August
8,
2001
Mr.
Anthony
Britten
Special
Review
and
Reregistration
Division
Office
of
Pesticide
Programs
(7508C)
U.
S.
Environmental
Protection
Agency
Room
266A,
Crystal
Mall
2
1921
Jefferson
Davis
Highway
Arlington,
Virginia
22202
Corr.
#
daL075­
01
Dear
Mr.
Britten,

Re:
Carbaryl;
Chemical
number
56801;
EPA
Reg.
No.
264­
324
Review
of
the
Preliminary
Risk
Assessments
for
Human
Health
and
for
Environmental
Fate
and
Ecological
Effects
for
the
Reregistration
Eligibility
Decision
on
Carbaryl
We
have
completed
our
review
of
the
preliminary
human
health
and
environmental
fate
and
ecological
effects
risk
assessments
for
the
Reregistration
Eligibility
Decision
of
carbaryl.
Our
comments
on
the
Agency's
assessments
are
enclosed
in
the
following
documents:
 
Review
of
the
Draft
Human
Health
Risk
Assessment
 
Review
of
the
Draft
Environmental
Fate
and
Ecological
Risk
Assessment
for
the
Reregistration
of
Carbaryl
Hard
copies
of
the
above
documents
will
be
sent
by
express
mail
within
the
next
week.

Key
issues
discussed
in
these
documents
are
as
follow:

DRAFT
HUMAN
HEALTH
RISK
ASSESSMENT
FQPA
Considerations
A
2­
Generation
reproduction
study
was
recently
completed
and
submitted
to
the
Agency.
The
No
Observable
Effect
Levels
(NOELs)
for
carbaryl
administered
in
the
diet,
in
CD®
(SD)
rats
under
the
conditions
of
this
study,
were:

°
For
parental
systemic
toxicity:
75
ppm
°
For
parental
reproductive
toxicity:
1500
ppm
°
For
offspring
toxicity:
75
ppm
In
addition,
Aventis
has
submitted
to
the
EPA
a
revised
Developmental
Neurotoxicity
Study
(DNT)
which
demonstrated
that
no
alterations
in
brain
morphometric
measurements
were
observed
in
the
offspring,
thus
no
increased
sensitivity
was
observed
in
the
pups.

Aventis
firmly
believes
that
with
the
submission
of
these
two
studies,
the
data
gap
for
the
2­
generation
reproduction
study
is
fulfilled
and
the
issue
with
the
developmental
neurotoxicity
study
is
resolved.
Thus,
the
Agency
has
the
required
data
to
justify
removing
the
extra
10X
FQPA
Safety
Factor
from
the
risk
calculations
throughout
the
Human
Health
Risk
Assessment
for
carbaryl.
Mr.
Anthony
Britten
August
8,
2001
Page
2
Additional
Uncertainty
Factors
For
the
chronic
study
in
dogs,
Aventis
does
not
agree
with
the
additional
3X
for
a
lack
of
a
NOEL
in
the
study.
According
to
the
policy
issued
in
August
2000
concerning
the
endpoint
for
selection
for
cholinesterase
inhibiting
compounds,
the
Agency
stated
that
the
red
blood
cell
(RBC)
inhibition
should
be
used
instead
of
the
plasma.
The
NOEL
for
RBC
cholinesterase
inhibition
was
125
ppm
(3.1
mg/
kg/
day).
Additionally,
Aventis
believes
that
the
brain
cholinesterase
inhibition
observed
at
the
125
ppm
level
was
slightly
above
background
level
and
was
not
of
toxicological
concerns
due
to
the
lack
of
clinical
signs
observed
at
this
dose
level.
Furthermore,
Aventis
performed
a
five­
week
study
in
dogs,
which
agreed
with
the
results
of
the
chronic
study.
Therefore,
it
is
our
opinion
that
all
calculations
based
on
the
chronic
dog
in
the
document
should
be
adjusted
to
reflect
the
removal
of
the
3X­
safety
factor.

Cancellation
of
Certain
Uses
of
Carbaryl
Use
on
Barley,
Oats,
Rye,
and
Cotton:
Changes
should
be
made
throughout
the
Human
Health
Risk
Assessment
and
associated
supporting
documents
to
reflect
the
cancellation
of
the
use
on
barley,
oats,
rye,
and
cotton.
It
should
be
noted
that
Aventis
CropScience
labels
for
the
technical
materials
and
the
end­
use
products
containing
carbaryl
were
amended
to
delete
these
uses.
The
Agency
has
already
approved
the
labeling
changes.

