thymol
being
applied
to
the
blocks.
One
thymol
block
was
placed
in
each
colony,
centered
on
top
of
the
frames
under
the
inner
cover.
Magic
3
Florist
Blocks.
Examining
the
Lindberg
et
al.
(2000)
LD
5
,
values
for
thymol
and
Magic
3,
16.4
ml
of
Magic
3
is
proportional
to
3.8
g
of
thymol
in
efficacy,
and
a
release
rate
of
1.2
ml
of
Magic
3
is
proportional
to
0.31
g
of
thymol
per
day.
As
used
in
this
experiment,
Magic
3
had
a
release
rate
from
vermiculite
blocks
of
0.6
ml
per
day
for
four
d
using
an
initial
dose
of
16.4
ml
of
Magic
3
on
the
block;
after
this
time
the
release
rate
was
variable
(unpublished
data).
From
these
data
we
determined
that
a
release
rate
of
0.6
ml
per
day
was
optimal
for
a
one-hive
body-sized
colony.
This
rate
was
achieved
by
pouring
16.4
ml
of
Magic
3
onto
the
florist
blocks
and
allowing
it
to
absorb.
All
florist
blocks
were
enclosed
in
wire
mesh
in
the
colonies
to
prevent
the
bees
from
chewing
them,
and
were
cen-
tered
on
top
of
the
frames
under
the
inner
cover.
Wick
System.
Fifty
milliliters
of
a
1.2%
thymol
in
canola
mixture
was
poured
into
an
LDPE
sleeve
and
absorbed
by
a
1.27-cm
cotton
wick
sewn
into
one
end
of
the
sleeve.
After
the
wick
had
absorbed
the
50
ml,
an
additional
350
ml
of
the
1.2%
thymol
in
canola
mixture
was
poured
into
the
LDPE
sleeve,
which
was
used
as
a
reservoir
for
the
fluid.
The
opposite
end
of
the
sleeve
was
closed
off
with
a
paper
clip
and
the
reservoir
was
placed
horizontally
centered
over
the
tops
of
the
frames
with
the
wick
dropping
down
be-
tween
the
frames
in
the
brood
rearing
area.
Experiment
2.
Comparison
of
Release
Device
Placements.
Colonies.
To
select
colonies
for
the
ex-
periment,
45
two
hive
body-sized
colonies
from
ex-
isting
Simon
Fraser
University
apiaries
were
sampled
for
varroa
using
a
24-h
natural
mite
drop
test;
that
is
a
test
in
which
no
acaracides
are
used,
and
mites
that
fall
naturally
off
the
bees
are
captured
and
recorded.
The
30
colonies
with
the
highest
mite
infestations
were
used
for
the
experiment
and
were
randomly
assigned
to
treatments.
Colonies
were
not
equalized
for
brood
or
adult
bees.
These
colonies
also
were
seeded
with
200
g
of
bees
from
colonies
that
were
known
to
be
highly
infested
from
a
separate
apiary
to
ensure
that
mites
were
present.
All
experimental
colonies
were
then
moved
to
an
apiary
at
the
British
Columbia
Min-
istry
of
Agriculture
in
Abbotsford
BC.
The
experimen-
tal
treatment
period
began
on
2
September
1999
and
ended
25
September
1999.
Colonies
were
treated
with
the
antibiotic
oxytetracycline
hydrochloride
(Oxytet-
25,
Medivet
Pharmaceuticals,
High
River,
Alberta,
Canada)
during
the
course
of
the
experiment
to
pre-
vent
American
foulbrood
disease,
and
were
fed
,
=6.7
liters
of
sugar
syrup
in
a
two
parts
sugar
to
one
part
water
concentration.
The
sugar
syrup
contained
the
antibiotics
oxytetracycline
hydrochloride
(Oxytet-
25)
and
fumagilin
(Fumagilin
B,
Medivet
Pharmaceu-
ticals).
Mites
were
sampled
from
the
colonies
in
the
manner
described
previously,
except
sheets
were
pulled
and
replaced
every
8
d
for
24
d.
