The influence of glucosinolates and sugars on feeding by the cabbage stem flea beetle, Psylliodes chrysocephala


Bartlet, E.; Parson, D.; Williams, I.H.; Clark, S.J.

Entomologia Experimentalis et Applicata 73(1): 77-83

1994


Agar was used as an artificial substrate to investigate the feeding behaviour of the cabbage stem flea beetle, Psylliodes chrysocephala L. (Coleoptera: Chrysomelidae), an important pest of winter oilseed rape (Brassica napus) in Europe. Both glucosinolates and sugars stimulated feeding when added to agar. The amount of feeding that occurred was affected by the type and concentration of glucosinolate and sugar and also by combinations of components. Although glucosinolates were potent feeding stimulants for P. chrysocephala, they were not a prerequisite for feeding, nor does it seem likely that glucosinolate profiles are used by this species to discriminate amongst cruciferous plants at the gustatory level.

Entomologia
Experimentalis
et
Applicata
73:
77-83,
1994.
77
©
1994
Kluwer
Academic
Publishers.
Printed
in
Belgium.
The
influence
of
glucosinolates
and
sugars
on
feeding
by
the
cabbage
stem
flea
beetle,
Psylliodes
chrysocephala
E.
Bartlet,
D.
Parsons,
I.
H.
Williams
&
S.
J.
Clark
AFRC
Institute
of
Arable
Crops
Research,
Rothamsted
Experimental
Station,
Harpenden,
Herts.
AL5
2JQ,
UK
Accepted:
March
1,
1994
Key
words:
Psylliodes
chrysocephala,
cabbage
stem
flea
beetle,
oilseed
rape,
host
plant
selection,
glucosinolates,
sugars,
feeding
stimulant
Abstract
Agar
was
used
as
an
artificial
substrate
to
investigate
the
feeding
behaviour
of
the
cabbage
stem
flea
beetle,
Psylliodes
chrysocephala
L.
(Coleoptera:
Chrysomelidae),
an
important
pest
of
winter
oilseed
rape
(Brassica
nap
us)
in
Europe.
Both
glucosinolates
and
sugars
stimulated
feeding
when
added
to
agar.
The
amount
of
feeding
that
occurred
was
affected
by
the
type
and
concentration
of
glucosinolate
and
sugar
and
also
by
combinations
of
components.
Although
glucosinolates
were
potent
feeding
stimulants
for
P.
chrysocephala,
they
were
not
a
prerequisite
for
feeding,
nor
does
it
seem
likely
that
glucosinolate
profiles
are
used
by
this
species
to
discriminate
amongst
cruciferous
plants
at
the
gustatory
level.
Introduction
The
feeding
of
phytophagous
insects
is
influenced
by
the
presence
in
plant tissue
of
secondary
plant
sub-
stances,
chemical
components
which
are
only
found
in
particular
plant
taxa
(Fraenkel,
1969).
The
best
known
examples
are
the
glucosinolates,
secondary
plant
sub-
stances
characteristic
of
the
Brassicacae
(formerly
Cru-
ciferae).
Glucosinolates
initiate
or
prolong
feeding
in
many
insect
species
which
feed
specifically
on
cru-
cifers
(Mitchell,
1988).
Crucifer
feeding
insects
may
discriminate
amongst
cruciferous
plants
by
responding
to
species-specific
glucosinolate
profiles,
i.e.
the
types
and
proportions
of
constituent
glucosinolates
(Rodman
&
Chew,
1980;
Louda
&
Mole,
1991).
Host
plant
acceptability
is,
however,
not
solely
mediated
by
sec-
ondary
plant
substances
and
plant
damage
by
crucifer
feeding
insects
in
the
field
sometimes
correlates
bet-
ter
with
nutrient
levels
than
with
glucosinolate
levels
(Louda
&
Mole,
1991).
Sugars,
amino
acids,
pro-
teins,
vitamins,
sterols
and
phospholipids
have
all
been
reported
to
be
insect
phagostimulants
(Hsaio,
1985).
The
cabbage
stem
flea
beetle,
Psylliodes
chrysocephala,
is
an
important
pest
of
winter
oilseed
rape
in
Europe
(Bromand,
1990;
Winfield
1992).
Adults
move
into
the
newly
emerged
crops
in
August
to
September.
They
remain
there
throughout
the
winter,
where
they
feed
on
the
leaves
and
lay
eggs
in
the
soil.
