A new technique for estimating density of the field slug (Deroceras reticulatum (Muller))


Ferguson, C.M.; Barratt, B.I.P.; Jones, P.A.

Monograph - British Crop Protection Council 41: 331-336

1989


A quantitaive method developed in New Zealand for estimating the density of the slug Deroceras reticulatum in the field is described. A defined-area trap (DAT) was used in pasture and compared with accepted methods of cold water extraction from soil samples, and with surface searching. DATs gave results comparable to soil sampling methods but with considerable savings in time and labour. Surface searching understimated slug density.

1989
BCPC
MONO.
No.
41
SLUGS
AND
SNAILS
IN
WORLD
AGRICULTURE
A
NEW
TECHNIQUE
FOR
ESTIMATING
DENSITY
OF
THE
FIELD
SLUG
(DEROCERAS
RETICULATUM
(MULLER))
C.M.
FERGUSON,
B.I.P
BARRATT,
P.A.
JONES
Invermay
Agricultural
Centre,
Ministry
of
Agriculture
and
Fisheries,
Private
Bag,
Mosgiel,
New
Zealand
ABSTRACT
A
quantitative
method
for
estimating
slug
density
in
the
field
is
described.
A
defined-area
trap
(DAT)
was
used
in
pasture
and
compared
with
accepted
methods
of
cold
water
extraction
from
soil
samples,
and
with
surface
searching.
DATs
gave
results
comparable
with
soil
sampling
methods
but
with
considerable
savings
in
time
and
labour.
Surface
searching
underestimated
slug
density.
INTRODUCTION
The
need
for
an
accurate,
quantitative
sampling
method
for
agricultural
pest
species
of
slugs
was
emphasised
by
Barker
(1987)
who
documented
the
shortcomings
of
traditional
trapping
methods
and
called
for
the
standardization
of
sampling
techniques.
Non-quantitative
trapping
methods,
which
depend
upon
the
attraction
of
slugs
to
refuge
sites
(eg
Schrim
and
Byers,
1980)
are
popular
because
of
ease
of
use.
Soil
and
vegetation
sampling
is
more
labour
intensive,
requiring
the
extraction
of
slugs
from
soil
samples
by
hand
sorting,
by
soil
washing
and
flotation
or
by
flooding.
Hand
sorting
is
inefficient
(Thomas
1944)
and
soil
washing
and
flotation,
although
more
accurate
(South
1964,
Hunter
1968)
is
time-consuming.
Because
of
this
South
(1964)
developed
a
method
of
flooding
soil
samples
slowly
over
time.
South's
"Cold
Water
Extraction
Process"
which
recovers
up
to
99%
of
Deroceras
reticulatum
still
requires
collection,
transport,
processing
and
disposal
of
large
volumes
of
soil.
Barker
&
Pottinger
(1982)
desribed
a
modification
of
South's
method
where
soil
cores
were
flooded
to
extract
slugs
instead
of
larger
turves.
A
quantitative
technique
known
as
defined-area
trapping
has
been
developed
in
the
course
of
slug
control
research
at
Invermay
Agricultural
Centre,
Mosgiel,
New
Zealand
(Ferguson
et
al.,
1989)
and
this
method
incorporates
the
convenience
of
traps
with
the
efficiency
of
soil
sampling.
The
objectives
of
the
experiments
described
here
were
to
assess
the
accuracy
of
defined-area
trapping
in
pasture
at
a
number
of
sites
and
to
compare
it
with
other
current
quantitative
techniques.
METHODS
Sixteen
sites
located
on
pasture
near
Mosgiel,
New
Zealand
which
varied
in
soil
type,
aspect,
vegetative
composition
and
stock
management
were
used
for
this
study.
Sampling
methods
Defined
area
trapping
The
defined-area
trap
(DAT)
consisted
of
a
galvanised
iron
ring
357
mm
in
diameter
and
150
mm
deep
(Fig.la).
A
circular
steel
template,
held
331
a
--
b
i`
s
•a
d
;7)
-.e•
-
e•st
711IF
Y•fr.
-
"t'
••'!
.1,
'No
eAr
f
1
"
7
1
'
;kr:
,
-
-
•t•
4
,
e.e
,
JR,
4
WC'
Hardboard
Galvanised
iron
ring
Wet
sacking
Soil
surface
Wire
pins
_J
Scale
10cm
Figure
1.