Treatments
also
were
replaced
every
8
d.
Treatments.
Ten
colonies
were
used
in
each
of
two
treatments
and
a
control.
Three
LPDE
sleeves
per
hive
body
were
placed
in
colonies
with
16.4
ml
of
Magic
3
per
sleeve.
For
the
first
treatment,
10
colonies
had
the
sleeves
hung
with
loops
of
wire
between
the
outer
frames.
Sleeves
were
alternated
so
that
in
the
top
hive
body
two
sleeves
were
located
between
frames
1
and
2
and
one
sleeve
was
between
frames
9
and
10,
and
in
the
bottom
hive
body
the
sleeve
arrangement
was
reversed.
In
the
second
treatment,
10
colonies
had
three
sleeves
per
hive
body
with
the
same
dose
of
Magic
3,
16.4
ml.
These
sleeves
were
placed
horizon-
tally
across
the
top
bars
of
the
frames
directly
over
the
brood
rearing
area
in
both
the
top
and
bottom
hive
body.
The
remaining
10
colonies
were
used
as
an
untreated
control
and
were
left
undisturbed.
Experiments
1
and
2:
Data
Analysis.
All
treatments
were
assigned
randomly
to
colonies.
Mite
drop
due
to
treatments
was
calculated
as
the
number
of
mites
collected
during
the
time
when
treatments
were
in
place
divided
by
the
total
number
of
mites
collected
during
the
entire
experiment
(mites
killed
during
treatment
plus
mites
killed
during
Apistan
applica-
tion).
Brood
level
change
was
calculated
as
the
in-
crease
or
decrease
of
posttreatment
brood
levels
as
compared
with
pretreatment
brood
levels.
Similarly,
adult
bee
population
change
was
calculated
as
the
increase
or
decrease
of
posttreatment
adult
bee
pop-
ulation
as
compared
with
pretreatment
adult
bee
lev-
els.
Unless
otherwise
stated,
means
of
all
data
were
compared
using
an
analysis
of
variance
followed
by
the
Tukey-Kramer
honestly
significant
difference
(HSD)
multiple
comparison
test
(ac
=
0.05)
(JMP
IN,
SAS
Institute
1997).
Proportion
data
were
log
trans-
formed.
Results
Experiment
1.
Varroa
Mite
Efficacy.
There
were
significant
differences
in
efficacy
of
the
treatments
in
killing
varroa
(F
=
11.9871;
df
=
5,
42;
P
<
0.001).
The
Magic
3
LDPE
sleeve
was
significantly
more
effective
in
the
control
of
varroa
mites
than
any
of
the
other
treatments
(Fig.
1).
None
of
the
treatments
except
for
the
Magic
3
LDPE
sleeve
were
more
effective
than
the
untreated
control
(P
<
0.05).
Tracheal
Mite
Efficacy.
Tracheal
mite
levels
were
similar
among
colonies
at
the
beginning
of
the
exper-
iment
(F
=
0.3003;
df
=
5,
42;
P
=
not
significant).
At
the
end
of
the
experiment,
significant
differences
were
found
for
the
Magic
3
sleeve
(92%
decrease
from
pretreatment
to
posttreatment
infestation
levels),
the
Magic
3
florist
block
(80%
decrease),
and
the
thymol-
canola
wick
(75%
decrease)
all
had
significantly
lower
tracheal
mite
levels
than
the
untreated
control
(43%
increase)
(F
=
3.4525;
df
=
5,
39;
P
=
0.0112).
Neither
a
b
j_
b
b
b
T
b
T
T
Varroa
des
truc
to
r
Mo
r
ta
lity
(
%
3
S.
E.
)
20
60
40
0
16
12
Mean
Pre-Treatment
Tracheal
Mite
Infestation
S.E.)
8
1
ab
ab
b
b b
6"
5,
6
0
4.
°
*0
,
se
„,
o
e
J
c
,
o
0
C
JF
Fig.