After
hatching,
the
larvae
enter
the
plants
and
mine
the
stems
and petioles.
The
role
of
glucosinolates
in
the
feeding
behaviour
of
this
species
is
unclear.
Although
glucosinolate
levels
in
the
cotyledons
of
rape
are
sim-
ilar
to
those
in
the
seed
(Glen
et
al.,
1990),
Williams
(1989)
found
that
P
chrysocephala
damage
to
cotyle-
dons
of
different
rape
cultivars
did
not
relate
to
seed
glucosinolate
content.
In
previous
experiments,
feed-
ing
of
P.
chrysocephala
was
restricted
to
plants
contain-
ing
glucosinolates
but
the
addition
of
glucosinolate
to
non-host
plants
did
not
induce
the
beetle
to
feed
(Bart-
let
&
Williams,
1991).
It
was
thought
that
this
was
because
the
non-host
plants
contained
other
chemicals
which
were
inhibitory
to
feeding.
This
study
used
an
artificial
substrate
to
investi-
gate
how
glucosinolates
affect
the
feeding
of
adult
P.
chrysocephala.
Glucosinolates
were
tested
individ-
ually
and
in
mixtures.
The
effect
of
sugars
on
feeding
and
the
interaction
between
sugars
and
glucosinolates
were
also
investigated.
78
Materials
and
methods
Insects
Adult
beetles
were
collected
from
crops
of
win-
ter
oilseed
rape
and
kept
at
10
°C,
L12:D12.
Between
experiments
they
were
kept
in
plastic
boxes
(210
x
100
x
80
mm)
and
fed
on
oilseed
rape
leaves
(cv.
Ariana).
Chemicals
Rape
extract.
Lyophilized
oilseed
rape
leaves
(cv.
Ari-
ana;
5
g)
were
stirred
with
chloroform
(200
ml)
for
2
h,
the
mixture
filtered,
and
the
filtrate
concentrated
to
dryness
using
a
rotary
evaporator.
The
leaf
mate-
rial
was
then
placed
in
methanol
(200
ml)
and
the
above
procedure
repeated.
Remaining
leaf
material
was
stirred
in
water
(200
ml)
for
2
h,
filtered
and
the
filtrate
freeze
dried.
The
three
filtrates
were
stored
frozen
until
required,
when
they
were
combined
in
50%
aqueous
ethanol
solution
(10
ml)
and
added
to
agar
(300
ml).
Glucosinolates.
All
glucosinolates
used
were
potas-
sium
salts.
Sinigrin
(Allylglucosinolate)
98%
was
purchased
from
Aldrich
Chemical
Co.
Ltd.
Pro-
goitrin
((2R)-Hydroxybut-3-enylglucosinolate)
96%
was
purchased
from
Biocatalysts
Ltd.
Glu-
cotropaeolin
(Benzylglucosinolate),
gluconapin
(But-
3-enylglucosinolate)
and
gluconasturtiin
(Phenethyl-
glucosinolate)
were
synthesised
using
published
pro-
cedures
(Ettlinger
&
Hodgkins,1955;
Ettlinger
&
Lundeen,
1957;
Benn,
1963;
Benn,
1964;
Kjaer
&
Rosendal
Jensen,
1968).
Glucobrassicin
(Indo1-3-
ylglucosinolate)
was
synthesised
by
a
route
similar
to
that
described
by
Viaud
&
Rollin
(1990)
except
that
the
indole
ring
was
protected
by
N-acetylation,
enabling
pure
crystalline
material
to
be
obtained
at
each
stage
of
the
synthesis.
The
N-acetyl
protecting
group
was
removed
at
the
same
time
as
the
glucose-
0-acetyl
groups
(D.
Parsons,
unpubl.).
Epiprogoitrin
was
obtained
by
extraction
of
Crambe
abyssinica
seed
with
boiling
methanol
(Daxenbichler
et
al.,
1965).
After
clarification
using
celite,
the
crude
material
was
purified
by
flash
chromatography
from
reverse
phase
C18-
silica
(Hanley,
et
al.,
1983).
Each
glucosinolate
was
converted
to
its
crystalline
tetramethyl
ammonium
salt,
the
tetramethyl
ammonium
ion
being
exchanged
for
the
potassium
ion
when
required.
Desulphated
glu-
cosinolates
were
checked
for
purity
by
HPLC
(Sang
&
Truscott,
1984).
Individual
glucosinolates
were
tested
at
1
mM.