Defined-area
trap
apparatus:
(a)
trap
components;
(b)
ring
inserted
into
steel
template;
(c)
trap
being
driven
into
the
ground;
(d)
sacking
inside
trap;
(e)
trap
fully
assembled;
(f)
verticle
section
through
trap.
in
position
on
top
of
the
ring
by
internal
and
external
flanges
(Fig.
lb)
enabled
it
to
be
driven
approx
50
mm
into
the
soil thereby
defining
the
sample
unit
of
0.1m
2
(Fig.lc).
A
piece
of
wet
hessian
sacking
placed
on
the
ground
within
the
trap
(Fig.1d)
acted
as
refuge
for
the
slugs
to
shelter
beneath
and
also
maintained
high
humidity within
the
trap
encouraging
slugs
to
remain
above
ground.
A
square
of
hardboard
slightly
larger
than
the
trap
diameter
with
a
white
painted
surface
uppermost
was
placed
on
top
of
the
ring
and
held
in
place
at
the
corners with
wire
pegs
pushed
into
the
ground
(Fig.le,
f).
The
white
surface
was
intended
to
help
reduce
temperatures
inside
the
trap.
Deroceras
reticulatum,
although
utilising
the
burrows
of
other
animals,
cracks
and
crevices
in
the
ground
etc.,
cannot
burrow
(Stephenson,
1975)
and
so
slugs
are
unable
to
gain
entry
to
or
exit
from
the
DAT
via
the
lower
rim.
The
hardboard
fitted
closely
onto
the
upper
rim
and
therefore
prevented
slug
movement
in
or
out
of
the
trap.
Furthermore,
slugs
appeared
reluctant
to
travel
over
galvanised
iron
and
so
were
discouraged
from
trying
to
find
ways
into
or
out
of
the
trap.
If
the
vegetation
exceeded
50
-
60
mm
in
height,
it
was
trimmed
around
the
outside
of
the
ring
to
provide
an
extra
safeguard
against
slug
movement
into
the
trap.
Where
the
vegetation
within
the
ring
was
long
it
was
cut,
searched
and
discarded
to
facilitate
slug
counts.
The
traps
were
serviced
every
2
-
3
days
by
removing
the
hardboard
cover
and
searching
the
sacking
and
the
area
within
the
ring.
Any
slugs
found
were
recorded
and
removed,
essentially
'trapping
out'
the
defined
area.
The
sacking
was
remoistened
when
necessary.
During
Sep
-
Oct
1987,
25
traps
were
set
at
sites
1
-
4;
in
May
1986
20
traps
at
site
5,
and
in
Nov
-
Dec
1988,
8
traps
at
sites
6
-
16
were
set
up
and
serviced
as
above
over
an
8
day
period.
At
sites
1
-
5
after
trapping
the
turf
was
lifted
to
a
depth
of
100
mm
and
any
slugs
remaining
extracted
using
the
flooding
technique
detailed
below.
Turf
sampling
Twenty-five
0.1
m
2
turves
100
mm
deep
from
sites
1
-
4,
20
from
site
5
and
8
from
sites
6
-
16
were
dug
and
placed
individually
in
large
plastic
bags
which
were
tightly
closed.
In
the
laboratory
the
bags
were
opened,
water
added
until
the
turves,
vegetation
uppermost,
were
standing
in
30
mm
of
water
and
then
reclosed.
The
following
day
the
turves
were
searched,
slugs
removed
and
counted
and
water
added
to
a
level
of
70
mm.
After
2
days
the
turves
were
searched
again
as
above
and
the
water
level
raised
until
it
was
1
cm
from
the
soil
surface.
Slugs
were
counted
for
the
final
time
after
two
days.
Core
Sampling
At
sites
1
-
4
318
cores
and
from
site
5
182
cores
100
mm
in
diameter
and
100
mm
deep
were
taken
and
stored
in
separate
closed
plastic
bags.
The
core
numbers
from
sites
1
-
4
gave
a
sampling
area
equivalent
to
the
previous
two
methods.
In
the
laboratory
the
bags
were
opened
and
the
cores
within
the
bags
were
placed
individually
in
200
mm
x
200
mm
x
125
mm
deep
plastic
containers.
The
cores,
lying
on
their
sides,
were
flooded
up
to
30
cm
with
water
and
the
bags
closed.