Use
on
Poultry:
Aventis
CropScience
will
no
longer
support
the
use
of
carbaryl
for
direct
application
to
poultry,
as
well
as
the
poultry
quarters
treatment.
We
will
submit
a
request
for
cancellation
of
these
uses
in
accordance
with
section
6(
f)(
1)
of
the
Federal
Insecticide,
Fungicide
and
Rodenticide
Act
(FIFRA).
The
cancellation
of
this
use
causes
a
significant
improvement
in
the
risk
assessment
since
poultry
are
the
major
contributor
to
the
risk
estimate.
We
believe
that
only
the
dietary
analyses
conducted
without
poultry
should
be
included
in
the
HED
Chapter.
Analyses
including
poultry
would
be
unrealistic
and
misleading.

Dietary
Exposure
Assessment
An
industry­
sponsored
Carbamate
Market
Basket
Survey
(CMBS)
was
conducted
and
submitted
to
the
Agency
for
use
in
the
dietary
exposure
and
risk
assessment
for
carbaryl.
The
Carbamate
Market
Basket
Survey
Task
Force
met
with
EPA/
HED
before
the
start
of
the
study
to
review
the
purpose
and
design
of
the
study.
The
EPA
assured
the
task
force
that
the
study
data
would
be
used
in
risk
assessments,
taking
precedence
over
all
other
monitoring
or
field
trial
data
available
for
those
crops.
A
surrogation
plan
according
to
HED
SOP
99.3
was
also
assured.
These
data
provide
more
realistic
measures
of
residues
to
which
consumers
are
exposed
and
they
should
be
included
in
the
dietary
analyses
as
agreed
upon
by
the
CMBS
Task
Force
and
the
EPA.

Water
Exposure
Assessment
EPA
has
based
its
assumptions
about
concentrations
of
carbaryl
in
drinking
water
upon
model
simulations.
Aventis
CropScience
conducted
a
drinking
water
monitoring
program
that
provides
the
best
estimate
of
concentrations
of
carbaryl
in
drinking
water.
The
study
uses
the
sampling
design
for
acute
endpoints
recommended
in
industry/
EPA
meetings
during
1999
(weekly
sampling
during
times
of
peak
concentrations
over
a
three­
year
period).
Twenty
sites,
representative
of
the
most
vulnerable
community
water
systems
in
the
highest
use
areas
of
carbaryl,
were
selected.
These
included
16
sites
in
agricultural
areas
and
4
locations
in
urban
areas.
Samples
were
collected
from
the
inlet
and
outlet
water
at
each
sampling
interval.
Outlet
samples
were
only
analyzed
when
residues
were
present
in
the
inlet
samples.
The
analytical
method
has
a
limit
of
quantification
of
0.030
ppb
and
a
limit
of
detection
of
0.002
ppb.

The
study
provides
information
on
concentrations
of
carbaryl
in
community
water
systems
most
likely
to
contain
the
highest
concentrations
of
carbaryl.
Residues
of
carbaryl
in
other
areas
would
be
expected
to
be
Mr.
Anthony
Britten
August
8,
2001
Page
3
lower.
The
study
should
be
used
in
the
carbaryl
exposure
calculations
as
it
provides
the
drinking
water
concentrations
needed
for
FQPA
dietary
assessments.

Occupational
and
Residential
Exposure
Assessment
Aventis
CropScience
has
conducted,
or
is
in
the
process
of
conducting,
studies
relevant
to
the
refinement
of
the
occupational
and
residential
risk
assessment.
These
studies
are
as
follows:

Study
Anticipated
Completion
Date

4­
Week
Repeated­
Dose
Dermal
Toxicity
Study
with
SEVIN
®
XLR
Plus
in
Rats.
September
2001

4­
Week
Repeated­
Dose
Dermal
Toxicity
Study
with
SEVIN
®
80S
in
Rats.
September
2001

4­
Week
Repeated­
Dose
Dermal
Toxicity
Study
with
Carbaryl
Technical
in
Rats.
September
2001

Measurement
of
Pesticide
Exposure
of
Suburban
Residents
Associated
with
the
Residential
Use
of
Carbaryl.
March
2002

Biological
monitoring
of
citrus
orchard
application
and
post­
application
apple
thinning
and
cherry
harvesting
job
functions
March
2002
DRAFT
ENVIRONMENTAL
FATE
AND
ECOLOGICAL
RISK
ASSESSMENT
Carbaryl
Studies
to
Fulfill
Requirements
Aventis
CropScience
is
in
the
process
of
conducting,
or
has
scheduled,
studies
relevant
to
the
refinement
of
the
environmental
risk
assessments
for
carbaryl
and
the
major
degradate
1­
naphthol.
These
studies
are
as
follows:

 
Rate
and
Route
of
Aerobic
Degradation
in
Soils.
These
studies
have
been
initiated
with
parent
carbaryl
applied
to
four
diverse
U.
S.
soils.
The
data
are
intended
to
provide
additional
half­
life
determinations
for
parent
carbaryl
and
the
major
degradate
1­
naphthol.
Expected
completion
date:
March
2002
 
Aerobic
Aquatic
Metabolism
in
Two
Water/
Sediment
Systems.
These
studies
have
been
initiated
with
parent
carbaryl
applied
to
two
distinct
U.
S.
water/
sediment
systems.
The
data
are
intended
to
provide
additional
half­
life
determinations
for
parent
carbaryl
and
the
major
degradate
1­
naphthol.
In
addition,
further
identification
of
additional
degradation
products
is
anticipated.
Expected
completion
date:
March
2002
Mr.
Anthony
Britten
August
8,
2001
Page
4
 
Adsorption
and
Desorption
of
1­
Napthol
to
five
soils.
This
study
has
been
scheduled
to
evaluate
the
adsorption
and
desorption
of
the
major
carbaryl
degradate
to
five
soils/
sediment.
The
data
are
intended
to
provide
information
necessary
to
evaluate
the
environmental
risks
from
1­
naphthol
in
standard
models.
Expected
completion
date:
March
2002
For
the
reregistration
process
in
the
EU,
Aventis
CropScience
is
in
the
process
of
conducting,
or
has
scheduled,
studies
relevant
to
the
refinement
of
the
ecotoxicological
risk
assessments
for
carbaryl
and
the
major
degradate
1­
naphthol.
These
studies
are
as
follows:

 
Studies
with
Carbaryl:
Acute
oral
LD50
in
mallard
ducks
Dynamic
acute
LC50
in
bluegill
sunfish
Acute
toxicity
in
Daphnia
Acute
toxicity
in
Chironomus
riparius
Toxicity
in
Selenastrum
capricornutum
Acute
oral
and
contact
toxicity
in
honeybees
14­
d
toxicity
in
earthworms
Effects
on
soil
microorganisms
(nitrification/
carbon
cycle)
Effect
on
sewage
treatment
 
Studies
with
1­
naphthol
Early
life­
stage
study
in
fathead
minnows
Acute
toxicity
in
Daphnia
Acute
toxicity
in
Daphnia
in
presence
of
sediment
Chronic
toxicity
in
Daphnia
14­
d
toxicity
in
earthworms
 
Formulated
Product
Vegetative
Vigor
Toxicity
in
Selenastrum
capricornutum
Acute
oral
and
contact
toxicity
in
honeybees
Effect
on
non­
target
arthropods
14­
d
toxicity
in
earthworms
Effects
on
soil
microorganisms
(nitrification/
carbon
cycle)

Ecotoxicological
Risk
Assessments
Aventis
has
pointed
out
several
errors
in
the
PRZM
input
parameters
(see
comments
made
to
Tables
5
and
6
of
the
draft
RED).
Overly
conservative
estimates
of
foliar
dissipation
half­
lives
and
changes
in
ecotoxicology
study
endpoints
would
dictate
a
re­
calculation
of
the
EECs
and
risk
quotients
is
warranted
in
a
number
of
instances.
Mr.
Anthony
Britten
August
8,
2001
Page
5
Endocrine
Disruption
Reports
in
the
open
literature
on
the
reproductive
effects
of
carbaryl
in
wild
mammals
are
at
best
ambivalent.
The
recently
submitted
2­
generation
study
in
rats
demonstrates
the
absence
of
reproductive
effects.
As
EPA
pointed
out,
findings
reported
in
the
literature
were
made
at
concentrations
well
above
the
highest
peak
concentration
modeled.
Therefore
these
findings
are
irrelevant
for
a
risk
assessment
and
at
the
current
stage
of
discussion
about
endocrine
disruption.
If
the
concern
about
the
endocrine
potential
of
carbaryl
persists,
the
issue
should
be
revisited
once
the
Agency's
endocrine
disrupter
screening
and
testing
program
as
well
as
a
policy
on
how
to
incorporate
positive
findings
into
an
ecological
risk
assessment
have
been
fully
developed.