Three
different
glucosinolate
mixtures
were
made
up
(Table
1).
Mixture
A
contained
the
five
glu-
cosinolates
used
in
dual
choice
experiments
in
equal
proportions.
Mixture
B
was
made
up
using
the
propor-
tions
and
concentrations
of
glucosinolates
found
in
the
leaves
of
young
rape
(Milford
et
al.,
1989).
We
were
unable
to
obtain
all
the
glucosinolate
types
present
in
rape
leaves
but
accounted
for
76%
of
the
total
glucosi-
nolate
content
(Table
1).
Mixture
C
contained
the
same
glucosinolates
as
mixture
B,
at
the
same
total
concen-
tration
but
in
equal
proportions.
Sugars.
Analysis
of
the
sugar
content
of
leaves
of
young
rape
(Ariana
cv.)
by
gas
chromatography
(as
in
Paul
et
al.,
1990)
found
fructose,
glucose,
rham-
nose,
inositol,
sucrose
and
raffinose.
Fructose
was
the
most
abundant
sugar
and
raffinose
the
least.
The
total
concentration
of
sugars
was
80
mM
(E.
Bartlet
&
J.
Fieldsend,
unpubl.).
Sugars
were
tested
at
40
mM.
Fructose,
glucose
and
rhamnose
(Sigma)
and
sucrose
(BDH)
were
used,
raffinose
and
inositol
were
not
test-
ed.
Feeding
experiments
Dual
choice
experiments.
Agar
was
used
as
the
feed-
ing
substrate.
Test
materials
were
stirred
into
agar
(2%)
that
had
cooled
to
below
70
°C.
Different
treatments
were
placed
in
either
of
two
equal
compartments
in
Petri
dishes
(9
cm
diameter).
The
agar
was
poured
until
its
surface
was
level
with
the
top
of
the
central
division
of
the
dish
(9
mm
deep)
and
allowed
to
solidify.
Five
male
and
five
female
beetles
were
starved
for
three
days
before
being
placed
in
each
dish.
The
dishes
were
then
covered
with
a
lid
with
a
nylon
mesh
(aperture
0.5
mm)
lined
hole
(5
cm
diameter).
After
three
days
the
amount
of
feeding
was
assessed
by
examining
the
agar
under
a
binocular
microscope,
using
transmitted
light,
and
counting
the
bite
marks
made
by
the
beetles.
Their
mandibles
left
paired,
triangular
marks
in
the
agar.
Occasionally,
more
intensive
feeding
occurred
in
which
case
an
area
of
feeding
1
mm
2
was
estimated
as
20
bite
marks.
Feeding
only
occurred
in
the
top
0.5
mm
of
the
agar.
The
treatments
were
replicated
10-25
times.
A
preference
index
was
calculated
as
in
Nielsen
et
al.
(1979);
Q
=
(A
B)/(A
+
B),
where
A
and
B
=
the
number
of
bite
marks
from
treatment
A
and
B,
respectively.
Q
was
not
calculated
for
replicates
where
less
than
five
bite
marks
were
found.
Differences
between
treatments
were
assessed
using
Wilcoxon's
signed
rank
test
(Siegel,
1956).
Comparisons
between
79
Table
1.
Composition
of
glucosinolate
mixtures
Glucosinolate
Concentration
(µM)
Mixture
A
1
Mixture
B
2
Mixture
C
3
Sinigrin
200
Glucotropaeolin
200
Gluconapin
200
34
242.5
Gluconasturtin
200
350
242.5
Glucobrassicin
200
650
242.5
Progoitrin
128
242.5
Epiprogoitrin
51
242.5
Total
glucosinolate
contentration
1000
1213
1212.5
1
5
glucosinolates
previously
tested
in
dual
choice
tests
2
glucosinolate
profile
of
young
rape
leaves
3
the
same
mixture
as
B
but
component
glucosinolates
in
equal
proportions
no.
of
bitemarks
350
300
250
200
sinigrin
150
--
H
sucrose
100
50
0
-6
-5
-4
-3
-2.3
log
concentration
(M)
Fig.
I.
Effect
of
concentration
of
sinigrin
or
sucrose
added
to
agar
on
the
no.
of
bitemarks.
Dual
choice
experiments.
value
1
0.8
0.6
0.4
0.2
0
—sinigrin
-0.2
+
sucrose
-0.4
-6
-5
-4
-3
-2.3
log
concentration
(M)
Fig.
2.
