9/
2004
Page
1
of
17
Response
Of
Propanil
Task
Force
II
To
The
Review
Of
The
U.
S.
Environmental
Protection
Agency
I.
INTRODUCTION
This
document
responds
to
the
review
by
the
U.
S.
Environmental
Protection
Agency
("
EPA"
or
"
Agency")
of
the
final
reports
from
the
"
Evaluation
of
the
Potential
Exposure
of
Workers
to
Propanil
During
Mixing/
Loading
and
Aerial
Application
to
Rice
Fields
Using
Simultaneous
Dermal
Dosimetry
and
Biological
Monitoring
Techniques"
and
the
"
Propanil
Exposures
and
Risk
Assessment
Based
on
Data
From
an
Aerial
Application
Study
in
Rice
with
Liquid
Formulations".

The
Propanil
Task
Force
II
(
the
"
Task
Force")
believes
that
the
passive
dosimetry
data
developed
in
the
study
reflects
the
typical
occupational
exposure
to
propanil.
Further,
the
Task
Force
believes
that
these
data
are
the
most
reliable
and
relevant
evidence
of
such
exposure
and
should
be
used
by
the
Agency
for
propanil
risk
assessment
purposes.

Regarding
the
limited
bio­
monitoring
information
collected
in
the
study,
the
Agency
reviewer
correctly
noted
that
"
the
Propanil
Task
Force
II
(
PTF
II)
did
not
attempt
to
use
the
bio­
monitoring
data
to
perform
a
quantitative
assessment
of
mixers/
loaders
and
pilots
exposures
to
propanil".
The
Task
Force
has
determined
that
quantitative
assessment
of
exposure
is
not
possible
based
on
the
bio­
monitoring
data
from
this
study
and
discusses
this
issue
below.

In
addition,
the
Agency
reviewer
noted
a
number
of
issues
related
to
the
conduct
of
the
study.
Explanation
and
clarification
of
each
of
those
issues
is
provided
in
this
document.

II.
A
QUANTITATIVE
RISK
ASSESSMENT
BASED
ON
THE
URINE
DATA
FROM
WORKERS
IS
NOT
RELIABLE
OR
APPROPRIATE
In
"
Evaluation
of
the
Potential
Exposure
of
Workers
to
Propanil
During
Mixing/
Loading
and
Aerial
Application
to
Rice
Fields
Using
Simultaneous
Dermal
Dosimetry
and
Biological
Monitoring
Techniques",
the
Task
Force
attempted
to
use
the
urine
monitoring
data
to
perform
a
semi­
quantitative
evaluation
of
propanil
exposure
recognizing
that
there
were
flaws
in
such
an
assessment.
In
the
"
Propanil
Exposures
and
Risk
Assessment
Based
on
Data
From
an
Aerial
Application
Study
in
Rice
with
Liquid
Formulations",
the
approach
for
the
semi­
quantitative
evaluation
of
exposure
was
reevaluated
and
it
was
concluded
that
any
type
of
quantitative
assessment
of
exposure
using
the
urine
screening
results
was
inappropriate.
The
basis
for
this
conclusion
was:

1.
The
semi­
quantitative
assessment
was
based
on
a
cursory
evaluation
of
propanil
metabolites
that
may
be
convertible
to
3,4­
dichloroaniline
(
3,4­
DCA)
by
acid
hydrolysis
of
rat
urine
from
a
metabolism
study.
In
the
absence
of
human
9/
2004
Page
2
of
17
pharmacokinetics
data,
it
is
inappropriate
to
use
rat
metabolism
data
combined
with
the
human
urine
monitoring
data
to
quantify
human
exposure
to
propanil.
2.
Most
of
the
workers
had
propanil
exposure
prior
to
or
shortly
after
the
study
application
day
which,
even
with
human
pharmacokinetic
data
to
predict
excretion
rates
and/
or
steady
state
of
non­
study
related
propanil
exposure,
makes
interpretation
of
the
data
difficult
at
best.

Each
of
these
items
is
discussed
below.

A.
Propanil/
DCA
Analysis
of
Urine
Samples
and
Propanil
Metabolism
Analysis
of
urine
samples
for
3,4­
DCA
from
mixer/
loaders
and
pilots
was
included
in
the
"
Evaluation
of
the
Potential
Exposure
of
Workers
to
Propanil
During
Mixing/
Loading
and
Aerial
Application
to
Rice
Fields
Using
Simultaneous
Dermal
Dosimetry
and
Biological
Monitoring
Techniques"
to
provide
a
screening
tool
for
propanil
exposure.
A
rat
metabolism
study
(
MRID
Nos.
41796401
and
41796402)
using
radiolabelled
propanil
showed
that
3,4­
DCA
is
a
minor
metabolite
of
propanil
representing
<
1%
of
the
total
radioactivity
in
the
urine.
From
a
toxicological
standpoint
however,
DCA
represents
a
significant
residue
of
concern
for
propanil
since
both
propanil
and
3,4­
DCA
are
known
inducers
of
methemoglobinemia,
the
most
sensitive
endpoint
of
concern
for
propanil.
On
this
basis,
the
Agency's
Health
Effects
Division
(
HED)
Metabolism
Assessment
Review
Committee
(
MARC)
concluded
that
the
propanil
residues
to
be
regulated
in
plants
and
animals
are
propanil
and
any
residues
convertible
to
3,4­
DCA.
There
was
no
need
for
individual
quantitation
of
propanil
metabolites
since
the
residue
analytical
method
in
plants
and
animals
determines
base­
releasable
total
DCA
expressed
as
propanil
equivalent
residues.

Following
a
single
oral
dose
of
propanil,
the
urine
results
from
rat
oral
metabolism
study
showed
that
there
are
two
distinct
metabolic
pathways:
­
Omega­
oxidation
of
the
propionate
side
chain
to
form
dicarboxylic
acids,
followed
by
conjugation
with
glucuronide.
­
Side
chain
cleavage
with
aromatic
ring
hydroxylation
at
the
6­
position
with
subsequent
conjugation
with
sulfate.

The
results
from
the
rat
metabolism
study
clearly
demonstrated
that
3,4­
DCA
was
a
minor,
transient
metabolite
of
propanil.
Recognizing
the
likelihood
that
only
low
levels
of
free
DCA
would
be
present
in
the
workers'
urine,
the
method
of
El
Marbouh
et
al.
(
2002)
was
used.
This
method
involved
acid
hydrolysis
of
urine
which
was
expected
to
hydrolize
DCA
metabolites
to
parent
DCA.
The
method
was
developed
to
provide
a
urine
screening
test
for
occupational
exposure
to
3,4­
DCA.
The
intent
of
the
method
was
not
to
quantify
exposure.