2.
Efficacy
of
each
treatment
against
A.
woodi
in-
festation.
Means
with
the
same
letters
are
not
significantly
different
(Tukey's
HSD,
cc
=
0.05).
Trac
hea
l
Mite
In
fes
ta
tion
(
%
t
S.
E.
)
a
a
b
ent
ood
Ares
S.
Pre
Trea
I
Mean
a
ire-Treatment
)Thrk
Popu
atioi
S.
E.)
ee'
•,.b
G
a
"
Fig.
3.
Sealed
brood
area
(A)
and
adult
population
(B)
of
treatment
groups
at
the
end
of
the
treatment
period,
compared
with
pretreatment
brood
area
and
adult
popula-
tion.
Means
with
the
same
letters
are
not
significantly
dif-
ferent
(Tukey's
HSD,
cc
=
0.05).
3000
Broo
d
Area
(c
m
2
±
S.
E.
)
2000
1000
Num
be
r
o
f
Bees
x
1000
(
±
S.
E.
)
10
224
JOURNAL
OF
ECONOMIC
ENTOMOLOGY
Vol.
95,
no.
2
®
e
\
<<
4
c-P
('
4"
b.
6s
&
Ga
r
Fig.
1.
Relative
efficacy
of
each
treatment
against
V.
destructor.
Means
with
the
same
letters
are
not
significantly
different
(Tukey's
HSD,
cc
=
0.05).
the
calcium
carbonate/Magic
3
paste
(43%
decrease),
nor
the
thymol
florist
blocks
(58%
decrease)
were
significantly
different
from
the
other
treatments
(Fig.
2).
Posttreatment
Colony
Size.
At
the
beginning
of
the
experiment
there
were
no
differences
in
colony
pop-
ulation
with
respect
to
brood
area
(F
=
0.1654;
df
=
5,
42;
P
=
ns)
or
adult
bee
population
(F
=
0.5165;
df
=
5,
42;
P
=
not
significant).
For
all
treatments
the
adult
population
increased
over
the
course
of
the
experi-
ment,
but
the
Magic
3
LDPE
sleeve
had
fewer
adults
(although
not
significantly
less)
than
the
other
treat-
ments
(Fig.
3B).
All
brood
areas
increased except
for
the
Magic
3
LDPE
sleeve,
which
had
little
change
from
the
beginning
to
the
end
of
the
experiment,
and
sig-
nificantly
less
brood
than
the
Magic
3
florist
block
and
the
untreated
control
at
the
end
of
the
experiment
(F
=
2.7792;
df
=
5,
42;
P
=
0.0295)
(Fig.
3A).
Experiment
2. Varroa
Efficacy.
There
were
no
sig-
nificant
differences
in
the
efficacy
of
the
treatments
in
killing
varroa
(F
=
0.7446;
df
=
2,
25;
P
=
not
signif-
icant).
None
of
the
treatments
were
significantly
dif-
ferent
from
the
untreated
control
(Fig.
4A).
Posttreatment
Colony
Size.
There
was
no
difference
between
the
colonies
in
colony
size
at
the
beginning
of
the
experiment
with
respect
to
brood
area
(F
=
1.2459;
df
=
2,
27;
P
=
not
significant).
The
brood
area
decreased
in
all
treatments,
but
no
treatment
was
significantly
different
than
the
others
(F
=
1.2964;
df
=
2,
27;
P
=
not
significant)
(Fig.
4B).
Discussion
Magic
3
showed
the
highest
varroa
control
when
administered
as
a
sleeve
treatment
during
the
first
experiment,
but
these
higher
levels
of
control
were
offset
by
decreased
adult
populations
and
significantly
decreased
brood
production.
Tests
with
sleeve
place-
ment
in
experiment
2
showed
that
altering
the
location
Varr
oa
des
truc
tor
Mor
ta
lity
(
%
±
S.
E.
)
Broo
d
Area
(cm
z
±
S.
E.
)
a
A.
a
Mean
Pre-Treatment
Brood
Area (±S.E.)