Effect
of
concentration
of
sinigrin
and
sucrose
on
the
feeding
preference
index,
Q.
Dual
choice
experiments.
different
Q
values
were
only
made
within
experiments.
Multiple
choice
experiments.
Agar
was
poured
into
plastic
cylinders
(1.5
cm
diameter,
1
cm
deep)
and
allowed
to
cool.
One
cylinder
of
each
treatment
was
placed,
in
random
order,
around
the
edge
of
a
Petri
dish
(9
cm
diameter)
lined
with
damp
filter
paper,
and
exposed
to
feeding
by
the
beetles
as
for
the
dual
choice
experiments.
Each
set
of
treatments
was
replicated
22-
30
times.
The
results
were
normalised
by
taking
log-
arithms
(log
io
(n
+
1))
and
analysed
by
ANOVA
using
Genstat
5
(Payne
et
al.
1987).
Treatment
effects
were
split
into
two
orthogonal
contrasts
which
compared
feeding
on
the
untreated
agar
to
overall
feeding
on
treated
agar,
and
differences
between
the
treatments,
respectively.
Results
Feeding
within
treatments
was
highly
variable.
Because
the
selectivity
of
the
beetles
decreased
as
they
aged,
comparisons
should
not
be
made
between
results
obtained
in
different
experiments.
The
presence
of
rape
extract
in
the
agar
induced
high
levels
of
feeding
(experiment
(i),
Table
2).
All
80
Table
2.
Feeding
of
Psylliodes
chrysocephala
on
agar,
to
which
chemical
components
of
oilseed
rape
have
been
added,
in
dual
choice
experiments
Experiment
Addition(s)
to
agar
on
side
A
Addition(s)
to
agar
on
side
B
n
Mean
Q
value±s.e.
Mean
no.
of
bites±s.e.
(i)
Rape
extract
None
7
0.9810.02*
255132
(ii)
Sinigrin
(1
mM)
None
8
0.4710.32
32118
Glucotropaeolin
(1
mM)
None
8
0.7110.14*
93142
Gluconapin
(1
mM)
None
8
0.6010.16*
128140
Gluconasturtiin
(1
mM)
None
8
0.8510.07*
71114
Glucobrassicin
(1
mM)
None
7
0.8410.12*
177141
(iii)
Fructose
(40
mM)
None
10
0.8710.06*
58119
Glucose
(40
mM)
None
10
0.6610.14*
168146
Sucrose
(40
mM)
None
10
0.9710.03*
344176
Rhamnose
(40
mM)
None
10
0.0110.36
23111
(iv)
Sucrose
(40
mM)
None
10
0.99±0.01*
115137
Sinigrin
(1
mM)
None
10
0.7610.16*
53112
Sucrose
(40
mM)
+
Sinigrin
(1
mM)
None
10
0.9710.03*
4801134
(v)
Fructose
(40
mM)
None
10
0.7010.15
46135
Sinigrin
(1
mM)
None
10
0.4210.30
2010
Fructose
(40
mM)
+
Sinigrin
(1
mM)
None
10
0.9710.02*
207155
(vi)
Glucosinolate
mixture
A
None
12
0.6610.10*
113129
Glucosinolate
mixture
A
Glucobrassicin
(1
11[1M)
10
0.7410.07*
84115
(vii)
Glucosinolate
mixture
B
Glucosinolate
mixture
C
25
—0.0810.08
92115
*
Feeding
on
side
A
significantly
higher
than
on
side
B
P<0.05
glucosinolates
tested
increased
feeding
on
the
agar
(experiment
(ii),
Table
2;
experiment
(viii),
Table
3).
Sinigrin
was
tested
at
1
mM
in
four
different
experi-
ments
(experiment
(ii),
(iv)
&
(v),
Table
2,
and
Fig.
1
&
2)
but
only
in
one
experiment
(experiment
(iv))
did
it
have
a
significant
effect
on
feeding
at
this
concen-
tration.
Sinigrin
had
the
least
effect
on
feeding
in
both
dual
and
multiple
choice
experiments
and
glucobras-
sicin
the
most.
A
significant
difference
(P<0.001)
in
effectiveness
was
found
amongst
the
different
glucosi-
nolates
when
they
were
compared
in
a
multiple
choice
experiment
(experiment
(viii),
Table
3).
All
sugars
except
rhamnose
stimulated
feeding
(experiment
(iii),
Table
2).
Sucrose
was
the
most
effec-
tive
sugar
tested.