The
same
procedure
as
for
the
large
turves
was
followed
with
gradual
flooding
of
the
cores
over
a
6-day
period.
Surface
searching
At
sites
1
-
4
25
x
0.1
m
2
randomly
selected
quadrats
of
pasture
were
hand-searched
and
slugs
counted.
333
Slugs
found
by
all
methods
from
sites
1
-
4
were
measured
and
allocated
to
one
of
four
size
classes;
very
small
(less
than
10
mm
fully
extended;
small
(11
-
20
mm);
medium
(21
-
30
mm)
and
large
(greater
than
30
mm).
Data
analysis
The
slug
sampling
techniques
were
subjected
to
a
comparison
of
methods
analysis
(Altman
and
Bland,
1983)
in
which
the
differences
between
two
methods
was
plotted
against
the
average
of
the
two
methods
to
determine
whether
the
differences
between
techniques
varied
with
population
size
and
whether
any
of
the
techniques
was
biased
towards
over
or
under
estimation
relative
to
other
techniques.
The
size
distribution
of
slugs
found
by
the
different
sampling
methods
at
sites
1
-
4
was
compared
using
generalised
linear
models
with
a
poisson
distribution
and
a
log
link
(McCullagh
and
Nelder
1983).
RESULTS
AND
DISCUSSION
At
sites
1
-
5
there
was
no
difference
between
mean
slug
density
estimated
by
DATs
and
the
cold
water
extraction
of
soil
samples
(Table
1).
Surface
searching
estimates
were
significantly
lower
at
sites
1
and
2.
At
sites
6
-
16,
where
DATs
and
cold
water
extraction
of
large
turves
were
compared,
there
was
no
sigificant
difference
between
mean
slug
density
estimated
by
the
two
methods
(DATs
25.8;
turves
23.6,
SEM
5.6).
These
data
are
included
with
those
from
sites
1
-
5
for
the
comparison
of
methods
analysis
(Fig.2)
which
showed
no
indication
that
either
of
the
soil
sampling
methods
consistently
over-
or
underestimated
slug
density
in
comparison
with
the
DATs
or
that
estimates
varied
with
population
size.
However,
surface
searching
both
underestimated
slug
density
at
sites
1
-
4
and
showed
a
tendency
for
the
differences
between
these
and
DAT
estimates
to
increase
in
magnitude
as
density
increased.
TABLE
1
Mean
slug
densities
(number/m
2
(S.E.M))
for
each
sampling
method
at
each
site.
Site
Defined
Area
Trapping
Cold
Water
Extraction
Large
Turves
Cores
Surface
Searching
1
43.3
(5.6)
38.4
(7.3)
36.8
(5.4)
4.4
(2.0)
2
25.2
(3.8)
30.0
(3.8)
20.4
(3.4)
3.6
(1.8)
3
8.8
(1.9)
4.0
(1.3)
2.8
(1.2)
2.0
(1.0)
4
8.8
(2.7)
6.8
(2.1)
9.0
(2.0)
4.0
(1.6)
5
31.5
(9.5)
34.0
(10.0)
35.8
(10.0)
Mean
23.5
22.6
19.8
3.5
LSR
=
1.82
(P<0.05)
*
least
significant
ratio
334
Di
ffe
re
nce
be
twee
n
me
t
ho
ds
17-
16-
15-
14-
13-
12-
11-
10-
9-
8-
7-
6-
5-
4-
3-
2-
1
-
0-4
-1
-
-2
-
-3-
-4
-
-5
Cold
water
extraction
of
the
turves
from
within
the
DATs
after
the
trapping
period
2
at
sites
1
-
5
yielded
a
further
3.6
slugs/m
2
at
site
1
and
2.8
slugs/m
at
site
2
but
no
more
at
sites
3,
4
or
5.
This
suggests
an
average
sampling
efficiency
of
95%.
There
were
no
significant
differences
between
the
size
class
distributions
of
slugs
recovered
using
DATs
and
soil
sampling
methods,
but
surface
searching
significantly
underestimated
the
proportion
of
slugs
in
the
smallest
size
class
(<
10
mm
long).
Defined-area
traps
and
large
turves
Defined-area
traps
and
cores
Di
ffe
re
nce
be
twe
e
n
me
t
ho
ds
Di
ffere
nce
be
twee
n
me
t
ho
ds
1
0
10
20
30
40
50
Average
of
methods
Figure
2.