In
the
initial
Task
Force
Study
report,
the
semi­
qualitative
assessment
of
propanil
exposure
assumed
that
acid
hydrolysis
of
urine
would
convert
specific
propanil
metabolites,
including
products
of
detoxification,
to
DCA
and
that
the
urine
metabolite
9/
2004
Page
3
of
17
profiles
would
be
similar
between
rats
and
man.
A
cursory
evaluation
of
the
metabolites
seen
in
rat
urine
from
the
rat
metabolism
study
suggested
that
approximately
47%
of
the
metabolites
would
most
likely
convert
to
DCA
upon
acid
hydrolysis
and
this
value
was
inappropriately
applied
to
give
a
semi­
quantitative
measure
of
propanil
exposure.

The
supplemental
Task
Force
Study
report
recognized
significant
flaws
in
this
approach.
First,
the
absence
of
actual
data
regarding
the
conversion
of
propanil
metabolites
to
total
DCA
makes
application
of
any
percentage
factor
unreliable.
Second,
the
urine
results
from
the
rat
metabolism
study
were
based
on
total
excretion
over
a
72­
hour
period
and
provided
no
pharmacokinetic
evaluation
of
propanil
excretion
rate
and
half
lives.
These
factors
are
important
in
determining
anticipated
metabolite
levels
at
shorter
and
longer
time
periods
such
as
used
in
the
worker
exposure
study.
Third,
the
rat
metabolism
study
was
conducted
by
the
oral
route,
whereas,
worker
exposure
in
the
study
would
be
via
the
dermal
and
inhalation
routes.
Fourth,
the
rate
of
metabolism
for
many
xenobiotics
for
rodent
species
have
been
shown
to
be
different
from
other
non­
rodent
species
including
man.
Even
if
a
pharmacokinetics
study
in
rats
was
available,
the
applicability
of
the
half
life
and
excretion
rates
in
rats
to
humans
could
not
be
assumed
to
reflect
those
of
humans.
Thus,
in
the
absence
of
human
pharmacokinetics
data,
application
of
any
conversion
factors
based
on
a
rat
metabolism
study
is
inappropriate
and
speculative.

Thus,
available
data
are
not
sufficient
to
quantify
propanil
exposure
to
humans
based
on
urine
levels
of
3,4­
DCA.
Human
pharmacokinetics
data
are
necessary
to
provide
a
valid
approach
for
quantification.

B.
Non­
study
Exposures
to
Propanil
The
protocol
specified
that
urine
would
be
collected
from
each
study
subject
on
the
following
schedule:
­
Day
­
1,
starting
with
the
first
void
approximately
24­
hours
prior
to
propanil
application.
­
Day
0
(
application
day)
for
time
periods
0­
12
hours
and
12­
24
hours.
­
Day
1
and
Day
2
post
application.

An
important
component
of
urine
screening
for
3,4­
DCA
was
the
specification
that
study
participants
avoid
contact
with
any
product
containing
propanil
for
at
least
3
days
prior
to
and
3
days
following
the
study
application
day.
Unfortunately,
these
circumstances
could
not
be
controlled
in
this
worker
exposure
study.
The
worker
exposure
study
was
conducted
with
actual
propanil
applications
to
rice
fields
based
on
actual
propanil
orders
to
aerial
applicator
facilities
from
rice
growers.
During
the
spring/
summer
of
2003,
the
rice­
growing
regions
in
Louisiana,
Texas
and
Arkansas
experienced
a
great
deal
of
rain
and
windy
weather
that
resulted
in
the
delay
of
many
applications.
The
application
would
go
forward
only
when
rain
was
not
expected
and
specific
conditions
regarding
wind
speed
and
wind
direction
were
met.
If
these
conditions
occurred,
the
applicators
moved
quickly
with
the
propanil
application
during
the
particular
"
window
of
opportunity."
Because
of
the
weather
conditions,
it
was
not
possible
to
predict
when
a
9/
2004
Page
4
of
17
commercial
aerial
application
to
rice
would
take
place.
Since
the
study
was
conducted
during
the
peak
propanil
application
period,
it
was
anticipated
that
the
exposure
data
collection
would
occur
over
approximately
6
weeks.
Due
to
the
adverse
weather
conditions,
the
exposure
data
collection
period
was
approximately
12
weeks.
As
a
result
of
uncertainties
regarding
the
application
day,
a
number
of
pre­
study
urines
were
taken
on
subjects
only
to
have
the
applications
in
which
they
were
to
be
involved
delayed.
Further,
applicators
refused
to
not
handle
or
apply
propanil
prior
to
or
following
the
study
applications
since
their
livelihood
was
based
on
completing
outstanding
propanil
orders
within
the
weed
growth
periods
specified
by
the
label.
Thus,
a
majority
of
the
test
subjects
handled
or
applied
propanil
during
the
3
days
prior
to
and
3
days
following
the
study
application
day.

The
outcome
was
that
most
of
the
workers
handled
propanil
on
the
days
prior
to
the
study
application
day,
the
days
following
the
study
application
or
in
some
cases
both
time
periods.
These
exposures,
coupled
with
the
lack
of
human
pharmacokinetic
data,
make
it
impossible
to
validly
use
the
urine
DCA
concentrations
seen
in
the
worker
study
for
assessing
the
risk
of
propanil
to
workers.

III.
DESCRIPTION
FOR
PROPANIL
USE
PRACTICES­­
APPLICATION
TO
RICE
FIELDS
The
objective
of
the
Task
Force
study
was
to
measure
potential
exposure
to
mixer/
loaders
and
pilots
from
propanil
that
was
mixed/
loaded
and
applied
in
a
realistic
and
routine
fashion.
Achieving
this
objective
allows
exposure
assessments
to
be
performed
in
a
way
that
determines
an
accurate
risk
to
mixer/
loaders
and
pilots
handling
Propanil.

Information
about
Propanil
use
pattern
on
rice
fields
is
significant
to
a
reliable
understanding
of
potential
exposure
issues
to
mixer/
loaders
who
handle
the
concentrated
product
and
to
understanding
potential
exposure
to
pilots
who
apply
the
diluted
product.
The
following
is
a
description
of
the
use
pattern
of
propanil
on
rice
fields
and
information
that
should
be
considered
and
examined
when
considering
exposure
scenarios
to
both
mixer/
loaders
and
pilots
handling
propanil.