It
was
so
strongly
preferred
in
a
mul-
tiple
choice
experiment
(experiment
(ix),
Table
3)
that
little
feeding
occurred
on
all
the
other
treatments.
When
sinigrin
was
presented
with
sucrose
or
fruc-
tose
a
synergistic
effect
on
the
number
of
bite
marks
the
beetles
made
was
found
(experiments
(iv)
&
(v),
Table
2).
Glucosinolate
mixture
A
was
significantly
more
effective
than
either
agar
alone
or
glucobrassicin
(experiment
(vi),
Table
2).
No
difference
was
found
between
the
amount
of
feeding
induced
by
glucosino-
late
mixtures
B
and
C
(experiment
(vii),
Table
2).
The
number
of
bitemarks
made
increased
with
the
concentration
of
both
sinigrin
and
sucrose
(Fig.
1).
A
large
increase
in
the
proportion
of
bites
occurred
on
the
treated
side
of
the
dish
(Q)
when
the
concentration
of
both
sinigrin
and
sucrose
was
raised
from
1
µ
M
to
10
it
M
but
increased
little
when
the
concentration
was
raised
further
(Fig.
2)
Sinigrin
increased
feeding
significantly
only
at
the
highest
concentration
tested
(5
mM).
Sucrose
increased
feeding
at
all
concentra-
tions
above
1
it
M.
Discussion
When
tested
using
an
artificial
substrate,
glucosino-
lates
stimulated
the
feeding
of
P.
chlysocephala.
Since
glucosinolates
do
not
readily
decompose
at
neutral
pH
81
Table
3.
Feeding
of
Psylliodes
chrysocephala
on
agar,
to
which
glucosinolates
or
sugars
have
been
added,
in
multiple
choice
experiments
Experiment
Addition(s)
to
agar
Mean
(log.no.
of
bites
+
1)
Mean
no
of
bites
(viii)
Sinigrin
1.55
47.5
Glucotropaeolin
1.82
76.3
Gluconapin
1.73
78.5
Gluconasturtiin
1.73
80.7
Glucobrassicin
2.02
119.8
None
1.30
36.1
s.e.d.
0.07
Fructose
0.11
1.5
(ix)
Glucose
0.03
0.1
Sucrose
1.58
70.0
Rhamnose
0.17
1.4
None
0.11
1.1
s.e.d.
0.12
-
*
Significant
difference
in
effectiveness
of
different
glucosinolates
P<0.01
n
=
30
*
Significant
difference
in
effectiveness
of
different
sugars
P<0.001
n
=
22
levels
in
the
absence
of
myrosinase
(Larsen,
1981),
this
work
demonstrates
the
role
of
intact
glucosinolates
rather
than
metabolites
of
the
glucosinolates,
such
as
isothiocyanates.
Different
glucosinolates
differed
sig-
nificantly
in
the
amount
of
feeding
they
induced,
with
glucobrassicin
being
the
most
effective
and
sinigrin
the
least.
This
demonstrates
the
value
of
investigat-
ing
a
greater
range
of
glucosinolates
than
those
readily
available
commercially.
Larsen
et
al.
(1985)
found
glucobrassicin
to
be
the
most
potent
glucosinolate
to
stimulate
the
feeding
of
flea
beetles
of
the
genus
Phyl-
lotreta.
Glucobrassicin
is
the
dominant
glucosinolate
in
the
leaves
of
early
vegetative
rape,
(Milford
et
al.,
1989).
It
seems
unlikely,
however,
that
P
chrysocephala
discriminates
between
different
cruciferous
host
plants
on
the
basis
of
glucosinolate
profiles,
since
the
glucosi-
nolate
profile
of
rape
was
no
more
effective
at
stimulat-
ing
feeding
than
a
mixture
of
the
same
glucosinolates
in
equal
proportions.
Furthermore,
there
is
little
dif-
ference
between
the
glucosinolate
profiles
of
Brassica
nigra
and
Alliaria
petiolata
(Nielsen
et
al.,
1979)
but,
in
dual
choice
tests
with
P.
chrysocephala,
oilseed
rape
was
strongly
preferred
to
A.
petiola
whilst
feeding
on
oilseed
rape
and
B.
nigra
was
approximately
equal
(Bartlet
&
Williams,
1991).