I
B.
a
T
a
T
a
Horizontal
Outside
Frames
Untreated
April
2002
RICE
ET
AL.:
BOTANICAL
OILS
FOR
HONEY
BEE
PARASITE
CONTROL
225
12
8
4
0
3000
2000
1000
0
Fig.
4.
Relative
efficacy
of
each
treatment
against
V.
destructor
(A)
and
the
adult
population
(B)
at
the
end
of
the
treatment
period
in
comparison
to
pretreatment
adult
pop-
ulation.
Means
with
the
same
letters
are
not
significantly
different
(Tukey's
HSD,
cc
=
0.05).
of
the
treatment
produced
no
effect
on
brood
pro-
duction
relative
to
the
control,
but
lowered
the
ca-
pacity
of
Magic
3
to
kill
mites
so
severely
as
to
render
the
product
ineffective
in
those
locations.
Although
we
did
see
less
brood
death
in
experiment
2
than
in
experiment
1,
Magic
3
placement
locations
in
these
experiments
cannot
justifiably
be
compared
because
of
the
constraints
of
outside
variables
including
time
of
application
and
the
lack
of
a
comparable
center
frame
application.
In
experiment
1,
it
is
not
clear
whether
Magic
3
itself
or
simply
physical
interference
from
the
sleeve
placed
next
to
the
brood
caused
the
drop
in
brood
production.
The
variability
in
treatment
efficacy
found
in
pre-
vious
studies
with
botanical
oil
applications
was
also
evident
in
our
results.
The
standard
(positive
control)
of
the
thymol
block
(Calderone
et
al.
1997,
Whitting-
ton
et
al.
2000)
showed
approximately
35%
control
of
varroa
mites,
which
is
at
the
lower
end
of
the
30
-70%
range
found
in
previous
studies
(Calderone
et
al.
1997,
Whittington
et
al.
2000).
Tracheal
mite
control
with
Magic
3
and
with
thymol
infused
canola
oil
was
dramatic,
and
either
of
these
applications
could
be
used
given
the
proper
dosage
and
placement.
Future
experiments
should
examine
a
range
of
Magic
3
doses
either
in
the
field
or
in
a
bioassay
system,
similar
to
that
of
Lindberg
et
al.
(2000),
to
determine
the
dosage
required
to
provide
tracheal
mite
control.
A
thymol/canola
oil
mixture
administered
as
a
spray
can
control
tracheal
mites
but
is
hard
on
bees
(Whittington
et
al.
2000).
The
same
thymol/
canola
oil
mixture
was
put
into
the
wicks
in
our
experiments
and
also
showed
good
efficacy
against
tracheal
mites
without
the
detrimental
effects
to
bees
and
brood
of
a
spray.
In
both
experiments,
bees
were
able
to
come
into
direct
contact
with
the
mixture;
future
studies
of
bee
interactions
with
a
thymol/
canola
oil
release
device
would
be
useful.
Magic
3
has
better
potential
to
control
both
tracheal
and
varroa
mites
than
thymol
alone,
but
only
if
place-
ment
and
release
technology
can
be
developed
to
reduce
negative
effects
on
brood.
The
optimal
release
device
should
administer
a
constant
dose
over
a
set
period
and
be
easy
to
apply.
The
LDPE
sleeves
tested
here
were
a
first
step
toward
refining
this
release
technology.
These
can
be
applied
similarly
to
Apistan
strips
and
provide
a
specified
dose
for
up
to
8
d.
A
single
reservoir
releasing
a
higher
dose
per
day
might
be
more
useful
if
the
placement
can
be
optimized.
There
are
several
potential
applications
for
botan-
ical
oils
in
controlling
diseases
and
parasites
in
honey
bee
colonies,
but
the
use
of
botanical
oil
formulations
in
an
integrated
pest
management
(IPM)
approach
to
mite
control
needs
improvement
to
have
commercial
success.