Comparison
between
slug
density
estimates
using
defined-area
traps
with
soil
sampling
and
surface
searching
methods.
7
6
-
5
-
4
-
3
2
-
1
-
0-
-1-
-2
-
-3
-
-4
5
0
10
20
30
40
Average
of
methods
40-
30-
20-
10-
0
1
I
I
;
0
10
20
30
40
Average
of
methods
Defined-area
traps
and
surface
searching
In
practice
the
DATs
were
found
to
be
considerably
less
labour
intensive
than
the
soil
sampling
and
cold
water
extraction
processes.
To
335
install
the
traps,
service
and
remove
them
took
four
man-hours
per
site
at
sites
1-5.
The
large
turves
took
nine
man
hours
to
dig,
process
and
dispose
of
the
soil
and
the
cores
18
man-hours
to
sample,
process
and
dispose
of
the
soil
per
site.
The
use
of
DATs
avoided
the
need
to
transport
samples,
find
space
to
process
them
and
dispose
of
the
water-logged
soil
after
processing.
CONCLUSION
The
technique
of
estimating
slug
density
using
DATs
gave
results
comparable
with
those
obtained
by
the
accepted
techniques
of
soil
sampling
and
cold
water
extraction.
Surface
searching
in
the
field
was
considered
unsatisfactory
because
it
gave
comparatively
low
density
estimates
particularly
for
small
slugs.
The
advantages
of
the
defined-area
trapping
method
over
cold
water
extraction
were
that
many
more
samples
could
be
taken
because
there
was
less
sampling
time
involved,
there
was
no
requirement
for
transport
and
disposal
of
soil
samples
or
laboratory
storage
space
for
processing.
ACKNOWLEDGEMENT
The
authors
would
like
to
thank
Mr
P.D.
Johnstone
for
statistical
advice.
REFERENCES
Altman,
D.G.;
Bland,
J.M.
1983.
Measurement
in
Medicine:
the
Analysis
of
Method
Comparison
Studies.
The
Statistician
32,
307
-
317.
Barker,
G.M.
1987.
Biology
of
pest
slugs
and
their
significance
in
conservation-tillage
systems.
In
:
Proceedings
of
an
International
Symposium
on
Establishment
of
Forage
Crops
by
Conservation-Tillage
Methods
:
Pest
Management.
1987.
R.R.
Hill,
Jr.,
R.O.
Clements,
A.A.
Hower
Jr.,
T.A.
Jordon
and
K.E.
Zeiders
(Eds).
United
States
Regional
Pasture
Research
Laboratory,
University
Park,
Pennsylvania
pp.
83-106.
Barker,
G.M.;
Pottinger,
R.P.
1982.
Sampling
slug
populations
:
procedures
adopted
at
Ruakura.
Proceedings
of
the
3rd
Australasian
Conference
on
Grassland
Invertebrate
Ecology,
Adelaide
30
Nov
-
4
Dec
1981,
K.E.
Lee
(Ed.)
SA
Govt.
Printer,
Adelaide.
p.
398.
Ferguson,
C.M.;
Barratt,
B.I.P.;
Jones,
P.A.
(1989).
Control
of
the
grey
field
slug
(Deroceras
reticulatum)
(Muller))
by
stock
management
prior
to
direct-drilled
pasture
establishment.
Journal
of
Agricultural
Science
(in
press).
Hunter,
P.J.
1968.
Studies
on
slugs
of
arable
ground.
I.
Sampling
methods.
Malacologia,
6,
369-377.
McCullagh,
P.;
Nelder,
J.A.
1983.
Generalised
Linear
Models.
Chapman
and
Hall,
London,
pp.
261.
Schrim,
m.;
Byers,
R.A.
1980.
A
method
for
sampling
three
slug
species
attacking
sod-seeded
legumes.
Melsheimer
Entomology
Series
No.
29,
9-11.
South,
A.
1964.
Estimation
of
slug
populations.
Annals
of
Applied
Biology
53,
251-258.
Step
enson,
J.W.
1975.
Laboratory
observations
on
the
effect
of
soil
compaction
on
slug
damage
to
winter
wheat.
Plant
Pathology
24,
9-11.
Thomas,
D.C.
1944
Field
sampling
for
slugs.
Annals
of
Applied
Biology
31,
163-164.
336