Although
the
glucosino-
late
profiles
of
cruciferous
plants
are
reported
as
being
relatively
stable
(Louda
&
Mole,
1991),
several
stud-
ies
have
shown
that
damage
to
oilseed
rape,
including
feeding
damage
by
P.
chrysocephala,
induces
major
changes
in
glucosinolate
composition
(Koritsas
et
al.,
1989;
Birch
et
al.,
1990;
Bodnaryk
et
al
,
1992).
In
tests
using
Phyllotreta
armoraciae,
Nielsen
et
al.
(1979)
could
not
relate
the
acceptability
of
12
spp.
of
crucifer-
ous
plants
to
the
stimulatory
activity
of
glucosinolate
mixtures
extracted
from
those
plants
and
concluded
that
other
chemicals
acted
as
either
feeding
stimulants
or
feeding
inhibitors
in
host
plant
selection.
All
sugars
except
rhamnose
stimulated
the
feed-
ing
of
P.
chrysocephala.
Sugars
are
phagostimulants
for
most
insects,
including
specialist
crucifer
feeders.
Sucrose
was
found
to
be
a
more
effective
feeding
stim-
ulant
than
sinigrin
for
both
Pieris
brassicae
and
Brevi-
coryne
brassicae
(Schoonhoven,
1972).
Different
components
had
an
interactive
effect
on
feeding
behaviour
when
presented
together.
Both
sucrose
and
fructose
had
a
synergistic
effect
on
the
amount
of
feeding
that
occurred
in
the
presence
of
sin-
igrin.
Sucrose
has
previously
been
reported
to
act
syn-
ergistically
with
glucosinolates
for
Meligethes
aeneus,
Plutella
maculipennis,
and
a
number
of
polyphagous
and
oligophagous
aphid
spp.
(Charpentier
&
Charp-
entier,
1986;
Klingauf,
1987).
Ma
(1972)
reported
that
sucrose
and
sinigrin
had
a
synergistic
effect
on
the
feeding
of
Pieris
brassicae,
but
Blom
(1978)
only
found
this
effect
to
be
additive.
If
the
effect
of
the
dif-
ferent
glucosinolates
was
additive,
mixture
A
would
be
expected
to
be
less
stimulatory
than
pure
glucobras-
82
sicin,
at
the
same
total
glucosinolate
concentration,
since
glucobrassicin
was
the
most
potent
component
of
mixture
A.
However,
mixture
A
was
fed
on
sig-
nificantly
more
than
glucobrassicin,
so
the
different
glucosinolates
in
mixture
A
must
be
interacting
syner-
gistically.
Although
the
amount
of
feeding
that
occurred
increased
with
sinigrin
concentration,
a
5000
fold
increase
in
concentration
led
to
only
a
seven
fold
increase
in
feeding.
Glen
et
al.
(1990)
found
that
the
glucosinolate
content
of
rape
seedlings
of
single
low
cvs
was
only
four
to
ten
times
greater
than
that
of
dou-
ble
low
cvs.
Given
the
variability
of
P
chrysocephala
feeding,
this
could
explain
the
lack
of
a
relationship
between
P
chrysocephala
damage
to
rape
cotyledons
and
the
seed
glucosinolate
content
of
a
range
of
rape
cultivars
(Williams,
1989).
Sinigrin
was
used
in
these
experiments
because
it
is
the
only
glucosinolate
avail-
able
in
sufficient
quantities
commercially,
but
it
was
the
least
potent
glucosinolate
tested.
Other
glucosi-
nolates,
or
glucosinolate
mixtures
could
show
more
pronounced
differences
in
feeding
with
concentration.
These
results
make
it
clear
that
glucosinolates,
although
highly
effective
as
feeding
stimulants
for
P.
chrysocephala,
are
not
prerequisites
for
feeding
by
this
beetle.
However,
plants
lacking
glucosinolates
were
rejected
by
P.
chrysocephala
in
previous
feed-
ing
experiments
(Bartlet
&
Williams,
1991);
further
work
is
under
way
to
investigate
whether
this
rejec-
tion
occurred
because they
contained
chemical
feeding
inhibitors.
Acknowledgments
We
thank
Jennifer
Campbell
for
her
help
with
this
work,
Alan
Todd
for
his
advice
and
help
with
the
sta-
tistical
analyses,
Jane
Fieldsend
for
sugar
analysis
of
rape
leaves,
and
the
Ministry of
Agriculture,
Fisheries
and
Food
for
their
financial
support
of
this
project.
References
Bartlet,
E.
&
I.
H.
Williams,
1991.