A.
Propanil
Use
Rates:

Propanil
used
for
weed
control
in
rice
is
sold
in
30­
35
gallon
drums
in
the
form
of
several
products
such
as
STAM*
M­
4,
Arrosolo
®
3­
3E,
WHAM!
EZ,
DUET
®
CA,
and
other
products.
Propanil
is
used
to
control
weeds
that
have
emerged
and
is
always
applied
to
drained
rice
fields.
For
this
reason
excessive
rain
can
delay
applications
since
draining
of
the
rice
fields
becomes
an
overriding
factor
in
scheduling
application
times.
Propanil
is
most
commonly
applied
at
3
to
4
lbs
ai/
Acre
when
grasses
are
actively
growing
in
the
3­
4­
leaf
stage
approximately
15­
25
days
after
planting.
Higher
rates
of
propanil
at
4­
6
lbs
ai/
Acre
are
sometimes
used
when
grasses
are
in
the
4­
6­
leaf
stage
about
20­
30
days
after
planting
or
for
emergency
or
"
rescue"
operations
for
older
tillering
grasses
about
30­
40
days
from
planting.
9/
2004
Page
5
of
17
B.
Methods
of
Handling
Propanil
During
Mixing/
Loading
of
the
Product:

The
loading
of
propanil
products
into
a
spray
tank
of
an
airplane
occurs
in
two
stages.
Propanil
is
removed
from
the
product
drum
using
a
siphoning
device
commonly
called
a
"
STAM"
pipe.
This
pipe
has
graduations
on
the
outside
and
an
adjustable
stop
around
the
pipe.
This
pipe
is
attached
to
a
hose
that
goes
through
a
pump
and
into
a
mix
tank.
The
mixer/
loader
adjusts
the
stop
on
the
"
STAM"
pipe
up
or
down
until
the
gradation
on
the
pipe
indicates
how
many
quarts
or
gallons
will
be
removed
from
the
drum
of
product.
The
pipe
is
then
placed
into
the
hole
in
the
lid
of
the
drum
where
the
stop
on
the
pipe
allows
the
withdrawal
of
the
desired
volume
from
the
drum.
The
predetermined
amount
of
formulation
is
removed
from
the
drum
and
pumped
into
a
mix
tank
where
water
is
added.
The
siphoning
device
is
left
in
the
drum
until
the
drum
is
emptied.
It
is
then
removed
from
the
drum
and
set
aside
until
needed
again.
This
is
similar
methodology
to
that
represented
by
the
data
in
PHED.
Propanil
is
mixed
with
other
products
and
water
in
the
mix
tank
and
the
mixture
is
transferred
to
the
airplane
spray
tank
via
a
dry­
lock
connector
that
attaches
to
the
airplane
tank
from
the
mix
tank.
The
entire
mixing/
loading
process
is
usually
performed
by
the
mixer/
loader
while
the
pilot
stays
in
the
airplane.
The
entire
mixing/
loading
process
takes
approximately
5­
10
minutes
between
applications.

C.
Factors
Affecting
the
Daily
Mixing/
loading
and
Application
Time
of
Propanil
Products:

Propanil
applications
are
made
only
when
the
grower
places
an
order
at
the
aerial
application
facility.
The
application
facility
responds
to
the
order
for
application
based
on
several
factors:

1.
If
weather
conditions
have
allowed
floodwater
to
be
drained
off
of
the
field,
the
application
will
proceed.

2.
If
the
wind
is
below
10
mph
the
aerial
applicator
will
proceed
with
the
application.

3.
If
the
wind
direction
is
such
that
potential
drift
will
not
be
in
the
direction
of
neighboring
homes
or
non­
target
crops,
the
application
will
proceed.

Propanil
orders
and
wind
factors
are
very
real
and
practical
constraints
placed
on
the
aerial
applicator.
In
Louisiana
for
example
the
state
monitors
complaints
from
neighbors
who
may
take
exception
to
an
application
of
propanil
in
which
there
was
some
drift
onto
neighboring
homes
or
crops.
The
state
will
levy
fines
of
up
to
several
thousand
dollars
to
aerial
applicators
when
complaints
from
neighbors
reach
a
designated
level.

Wind
factors
come
into
play
when
planning
for
propanil
applications
and
play
a
major
role
in
how
long
a
mixer/
loader
loads
the
chemical
and
how
long
the
pilot
applies
the
product.
The
mixer/
loader
and
pilot
generally
begin
applying
propanil
as
soon
as
the
sun
comes
up
to
avoid
wind
problems.
Morning
fog
may
delay
the
initiation
of
application
9/
2004
Page
6
of
17
and
may
delay
the
application
completely
if
the
wind
becomes
excessive
after
the
fog
clears.
The
pilot
will
apply
loads
of
propanil
until
the
wind
reaches
the
limit
of
10
mph
or
until
the
wind
shifts
to
an
unfavorable
direction.
When
that
happens,
the
pilot
flies
back
to
the
aerial
applicator
facility
where
the
spray
boom
will
be
switched
to
a
fertilizer
rig
and
the
pilot
will
apply
fertilizer
the
remaining
time
during
the
day
while
the
wind
stays
above
10
mph.
Consequently,
under
realistic
conditions,
it
is
not
unusual
for
pilots
to
only
apply
propanil
and
mixer/
loaders
will
only
load
propanil
for
an
average
of
about
2
hours
per
day.

A
pilot
can
treat
about
40
acres
per
load.
Each
load
will
take
about
15
minutes
of
flying
time.

Mixer/
loaders
load
Propanil
for
about
5­
10
minutes
for
each
load
and
will
continue
to
load
the
plane
as
it
comes
in
between
applications.
Mixer/
loaders
will
load
fertilizer
when
the
planes
rigs
are
switched
and
do
not
continue
to
handle
Propanil
during
the
remaining
part
of
the
day.

D.
Factors
Affecting
the
Conduct
of
the
2003
Propanil
Mixer/
Loader
and
Aerial
Applicator
Exposure
Study:

All
of
the
factors
outlined
above
came
into
play
during
the
2003
propanil
mixer/
loader
aerial
applicator
exposure
study.