It
is
still
not
clear
whether the
95-99%
levels
of
control
found
with
Apistan
and
coumaphos
need
to
be
maintained,
and
if
so
whether
such
levels
could
be
reached
using
combinations
of
less
conventional
methods
such
as
queen
selection,
drone
comb
traps,
physical
barriers
that
prevent
the
reentry
of
mites
into
the
colonies,
or
a
plethora
of
other
control
techniques
that
have
not
been
fully
explored.
We
encourage
the
research
community
to
shift
its
focus
from
the
more
singular
approach
of
pesticide
use
to
the
more
broad-
spectrum
approach
of
an
IPM
system.
Acknowledgments
We
thank
the
members
of
the
Simon
Fraser
University
"Swarm
Team"
for
their
valuable
assistance,
especially
Shawn
Devlin,
Kerri
Johannson,
Adony
Melathopolous,
Hien
Ngo,
Stephen
Pemal,
and
Holly
Sabara.
We
also
thank
Murray
Isman,
Rod
Bradbury,
and
Eco-Smart
and
Eco-Safe
Tech-
nologies
for
advice,
preparation
of
release
devices,
and
do-
nation
of
Magic
3,
and
the
British
Columbia
Ministry
of
Agriculture
and
Food
for
the
use
of
an
apiary
site
in
Abbots-
ford,
BC.
Funding
was
provided
by
Eco-Smart
Technologies,
The
Agri-Food
Innovation
Fund
of
Saskatchewan
Agricul-
ture,
the
Natural
Sciences
and
Engineering
Research
Coun-
cil,
and
the
Canadian
Bee
Research
Fund.
References
Cited
Anderson,
D.
L.,
and
J.W.H.
Trueman.
2000.
Varroa
jazob-
soni
(Acari:
Varroidae)
is
more
than
one
species.
Exp.
Appl.
Acarol.
24:
165-189.
Balayannis,
P.
G.,
and
L.
A.
Santas.
1992.
Dissipation
of
mal-
athion
and
fluvalinate
residues
from
honey.
J.
of
Apic.
Res.
31(2):
70
-76.
Bogdanov,
S.,
A.
Imdorf,
and
V.
Kilchenmann.
1998.
Resi-
dues
in
wax
and
honey
after
Apilife
VAR
treatment.
Apidologie
29:
513-524.
Burgett,
M.,
and
I.
Burikam.
1985.
Number
of
adult
honey
bees
(Hymenoptera
Apidae)
occupying
a
comb:
a
stan-
226
JOURNAL
OF
ECONOMIC
ENTOMOLOGY
Vol.
95,
no.
2
dard
for
estimating
colony
populations.
J.
Econ.
Entomol.
78:
1154
-1156.
Calderone,
N.
W.,
W.
A.
Bruce,
G.
Allen-Wardell,
and
H.
Shimanuki.
1991.
Evaluation
of
botanical
compounds
for
control
of
the
honey
bee
tracheal
mite,
Acarapis
woodi.
Am.
Bee
J.
131:
589-591.
Calderone,
N.
W.,
W.
Wilson,
and
M.
Spivak.
1997.
Plant
extracts
used
for
control
of
the
parasitic
mites
Varroa
jacobsoni
(Acari:
Varroidae)
and
Acarapis
woodi
(Acari:
Tarsonemidae)
in
colonies
of
Apis
mellifera
(Hymenop-
tera:
Apidae).
J.
Econ.
Entomol.
90:
1080-1086.
Colin,
M.
E.
1990.
Essential
oils
of
Labiatae
for
controlling
honey
bee
varroosis.
J.
Appl.
Entomol.
110:
19-25.
De
Guzman,
L.
I.,
T.
E.
Rinderer,
and
G.
T.
Delatte.
1998.
Comparative
resistance
of
four
honey
bee
(Hymenop-
tera:
Apidae)
stocks
to
infestation
by
Acarapis
woodi
(Acari:
Tarsonemidae).
J.
Econ.
Entomol.
91:
1078
-1083.
De
Jong,
D.