Factors
restricting
the
feeding
of
the
cabbage
stem
flea
beetle
(Psylliodes
chrysocephala).
Ento-
mologia
Experimentalis
et
Applicata
60:
233-238.
Benn,
M.
H.,
1963.
A
new
mustard
oil
glucoside
synthesis.
Canadian
Journal
of
Chemistry
41:
2836-2838.
Benn,
M.H.,
1964.
The
synthesis
of
gluconasturtiin.
Journal
of
the
Chemical
Society
4072-4074.
Birch,
A.
N.
E.,
D.
W.
Griffiths
&
W.
H.
M.
Smith,
1990.
Changes
in
forage
and
oilseed
rape
root
glucosinolates
in
response
to
attack
by
turnip
root
fly
(Delia
floralis).
Journal
of
the
Science
of
Food
and
Agriculture
51:
309-320.
Blom,
F.,
1978.
Sensory
activity
and
food
intake:
a
study
of
input-
output
relationships
in
two
phytophagous
insects.
Netherlands
Journal
of
Zoology
28:
277-340.
Bodnaryk,
R.
P.,
1992.
Effects
of
wounding
on
glucosinolates
in
the
cotyledons
of
oilseed
rape
and
mustard.
Phytochemistry
31:
2671-2677.
Bromand,
B.,
1990.
Diversities
in
oilseed
rape
growing
within
the
Western
Paleartic
Regional
Section.
Bulletin
SROP/WPRS
-
Proceedings
of
the
OILB
working
group
'Integrated
Control
in
Oilseed
Rape',
Malmo,
Sweden,
21-22
March
1988,
pp.
7-31.
Charpentier,
R.
&
B.
Charpentier,
1986.
Resistance
to
insects
in
cruciferous
oil
crops:
influence
of
chemical
stimuli
on
the
host
plant
selection
of
the
pollen
beetle
(Meligethes
aeneus).
Eucarpia
IOBC
4th
meeting,
Hundestad,
Denmark,
p.
83.
Daxenbichler,
M.
E.,
C.
H.
Van
Etten
&
I.
A.
Wolff,
1965.
High
performance
liquid
chromatography
(HPLC)
of
desulphoglucosi-
nolates.
Biochemistry
4:
318.
Ettlinger,
M.
G.
&
J.
E.
Hodgkins,
1955.
The
mustard
oil
of
rape
seed,
allylcarbinyl
isothiocyanate,
and
synthetic
isomers.
Journal
of
the
American
Chemical
Society
77:
1831-1836.
Ettlinger,
M.
G.
&
A.
J.
Lundeen,
1957.
First
synthesis
of
a
mustard
oil
glucoside;
the
enzymatic
lossen
rearrangement.
Journal
of
the
American
Chemical
Society
79:
1764-1765.
Fraenkel,
G.,
1969.
Evaluation
of
our
thoughts
on
secondary
plant
substances.
Entomologia
Experimentalis
et
Applicata
12:
473-
486.
Glen,
D.
M.,
H.
Jones
&
J.
K.
Fieldsend,
1990.
Damage
to
oilseed
rape
seedlings
by
the
field
slug
Deroceras
reticulatum
in
relation
to
glucosinolate
concentration.
Annals
of
Applied
Biology
117:
197-207.
Hanley,
A.
B.,
R.
K.
Heaney
&
G.
R.
Fenwick,
1983.
Improved
isolation
of
glucobrassicin
and
other
glucosinolates.
Journal
of
the
Science
of
Food
and
Agricultural
34:
869-873.
Hsaio,
T.
H.,
1985.
Feeding
behaviour.
In:
C.
A.
Kerkut
&
L.
I.
Gilbert
(eds)
Comprehensive
insect
physiology,
biochem-
istry
and
pharmacology.
Pergamon
Press
Oxford,
pp.
471-512.
Kjaer,
A.
&
S.
Rosendal
Jensen,
1968.
Synthesis
of
the
3-
Butenylglucosinolate
ion.
Acta
Chemica
Scandinavica
22:
3324-
3326.
Klingauf,
F.
A.,
1987.
Host
plant
finding
and
acceptance.
In:
A.
K.
Minks
&
P.
Harrewijn
(eds),
Aphids:
their
biology,
nat-
ural
enemies
and
control.
Elsevier,
Amsterdam,
pp.
209-223.
Koritsas,
V.
M.,
J.
A.
Lewis
&
G.
R.
Fenwick,
1989.