1.
Early
Application
Season
Was
Chosen
for
the
Worker
Exposure
Study:

Early
season
application
of
Propanil
was
chosen
in
order
to
meet
the
protocol
requirement
of
finding
mixer/
loaders
and
pilots
who
did
not
have
previous
exposure
to
Propanil
in
2003.

The
season
for
planting
rice
crops
moves
from
south
Texas
to
Louisiana,
and
Arkansas
as
follows:

a.
Rice
planting
in
south
Texas
begins
around
mid
to
late
February.
Application
of
Propanil
begins
in
late
March
and
early
April.

b.
Rice
planting
in
Louisiana
begins
in
mid
March
to
early
April.
Applications
of
Propanil
begins
in
mid
April
to
early
May.

c.
Rice
planting
in
Arkansas
begins
in
mid
April
to
the
end
of
April.
Propanil
applications
begin
in
early
May
to
June.

The
above
sequence
of
Texas
to
Louisiana
to
Arkansas
was
followed
when
finding
test
volunteers
for
this
study
and
for
scheduling
applications
of
Propanil.
9/
2004
Page
7
of
17
2.
H.
E.
R.
A.
C.,
Inc.
Field
Scientists
Waited
for
Propanil
Orders
from
Growers
and
for
Wind
Conditions
to
be
Favorable
Before
Performing
each
Replicate
of
the
Study.

On
most
occasions
the
procedures
for
finding
test
subjects,
test
sites
as
well
as
scheduling
the
Propanil
treatments
were
as
follows:

a.
The
aerial
applicator
facility
was
contacted
and
a
meeting
was
set
up
with
the
owner
of
the
aerial
applicator
facility.
If
the
owner
of
the
aerial
application
facility
approved
the
use
of
his/
her
facility
for
performing
the
study,
the
owner
would
then
recommend
pilots
and
mixer/
loaders
for
the
study.

b.
A
meeting
was
held
with
the
pilots
and
mixer/
loaders
to
acquire
their
consent
to
be
involved
in
the
study.

c.
The
H.
E.
R.
A.
C.,
Inc.
field
scientist
would
then
wait
and
contact
the
owner
or
pilot
volunteer
to
determine
when
an
order
for
Propanil
came
in
and
when
the
product
was
likely
to
be
sprayed.
At
this
point
the
H.
E.
R.
A.
C,
Inc.
field
scientist
would
show
up
each
morning
at
about
5:
00
AM
at
the
aerial
applicator
facility
to
be
sure
that
all
the
test
equipment
was
in
place
if
the
wind
conditions
were
to
be
favorable
to
apply
Propanil.
This
procedure
is
time
consuming,
however,
it
is
the
only
way
to
be
certain
to
be
present
when
an
application
of
propanil
is
to
take
place.

On
some
occasions,
rain
delayed
the
scheduled
application
of
the
product.
In
northeastern
Arkansas
for
example,
excessive
rains
resulted
in
the
growers
being
about
2­
3
weeks
behind
in
their
planting.

IV.
EPA
COMMENTS:
The
protocol
states
that
"
mixer/
loaders
will
wear
new
or
freshly
laundered
long­
sleeved
shirts,
new
or
freshly
laundered
long
pants,
new
or
freshly
laundered
t­
shirt
and
brief,
new
chemical­
resistant
gloves,
new
shoes,
new
socks,
and
protective
eyewear
as
required
by
the
label."
However,
mixer/
loaders
in
the
study
wore
either
a
chemical
resistant
apron
or
a
Tyvek
coverall
over
the
"
outer"
dosimeter
(
cotton
coverall)
and
also
wore
chemical
resistant
footwear.
In
calculating
potential
dermal
exposures
to
mixer/
loaders,
the
study
author
does
not
factor
in
this
additional
personal
protective
equipment,
which
exceeds
the
requirements
of
the
product
labeling.

Triple
Layers
for
Mixer/
Loaders:
In
the
study
protocol,
where
the
use
of
a
protection
factor
was
presented
as
an
approach
to
calculating
exposure
to
the
torso,
arms,
and
legs,
the
study
authors
stated
that
mixer/
loaders
would
wear
new
or
freshly
laundered
long­
sleeve
shirt
and
long
pants
over
a
tee­
shirt
and
briefs.
9/
2004
Page
8
of
17
EPA
agreed
that
such
attire
would
permit
the
calculation
of
a
penetration
factor
for
estimating
risks
to
the
torso,
arms,
and
legs.
However,
in
the
actual
study,
all
mixer/
loaders
wore
either
a
chemical­
resistant
apron
or
a
tyvek
coverall
as
an
additional
layer
over
cotton
coveralls.
The
study
did
not
use
the
residues
on
the
apron
or
Tyvek
coveralls,
but
measured
residues
on
the
cotton
coverall
as
the
"
outer
dosimeter"
and
the
tee­
shirt
and
briefs
as
an
"
inner
dosimeter."
However,
the
apron
provided
a
chemical­
resistant
barrier
over
the
coverall
in
the
torso
section
and
presumably
reduced
penetration
to
the
coverall
itself
to
a
significant
degree.
Since
an
apron
would
not
provide
similar
exposure
to
the
legs
and
arms,
applying
a
penetration
factor
calculated
on
the
torso
exposures
to
the
residues
on
the
arms
and
legs
would
result
in
a
calculated
penetration
factor
that
would
be
artificially
low.
Similarly,
the
Tyvek
coverall
reduced
exposure
to
the
cotton
coverall
to
a
significant
degree.
Therefore,
any
calculated
penetration
factor
also
would
be
artificially
low.
HED
considered
calculating
a
penetration
factor
using
the
Tyvek
coverall
as
an
outer
dosimeter
and
the
cotton
coverall
as
an
inner
dosimeter.
However,
only
four
mixer/
loader
replicates
were
performed
with
Tyvek
coveralls,
which
would
not
provide
a
statistically
significant
number.
In
addition,
nonwoven
fabric,
such
as
Tyvek,
is
known
to
allow
significantly
less
penetration
than
a
cotton
coverall,
therefore
a
calculated
penetration
factor
would
not
be
representative
of
the
penetration
through
cotton
coveralls.
Due
to
many
practical,
considerations,
including
cost
and
heat
stress
concerns,
EPA
does
not
require
routine
use
by
mixer/
loaders
or
applicators
of
coveralls
made
from
Tyvek
or
other
nonwoven
fabrics.