1997.
Mites:
Varroa
and
other
parasites
of
brood,
pp.
279-327.
In
R.
A.
Morse
and
K.
Flottum
[eds.],
Honey
bee
pests,
predators,
and
diseases,
3rd
ed.
A.I.
Root,
Medina,
OH.
Delaplane,
K.
S.
1992.
Controlling
tracheal
mites
(Acari:
Tarsonemidae)
in
colonies
of
honey
bees
(Hymenoptera:
Apidae)
with
vegetable
oil
and
menthol.
J.
Econ.
Ento-
mol.
85:
2118-2124.
Delaplane,
K.
S.,
and
D.
F.
Mayer.
2000.
Benefits
of
bee
pollination,
pp.
1-3
In
Crop
pollination
by
bees.
CAB,
Wallingford,
Oxon,
UK.
Delfinado-Baker,
M.
1984.
Acarapis
woodi
in
the
U.S.
Am.
Bee
J.
124:
805-806.
Downey,
D.
L.,
T.
Higo,
and
M.
L.
Winston.
2000.
Honey
bee
colony
mortality
and
productivity
with
single
and
dual
infestations
of
parasitic
mite
species.
Insectes
Soc.
47:
171-176.
Eischen,
F.
A.
1996.
Botanical
acaricides
and
Varroa
con-
trol.
Am.
Bee
J.
136:
277-278.
Elzen,
P.
J.,
F.
A.
Eischen,
J.
B.
Baxter,
J.
Pettis,
G.
W.
Elzen,
and
W.
T.
Wilson.
1998.
Fluvalinate
resistance
in
Varroa
jacobsoni
from
several
geographic
locations.
Am.
Bee
J.
138:
674-676.
Feldlaufer,
M.
F.,
J.
S.
Pettis,
J.
P.
Kochansky,
and
H.
Shi-
manuki.
1997.
A
gel
formulation
of
formic
acid
for
the
control
of
parasitic
mites
of
honey
bees.
Am.
Bee
J.
137:
661-663.
Herbert,
E.
W.,
Jr.,
H.
Shimanuki,
and
J.
C.
Mathenius
Jr.
1987.
The
effects
of
two
candidate
compounds
on
Acara-
pis
woodi
in
New
Jersey.
Am.
Bee
J.
127:
776-778.
Imdorf,
A.,
V.
Kilchenmann,
C.
Maquelin,
and
S.
Bogdanov.
1994.
Optimierung
der
Anwendung
von
`Apilife
VAR'
zur
Bekampfung
von
Varroa
jacobsoni
Oud
in
Bienen-
voikern.
Apidologie
25:
49
-60.
Imdorf,
A.,
J.
D.
Charriere,
C.
Maquelin,
V.
Kilchenmann,
and
B.
Bachofen.
1996.
Alternative
Varroa
control.
Am.
Bee
J.
136:
189-193.
Kjer,
K.
M.,
D.
W.
Ragsdale,
and
B.
Furgala.
1989.
A
retro-
spective
and
prospective
overview
of
the
honey
bee
tra-
cheal
mite,
Acarapis
woodi
R.
Am.
Bee
J.
129:
25-28.
Kochansky,
J.,
and
H.
Shimanuki.
1999.
Development
of
a
gel
formulation
of
formic
acid for
control
of
parasitic
mites
and
honey
bees.
J.
Agric.
Food
Chem.
47:
3850
-
3853.
Kraus,
B.,
N.
Koeniger,
and
S.
Fuchs.
1994.
Screening
of
substances
for
their
effect
on
Varroa
jacobsoni:
attrac-
tiveness,
repellency,
toxicity
and
masking
effects
of
ethe-
real
oils.
J.
Apic.
Res.
33:
34-43.
Lindberg,
C.
M.,
A.
P.
Melathopoulos,
and
M.
L.
Winston.
2000.
Laboratory
evaluation
of
miticides
to
control
Var-
roa
jacobsoni
Oud.