Accumulation
of
indole
glucosinolates
in
Psylliodes
chrysocephala
L.
infested
or
damaged
tissues
of
oilseed
rape
(Brassica
napus
L.).
Experientia
45:
493-495.
Larsen,
P.
0.,
1981.
Glucosinolates.
In:
P.
K.
Stumpf
&
E.
E.
Conn
(eds)
The
biochemistry
of
plants
vol.
7.
Academic
press,
pp.
501-
525.
Larsen,
L.
M.,
J.
K.
Nielsen,
A.
Ploger
&
H.
Sorenson,
1985.
Responses
of
some
beetle
species
to
varieties
of
oilseed
rape
and
to
pure
glucosinolates.
In:
H.
Sfenson
(ed),
Advances
in
the
production
and
utilization
of
cruciferous
crops.
Martinus
Nijhoff
Publishers,
pp.
230-244.
Louda,
S.
&
S.
Mole,
1991.
Glucosinolates:
chemistry
and
ecolo-
gy.
In:
G.
A.
Rosenthal
&
M.
R.
Berenbaum
(eds)
Herbivores
and
their
interactions
with
secondary
plant
metabolites
2E
(1):
chemical
participants.
Academic
Press,
pp.
123-164.
83
Ma,
W.
C.,
1972.
Dynamics
of
feeding
responses
of
Pieris
brassi-
cae
as
a
function
of
chemosensory
input.
Mededelingen
Land-
bouwhogeschool
Wageningen.
162
pp.
Milford,
G.
F.
J.,
J.
K.
Fieldsend,
A.
J.
R.
Portert
&
C.
J.
Rawlinson,
1989.
Changes
in
glucosinolate
concentration
during
the
vegeta-
tive
growth
of
single
and
double
low
cultivars
of
winter
oilseed
rape.
Aspects
Applied
Biology
23:
83-90.
Mitchell,
B.
K.,
1988.
Adult
leaf
beetles
as
models
for
exploring
the
chemical
basis
of
host
plant
selection.
Journal
of
Insect
Physiol-
ogy
34:
213-225.
Nielsen,
J.
K.,
D.
L.
Dalgaar,
L.
M.
Larsen
&
H.
Sorenson,
1979.
Host
plant
selection
of
the
horseradish
flea
beetle
Phyllotreta
armoraciae
(Coleoptera:
Chrysomelidae):
Feeding
responses
to
glucosinolates
from
several
crucifers.
Entomologia
Experimen-
talis
et
Applicata
25:
227-239.
Payne,
R.
W.
&
Genstat
5
committee,
1987.
Genstat
5
Reference
Manual.
Clarendon
Press,
Oxford.
Paul,
M.
J.,
D.
W.
Lawlor
&
S.
P.
Driscoll,
1990.
The
effect
of
temperature
on
photosynthesis
and
carbon
fluxes
in
sunflower
and
rape.
Journal
of
Experimental
Botany
41:
547-555.
Rodman,
J.
E.
&
F.
S.
Chew,
1980.
Phytochemical
correlates
of
herbivory
in
a
community
of
native
and
naturalised
Cruciferae.
Biochemical
Systematics
and
Ecology
8:
43-50.
Sang,
J.
P.
L.
&
R.
J.
W.
Truscott,
1984.
Liquid
chromatography
determination
of
glucosinolates
in
rapeseed
as
desulfoglucosino-
lates.
Journal
of
the
Association
of
Official
Analytical
Chemists
67:
829-833.
Schoonhoven,
L.
M.,
1972.
Secondary
plant
substances
and
insects.
In:
V.
C.
Runeckles
&
T. T.
So
(eds),
Recent
advances
in
phyto-
chemistry,
5.
Academic
press,
pp.
197-224.
Siegel,
S.,
1956.
Nonparametric
statistics
for
the
behavioural
sci-
ences.
McGraw-Hill,
New
York,
pp.
75-83.
Viaud,
M.
C.
&
P.
Rollin,
1990.
First
synthesis
of
an
indole
glucosi-
nolate.
Tetrahedron
Letters
531:
1417-1418.
Williams,
I.
H.,
1989.
Pest
incidence
on
single
low
and
double
low
oilseed
rape
cultivars.
Aspects
of
Applied
Biology
23:
277-286.
Winfield,
A.
L.,
1992.
Management
of
oilseed
rape
pests
in
Europe.
Agricultural
Zoology
Review
5:
51-95.