Dermal
Exposures
to
Mixer/
Loaders
and
Pilots:
HED
used
the
propanil
passive
dosimetry
data
to
calculate
dermal
exposures
to
the
torso,
arms,
and
legs
for
the
pilot
scenarios
only.
HED
did
not
calculated
dermal
exposure
to
the
torso,
arms,
and
legs
for
the
mixer/
Loader
or
mixer/
loader/
applicator
scenarios,
since
an
appropriate
protection
factor
could
not
be
calculated
from
data
as
presented
in
the
study.
HED
calculated
the
dermal­
body
exposure
to
pilots
using
the
same
method
as
the
study
author.
First,
a
penetration
factor
was
derived
by
dividing
the
amount
of
residue
on
the
inner
dosimeter
(
tee
shirt
and
briefs)
by
the
residue
on
the
torso
section
of
the
outer
dosimeter.
Then
the
outer
dosimeter
residues
for
arms,
legs
and
torso
were
multiplied
by
the
penetration
factor.
For
the
pilot
study,
dermal
unit
exposures
averaged
1.05E­
04
mg/
lb
ai
for
the
arms,
4.82E­
05
mg/
lb
ai
for
the
legs,
and
7.46E­
05
mg/
lb
ai
for
the
torso.
HED
then
calculated
total
dermal
unit
exposure
for
pilots
by
summing
the
dermal
unit
exposures
to
hands
(
from
hand
washes),
to
face,
head,
and
neck
(
from
head
patches),
and
to
the
arms,
legs
and
torso
(
as
described
above).
Total
dermal
unit
exposure
estimates
averaged
1.27E­
03
mg/
lb
ai
handled
for
pilots.
The
total
dermal
unit
exposure
to
mixer/
loaders
and
to
mixer/
loaders/
applicator
were
not
calculated
by
HED,
since
an
appropriate
protection
factor
could
not
be
calculated
from
the
data
as
presented
in
the
study
for
dermal
exposure
to
the
torso,
arms,
and
legs
for
these
two
scenarios.
9/
2004
Page
9
of
17
A.
RESPONSE
TO
EPA
COMMENTS
ON
CALCULATION
OF
MIXER/
LOADER
PROPANIL
EXPOSURE.

1.
Clarification
of
Clothing
Scenario
Worn
by
Mixer/
Loaders:

EPA
refers
to
an
"
outer
coverall"
as
an
"
outer
dosimeter"
in
the
comments
shown
above.
"
However,
in
the
actual
study,
all
mixer/
loaders
wore
either
a
chemical­
resistant
apron
or
a
tyvek
coverall
as
an
additional
layer
over
cotton
coveralls.
The
study
did
not
use
the
residues
on
the
apron
or
Tyvek
coveralls,
but
measured
residues
on
the
cotton
coverall
as
the
"
outer
dosimeter"
and
the
tee­
shirt
and
briefs
as
an
"
inner
dosimeter."
However,
the
apron
provided
a
chemical­
resistant
barrier
over
the
coverall
in
the
torso
section
and
presumably
reduced
penetration
to
the
coverall
itself
to
a
significant
degree"
The
"
outer
dosimeter"
worn
by
mixer/
loaders
and
pilots
was
a
100%
cotton
long
sleeve
shirt
and
100%
cotton
long
pants.
These
garments
are
referred
to
on
pages
17,
19,
and
22
of
the
final
report:
"
Evaluation
of
the
Potential
Exposure
of
Workers
to
propanil
During
Mixing/
Loading
and
Aerial
Application
to
Rice
Fields
Using
Simultaneous
Dermal
Dosimetry
and
Biological
Monitoring
Techniques".

2.
Task
Force
Proposal
to
Recalculate
the
Exposure
of
Propanil
to
Mixer/
Loaders:

The
EPA
has
indicated
that
the
method
of
the
calculation
of
a
penetration
factor
for
the
arms
and
legs
of
the
mixer/
loaders
may
not
be
appropriate
using
the
inner
dosimeter
in
the
torso
area
of
the
mixer/
loader
since
wearing
a
chemical
resistant
apron
over
the
torso
area
could
lead
to
an
artificially
low
penetration
factor.
The
Task
Force
proposes
an
alternative
method
to
calculate
the
potential
dermal
exposure
to
the
mixer/
loaders
wearing
the
chemical­
resistant
apron.

The
proposed
method
to
recalculate
the
dermal
exposure
to
propanil
mixer/
loaders
uses
a
penetration
factor
of
10%.
Average
penetration
factors
ranging
from
2­
27%
percent
have
been
observed
in
studies
since
1981
(
References
1­
4).
Thus,
a
reasonable
average
penetration
factor
to
use
for
subsequent
calculations
would
be
10%
for
the
upper
and
lower
arms
as
well
as
the
back
of
the
legs
of
the
mixer/
loader.
This
10%
penetration
factor
is
a
conservative
estimate
of
penetration
since
the
front
top
of
the
thighs
down
to
the
ankle
of
the
mixer/
loader
were
covered
by
the
chemical
resistant
apron
worn
by
the
mixer/
loader
test
subjects
in
this
study.
This
covered
area
of
the
body
would
result
9/
2004
Page
10
of
17
in
much
less
than
10%
penetration
through
the
front
thigh
and
calf
areas
of
the
legs.

The
Task
Force's
proposed
method
of
calculation
of
exposure
to
the
mixer/
loaders
in
this
study
is
as
follows:

a.
The
residues
of
propanil
on
the
legs
and
arms
of
the
outer
dosimeter
long­
sleeved
cotton
shirt
and
long
cotton
pants
are
summed.
The
sum
is
then
multiplied
by
0.1
to
achieve
the
residue
amounts
given
a
10%
penetration
factor
for
the
cotton
garments.

b.
The
corrected
residue
on
the
arms
and
legs
of
the
mixer/
loader
are
then
added
to
the
residues
of
propanil
on
the
t­
shirt
and
brief
as
well
as
to
the
residues
in
the
handwash.
This
sum
is
then
added
to
the
head,
face
and
neck
residues
derived
from
the
head
patch
data
to
provide
a
total
dermal
body
exposure
for
the
worker.