(Acari:
Varroidae),
a
honey
bee
(Apis
mellifera
L.,
Hymenoptera
Apidae)
parasite.
J.
Econ.
Entomol.
93:
189
-198.
Lodesani,
M.,
A.
Pellacani,
S.
Bergomi,
E.
Carpana,
T.
Rabitti,
and
P.
Lasagni.
1992.
Residue
determination
for
some
products
used
against
Varroa
infestation
in
bees.
Apid-
ologie
23:
257-272.
Marchetti,
S.,
R
Barbattini,
and
M.
D'Agaro.
1984.
Compar-
ative
effectiveness
of
treatments
used
to
control
Varroa
jacobsoni
Oud.
Apidologie
15:
363-378.
Milani,
N.
1999.
The
resistance
of
Varroa
jacobsoni
Oud.
to
acaracides.
Apidologie
30:
229
-234.
Sammataro,
D.,
S.
Cobey,
B.
H.
Smith,
and
G.
R.
Needham.
1994.
Controlling
tracheal
mites
(Acari:
Tarsonemidae)
in
honey
bees
(Hymenoptera:
Apidae)
with
vegetable
oil.
J.
Econ.
Entomol.
87:
910
-916.
Sammataro,
D.,
G.
Degrandi-Hoffman,
G.
Needham,
and
G.
Wardell.
1998.
Some
volatile
plant
oils
as
potential
con-
trol
agents
for
Varroa
mites
(Acari:
Varroidae)
in
honey
bee
colonies
(Hymenoptera:
Apidae).
Am.
Bee
J.
138:
681-685.
Sanford,
T.,
K.
Flottum,
and
B.
Arthur.
1999.
Focus
on
Bayer
strips.
Bee
Culture
127:
32-35.
SAS
Institute.
1997.
User's
manual,
JMP
in
release
3.1.5.
SAS
Institute,
Cary,
NC.
Shimanuki,
H.,
and
D.
A.
Knox.
1991.
Diagnosis
of
honey
bee
diseases.
USDA
Agric.
Handb.
AH-690,
p.
53.
Sherman,
P.
W.,
T.
D.
Seeley,
and
H.
K.
Reeve.
1998.
Par-
asites,
pathogens,
and
polyandry
in
honey
bees.
Am.
Nat.
151:
392-396.
Smith,
A.
W.,
R.
E.
Page,
Jr.,
and
G.
R.
Needham.
1991.
Vegetable
oil
disrupts
the
dispersal
of
tracheal
mites
Acarapis
woodi
(Rennie),
to
young
bee
hosts.
Am.
Bee
J.
131:
44
-46.
Thrasyvoulou,
A.
T.,
and
N.
Pappas.
1988.
Contamination
of
honey
and
wax
with
malathion
and
coumaphos
used
against
the
Varroa
mite.
J.
Apic.
Res.
27(1):
55-61.
Wallner,
K.
1995.
The
use
of
Varroacides
and
their
influ-
ence
on
the
quality
of
bee
products.
Am.
Bee
J.
135:
817-821.
Whittington,
R.,
A.
P.
Melathopoulos,
M.
L.
Winston,
and
H.
Higo.
2000.
Evaluation
of
the
botanical
oils
neem,
thy-
mol,
and
canola
sprayed
to
control
Varroa
jacobsoni
Oud.
(Acari:
Varroidae)
and
Acarapis
woodi
(Acari:
Tarson-
emidae)
in
colonies
of
honey
bees
(Apis
mellifera
L.,
Hymenoptera
Apidae).
Am.
Bee
J.
140:
567-572.
Wilson,
W.
T,
J.
S.
Pettis,
C.
E.
Henderson,
and
R.
A.
Morse.
1997.
Tracheal
mites,
pp.
253-278.
In
R.
A.
Morse,
and
K.
Flottum
[eds.],
Honey
bee
pests,
predators,
and
diseases,
3rd
ed.
A.I.
Root,
Medina,
OH.
Received
for
publication
11
September
2000;
accepted
5
March
2001.