The
following
table
presents
the
calculation
of
dermal
exposure
and
total
dose
for
the
mixer/
loaders
in
this
study
(
corrected
for
dermal
penetration
of
20%).
Only
workers
wearing
chemicalresistant
aprons
are
considered
when
using
this
alternative
method
of
calculating
exposure.
The
four
workers
wearing
the
Tyvek
coverall
were
excluded
from
the
data
set
since
the
Tyvek
coverall
would
produce
in
theory
an
artificially
low
residue
on
the
outer
dosimeter
shirt
and
pants,
rendering
the
arms
and
leg
data
from
the
shirt
and
pants
outer
dosimeter
useless.
9/
2004
Page
11
of
17
Worker
ID
V1
V3
V5
V7
V9
V13
V15
V19
V23
V27
Application.
Date
4/
12/
03
4/
3/
03
4/
14/
03
5/
6/
03
4/
12/
03
4/
17/
03
4/
17/
03
4/
29/
03
5/
9/
03
5/
19/
03
Propanil
Residues
(
µ
g)
Outer
Arms
402.9
2761
7749
37464
1027
1774
225.8
5696
2587
496.2
Outer
Legs
152.6
196.7
2552
2156
1014
452.8
33.57
3017
531.4
136.1
Sum
555.5
2957.7
10301
39620
2041
2226.8
259.4
8713
3118
632.3
10%
Pen.
Factor
55.6
295.8
1030
3962
204
223
25.9
871
312
63.2
Tee
Shirt+
Brief
5.655
18.4
26.3
17.7
36.57
7.334
2.319
30.84
88.88
13.79
Handwash
88.2
504
40.7
73.8
52.7
7.74
3.05
118
215
7.01
Head
Patch
130
31.9
1778
2920
2829
14.4
2.36
106
1879
269
T.
Dermal
279.5
850
2875
6974
3122
253
33.6
1126
2495
353
20%
Dermal
Pen.
55.9
170
575
1395
624
50.6
6.72
225
499
70.6
Inhalation
Exposure
7.7
0.6
12.9
14.5
10.7
0.715
1.43
2.87
5.35
8.12
T.
Dose
(
µ
g)
63.6
171
588
1410
635
51.3
8.15
228
504
78.7
Kg
BW
65.8
72.6
99.9
65.8
129.4
62.7
84
95.3
68.1
99.9
µ
g/
Kg
BW/
day
0.97
2.4
5.9
21.4
4.9
0.82
0.097
2.39
7.4
0.79
Kg
ai
Handled
602.7
489.8
394.5
292
602.7
231.3
231.3
133.3
88.9
760.1
µ
g/
Kg
ai/
day
0.11
0.35
1.5
4.8
1.1
0.22
0.035
1.7
5.7
0.1
T.
Dermal
279.5
850
2875
6974
3122
253
33.6
1126
2495
353
Kg
ai
Handled
602.7
489.8
394.5
292
602.7
231.3
231.3
133.3
88.9
760.1
Total
Dermal
(
ug/
Kg
ai
Handled)
0.46
1.74
7.29
23.9
5.18
1.09
0.145
8.45
28.1
0.46
Lb
ai
Handled
1329
1080
870
644
1329
510
510
294
196
1676
Total
Dermal
(
ug/
lb
ai
Handled)
0.21
0.79
3.30
10.83
2.35
0.50
0.066
3.83
12.7
0.21
T.
Inhalation
7.7
0.6
12.9
14.5
10.7
0.715
1.43
2.87
5.35
8.12
Kg
ai
Handled
602.7
489.8
394.5
292
602.7
231.3
231.3
133.3
88.9
760.1
Total
Inhalation
(
ug/
Kg
ai
Handled)
0.013
0.0012
0.033
0.050
0.018
0.003
0.006
0.022
0.060
0.011
Lb
ai
Handled
1329
1080
870
644
1329
510
510
294
196
1676
Total
Inhalation
(
ug/
lb
ai
Handled)
0.0058
0.00056
0.015
0.023
0.0081
0.0014
0.0028
0.0098
0.0273
0.0048
The
results
of
the
recalculated
dermal
exposure
for
the
mixer/
loaders
in
this
study
show
that
the
total
dose
(
dermal
exposure
corrected
for
dermal
penetration
of
20%
and
normalized
for
worker
body
weight
in
Kg)
ranged
from
0.097
to
21.4
µ
g/
Kg
BW/
day
(
Geometric
Mean
=
2.1
µ
g/
Kg
BW/
day;
std
dev.
=
6.4
µ
g/
Kg
BW/
day).

When
total
dermal
and
inhalation
exposures
are
normalized
against
pounds
handled
the
resultant
Geometric
Means
are:
1.30
ug/
lb
ai
for
dermal
and
0.0065
ug/
lb
ai
for
inhalation.
Exposure
and
risk
9/
2004
Page
12
of
17
for
Mixer/
loaders
was
recalculated
based
on
the
proposed
method
using
a
default
penetration
factor
rather
than
a
calculated
one
and
only
those
workers
who
did
not
wear
Tyvek.
The
results
are
presented
in
the
following
table.
MOEs
are
acceptable
for
mixer/
loaders
who
load
for
as
many
as
1200
acres
a
day.

Study
Data
All
Workers
Study
Data
Without
Tyvek
MOE
Rate
(
lb
ia/
A)
1200
Acres
350
Acres
1200
Acres
350
Acres
Mixer/
Loader
3
880
3000
660
2200
6
440
1500
330
1100
Applicator
3
3100
11000
3100
11000
6
1600
5600
1600
5600
V.
EPA
COMMENTS:
The
study
states
that
most
mixers/
loaders
used
a
siphoning
device
to
transfer
the
propanil
from
the
drum
into
the
mix
tank
and
then
used
a
dry­
lock
system
(
an
engineering
control)
to
pump
the
dilute
mixture
into
the
airplane
spray
tank.
This
would
result
in
artificially
low
mixer/
loader
exposures
due
to
the
use
of
the
engineering
control
in
part
of
the
mix/
load
process.

A.
RESPONSE
TO
EPA
COMMENTS:
USE
OF
ENGINEERING
CONTROLS
PRODUCE
ARTIFICALLY
LOW
EXPOSURES
TO
MIXER­
LOADERS:

1.
Objective
of
Study:

One
of
the
objectives
of
this
study
was
to
measure
potential
exposure
to
mixer/
loader
and
pilots
from
propanil
that
was
mixed/
loaded
and
applied
in
a
realistic
and
routine
fashion.

2.
Standard
Routine
Mixing/
Applying
Procedures
for
Mixer/
loaders
and
Aerial
Applicators
Handling
and
Applying
Propanil
Products:

Propanil
is
first
removed
from
the
drum
into
a
mix
tank
using
a
siphoning
device
commonly
called
a
"
STAM"
pipe
(
see
III
DESCRIPTION
FOR
PROPANIL
USE
PRACTICES
 
APPLICATION
TO
RICE
FIELDS
for
discussion).
This
system
is
similar
to
those
included
in
PHED
for
Scenario
6
(
ALL
LIQUIDS,
CLOSED
MIXING
and
LOADING
(
MLOD)).
The
siphon
device
runs
on
a
pump
that
transfers
the
propanil
from
the
drum
into
a
mix
tank
where
the
propanil
is
mixed
with
water.
The
propanil
9/
2004
Page
13
of
17
water
mixture
is
then
pumped
into
the
airplane
spray
tank
via
a
closed
system.
This
closed
system
consists
of
a
dry­
lock
coupling
from
the
mix
tank
to
the
spray
tank
of
the
airplane.
This
mixing/
loading
procedure
is
a
routine
operation
and
would
not
produce
artificially
low
exposures.

VI.
EPA
COMMENTS:
The
protocol
states
"
all
workers
will
perform
their
work
tasks
for
a
typical
amount
of
time
that
represents
an
entire
workday.
One
replicate
will
be
an
entire
workday
for
each
test
subject 
the
air
sampling
pump
will
operate
for
the
entire
monitoring
replicate
(
estimated
to
be
6­
12
hours)."
However,
the
average
duration
of
actual
handling
for
each
replicate
in
the
study
was
only
2.2
hours
for
mixer/
loaders
and
only
1.7
hours
for
pilots.

Reduced
Likelihood
of
Penetration
of
Coverall:
Penetration
of
a
chemical
through
a
matrix
is
dependent
on
three
factors:

1.
Composition
of
the
matrix
2.
Concentration
of
the
residue
on
the
matrix
surface;
and
3.
Time
of
residue
contact
with
the
matrix
surface.

This
study
used
application
rates
lower
than
the
maximum
6
pounds
active
ingredient
listed
in
the
protocol
and
on
the
product
labeling
(
STAM
M4),
with
and
average
application
rate
in
the
study
less
than
3
pounds
active
ingredient
per
acre.
In
addition,
this
study
involved
much
lower
handling
times
than
were
listed
in
the
protocol.
The
average
actual
handling
time
for
mixer/
loaders
was
only
2.2
hours
and
for
pilots
were
only
1.7
hours
and
the
average
dermal
monitoring
time
was
less
than
4.5
hours
for
both
handling
tasks.
The
application­
rate
factor
would
be
expected
to
result
in
less
residue
being
deposited
on
the
outer
dosimeter
and
the
handling
time
factor
would
be
expected
to
result
in
less
time
for
the
residue
to
penetrate
the
outer
dosimeter
1.
RESPONSE
TO
EPA
COMMENTS:

A.
Timing
and
Length
of
Aerial
Applications
of
Propanil:

Propanil
aerial
applications
generally
take
place
in
the
early
morning
or
may
be
applied
in
late
evening
when
the
wind
is
less
than
10
mph.
Due
to
the
possibility
of
drift
onto
neighboring
crops,
liability
issues
play
a
major
role
in
decisions
to
apply
propanil
even
though
the
wind
may
be
at
a
favorable
velocity
but
at
an
unfavorable
direction.
In
Louisiana
for
example,
local
authorities
track
the
number
of
complaints
that
neighbors
file
concerning
drift
on
to
their
property
from
aerial
applications
to
adjacent
farms.
The
aerial
applicator
is
fined
or
penalized
proportionally
to
the
number
of
complaints
that
are
logged
by
the
authorities.
9/
2004
Page
14
of
17
For
the
above
reasons,
propanil
typically
is
only
mixed/
loaded
and
applied
during
those
periods
of
time
when
the
wind
conditions
are
favorable.
The
mixing/
loading
times
for
workers
in
this
study
ranged
from
0.37
hours
to
5.3
hours
(
median
=
2.33
hours
and
average
=
2.24
hours).
There
were
only
two
workers
who
mixed/
loaded
for
less
than
one
hour.
These
two
replicates
were
cut
short
by
wind
factors
or
lack
of
propanil
orders
on
the
scheduled
day
of
the
test.
Pilot
application
times
ranged
from
0.68
to
3.82
hours
(
median
=
1.59
hours
and
average
=
1.67
hours).
Only
two
pilots
applied
less
than
one
hour.
Again,
these
two
replicates
were
abbreviated
due
to
wind
factors
or
lack
of
orders
for
propanil
on
the
scheduled
day
of
the
test.
The
average
work
days
for
aerial
applicators
observed
in
this
study
of
2.2
hours
for
mixer/
loaders
and
1.7
hours
for
pilots
is
realistic
for
the
routine
application
of
propanil
during
the
rice
season.
The
protocol
also
states
that
the
study
would
monitor
actual
applications
as
they
occur.
A
pilot
may
well
spend
6+
hours
treating
rice
during
a
normal
day
but
most
of
that
time
will
be
spent
applying
fertilizer
after
weather
conditions
no
longer
allow
propanil
applications.
The
workers
monitored
in
this
study
did
work
appropriate
times
and
handled
appropriate
amounts
of
material.
Recreating
Scenarios
6
and
7
from
the
"
PHED
SURROGATE
EXPOSURE
GUIDE"
(
August,
1998)
allows
one
to
look
at
the
amounts
of
material
handled
for
both
the
mixer/
loader
and
the
pilot
as
well
as
the
acres
treated
by
the
pilot.
The
following
Tables
present
the
statistics
from
the
two
scenarios
for
pounds
handled.

Comparison
of
Hours
worked
in
PHED
vs.
Hours
worked
in
Propanil
Study.

PHED
Study
All
M/
L
Workers
Study
w/
o
Tyvek
Hours
Hours
Hours
Hours
Hours
Mixer/
Loader
Applicator
Mixer/
Loader
Applicator
Mixer/
Loader
Geo
Mean
2.1
1.1
1.9
1.5
1.6
Average
3.3
1.9
2.2
1.7
2.0
Median
4.2
0.73
2.3
1.6
2.0
75th
Percentile
4.6
2.9
2.8
1.7
2.7
90th
Percentile
4.8
4.6
3.4
2.1
3.4
Minimum
0.18
0.21
0.37
0.72
0.37
Maximum
4.8
9.4
5.3
3.8
3.4
9/
2004
Page
15
of
17
Comparison
of
Pounds
Handled
in
PHED
vs.
Pounds
Handled
in
Propanil
Study.
PHED
Study
All
M/
L
Workers
Study
w/
o
Tyvek
Pounds
Pounds
Pounds
Pounds
Pounds
Mixer/
Loader
Applicator
Mixer/
Loader
Applicator
Mixer/
Loader
Geo
Mean
222
44
629
544
626
Average
520
244
710
583
734
Median
127
19.5
644
579
610
75th
Percentile
1022
248
837
757
1027
90th
Percentile
1510
760
1229
798
1329
Minimum
3
2
364
195
196
Maximum
1569
1569
1328
870
1329
Comparison
of
Acres
Treated
in
PHED
vs.
Acres
Treated
in
Propanil
Study.
PHED
Propanil
Study
Applicators
Acres
Treated
Acres
Treated
Applicator
Applicator
Geo
Mean
147
125
Average
286
151
Median
200
138
75th
Percentile
366
200
90th
Percentile
784
246
Minimum
10
32
Maximum
1061
334
As
can
be
seen
from
the
above
data,
the
Geometric
Mean,
Average,
Median
and
Minimum
Pounds
Handled
for
the
propanil
study
all
exceed
those
same
values
for
the
data
in
PHED.
This
is
true
for
the
original
interpretation
of
the
study
data
and
for
the
interpretation
that
excludes
those
workers
wearing
Tyvek.
Further,
the
hours
worked
is
consistent
with
PHED
and
likewise
supports
that
the
propanil
data
from
the
Study
are
superior
to
PHED
with
its
pounds
handled.
Therefore,
the
propanil
study
is
not
only
appropriate
based
on
the
agronomic
and
climatological
realities
of
propanil
applications
in
rice
but
is,
since
it
was
conducted
on
the
molecule
of
interest
and
is
more
robust
than
PHED,
more
realistic
and
appropriate
for
estimation
of
exposure
to
workers
involved
in
propanil
use
in
rice.
9/
2004
Page
16
of
17
B.
National
Agricultural
Aviation
Association`
s
Survey
2003:

Data
from
the
National
Agricultural
Aviation
Association's
survey
in
2004
show
that
the
average
flying
time
for
airplanes
in
aerial
application
scenarios
on
a
given
day
is
3­
4
hours.
In
the
present
study,
mixing
loading
times
ranged
from
0.37
to
5.3
hours
(
median
=
2.33
hours)
and
application
times
ranged
from
0.68
to
3.82
hours
(
median
=
1.59
hours).
These
mixer/
loader
and
pilot
replicate
times
are
not
extremely
different
from
the
data
gathered
by
the
National
Agricultural
Aviation
Association
in
a
survey
performed
in
2004
with
aerial
applicators
throughout
the
US
(
Reference
5).
It
is
concluded
that
the
times
that
workers
handled
propanil
during
this
study
were
realistic
and
routine.

VII.
CONCLUSIONS
The
above
discussion
clearly
shows
that
the
worker
exposure
study
with
propanil
involving
passive
dosimetry
was
conducted
according
to
typical
current
use
practices
followed
in
the
rice
industry.
The
amounts
of
product
mixed
and
applied
and
the
exposure
times
for
mixer/
loaders
and
pilots
were
typical
of
propanil
uses
and
more
representative
of
propanil
uses
than
the
data
currently
in
PHED
for
the
liquid
mixer/
loader
and
aerial
applicator
scenarios.
There
were
no
special
considerations
made
regarding
either
the
clothing
or
PPE
worn
by
mixer/
loaders
or
pilots
and
the
procedures
for
mixing
and
transferal
of
propanil
to
and
from
mixing
tanks
and
to
airplanes
were
consistent
with
industry
standards.
Because
of
the
consistency
in
rice
industry
procedures
and
practices,
the
dosimetry
data
are
considered
valid
measures
of
worker
exposure.

Urine
samples
from
mixer/
loaders
and
pilots
were
collected
and
concentrations
of
3,4­
DCA
were
determined.
These
measurements
are
adequate
as
a
screening
assay
for
propanil
exposure
but,
in
the
absence
of
human
pharmacokinetics
data,
they
are
not
appropriate
for
the
quantification
of
propanil
exposure
in
humans.

VIII.
REFERENCES:

1.
El
Marbouh,
L.,
Areliano,
C.,
Philibert,
C.,
Evrard,
Poey,
J
and
Houin,
G.
(
2002).
Development
and
evaluation
of
an
HPLC
urinalysis
screening
test
for
occupational
exposure
to
3,4­
and
3,5­
dichloroanilines.
International
J.
of
Clin.
Pharmacol
and
Therapeutics,
40(
1):
41­
46.

2.
Orlando,
J.,
Branson
G.
Ayers,
and
Levitt,
P.,
"
The
Penetration
of
Formulated
Guthion
Spray
Through
Selected
Fabrics",
J.
Environ.
Sci.
Health,
B16
(
5):
617­
628,
(
1981).
9/
2004
Page
17
of
17
3.
Mumma,
R.,
Brandes,
G.
A.,
Brandes
G.
F.,
"
Exposure
of
Applicators
and
Mixer­
Loaders
During
Application
of
Mancozeb
by
Airplanes,
Air
Blast
Sprayers,
and
Compressed
Air
Backpack
Sprayers",
ACS
Symposium
Series
273,
Edited
by
R.
Honeycutt,
G.
Zweig,
and
N.
Ragsdale,
American
Chemical
Society,
pp
201­
220(
1985).

4.
Keeble,
V.,
Dupont,
P.,
Doucette,
W.,
Norton,
M.
"
Guthion
Penetration
of
ClothingMaterilas
During
Mixing
and
Spraying
in
Orchards",
Performance
of
Protective
Clothing,
Edited
by
S.
Mansdorf,
R.
Sager,
and
A.
Nielsen,
American
Society
for
Testing
Materials,
pp
573­
583
(
1988).

5.
Poppendorf.
W.,
Mechanisms
of
Clothing
Exposure
and
Dermal
Dosing
During
Spray
Application",
Performance
of
Protective
Clothing,
Edited
by
S.
Mansdorf,
R.
Sager,
and
A.
Nielsen,
American
Society
for
Testing
Materials,
pp
611­
624
(
1988).

6.
"
Pesticide
Use
Survey
Report
for
Agricultural
Aviation"
National
Agricultural
Aviation
Association,
Washington,
DC
(
May
2004).

WDC99
963577­
3.058275.0010