Mobility
The
classification
of
carbaryl
as
mobile
to
very
mobile
is
inconsistent
with
measured
Koc
values
of
177
to
249
(MRID
43259301).
According
to
the
widely
used
classification
scheme
of
McCall,
et
al.
(1980)
wherein
Koc
values
between
150
and
500
denote
medium
mobility
in
soil,
carbaryl
would
be
classified
as
having
medium
mobility
in
most
soils.
This
classification
of
medium
mobility
is
further
supported
by
the
acceptable
column
leaching
study
(MRID
43320701)
in
which
aged
carbaryl
residues
were
only
slightly
mobile
in
a
number
of
soils.
The
mobility
of
carbaryl
would
be
expected
to
be
higher
in
sandy
soils
or
in
soils
of
low
organic
matter.

1­
Napthol
Fate
and
Transport
The
Agency
is
requiring
additional
information
on
the
persistence
and
mobility
of
1­
naphthol,
a
major
environmental
degradate
of
carbaryl.
However,
a
half­
life
for
1­
naphthol
of
less
than
1
day
can
be
calculated
from
the
carbaryl
aerobic
soil
metabolism
study
(MRID
42785101).
The
data
from
this
study
demonstrate
that
under
aerobic
soil
conditions
the
formation
and
decline
of
1­
naphthol,
starting
from
parent
carbaryl,
is
complete
in
less
than
14
days.
This
half­
life
can
be
used
for
preliminary
environmental
fate
modeling
to
estimate
EECs
for
1­
naphthol.

The
EPA
suggested
that
1­
naphthol
is
not
strongly
sorbed
to
soil.
Additional
information
available
in
the
literature
demonstrates
that
the
sorption
of
1­
naphthol
to
soil
is
stronger
than
that
seen
for
carbaryl
itself.
Hassett
et
al.
(1981)
has
demonstrated
that
the
sorption
of
1­
naphthol
was
the
result
of
sorption
to
organic
carbon
resulting
in
Koc
values
between
431
and
15,618.
These
data
indicate
that
1­
naphthol
is
less
mobile
and
less
susceptible
to
leaching
than
carbaryl
itself,
and
they
demonstrate
that
at
least
a
portion
of
the
1­
naphthol
residue
is
tightly
sorbed
to
soil
constituents.
To
meet
the
requirement
for
information
on
the
adsorption
and
desorption
of
1­
naphthol
by
the
Agency,
the
registrant
is
conducting
an
adsorption/
desorption
study
to
meet
the
163­
1
guideline.
Study
results
should
be
available
for
submission
to
the
Agency
in
the
first
quarter
of
the
calendar
year
2002.

Ground
Water
EPA
summarized
information
on
the
detection
of
carbaryl
in
groundwater
from
the
EPA
Pesticides
in
Groundwater
Database,
the
EPA
STORET
database
and
the
NAWQA
database.
Each
of
the
databases
shows
a
pattern
of
very
low
levels
of
carbaryl
detection
in
few
groundwater
resources.
These
analyses
confirm
several
statements
made
by
the
Agency
that
carbaryl
have
limited
potential
to
impact
groundwater
resources.
However,
on
page
2
of
the
Memorandum
issued
June
28,
2001,
in
conjunction
with
the
EFED
RED
chapter
for
carbaryl,
EPA
is
requiring
additional
information
on
"Surface
and
groundwater
monitoring
in
urban
and
suburban
use
areas
(non­
guideline)."
Based
on
the
characteristics
of
carbaryl
and
the
available
data
demonstrating
limited
impact
of
carbaryl
on
ground
water
resources,
additional
studies
to
evaluate
the
potential
for
carbaryl
to
contaminate
groundwater
are
unnecessary
and
unwarranted.
Mr.
Anthony
Britten
August
8,
2001
Page
6
Please
let
me
know
if
you
need
any
additional
information.
My
phone
number
is
(919)
549­
2718.

Sincerely,

Danielle
A.
Larochelle
Danielle
A.
Larochelle
Registration
Manager
1
CARBARYL
Chemical
ID
No.
056801;
Case
0080
Review
of
the
Draft
Human
Health
Risk
Assessment
August
8,
2001
Aventis
CropScience
P.
O.
Box
12014,
2
T.
W.
Alexander
Drive
Research
Triangle
Park,
NC
27709
2
TABLE
OF
CONTENTS
GENERAL
COMMENTS
........................................................................................................
3
FQPA
Considerations
.........................................................................
3
Additional
UncertaintyFactors
...........................................................
3
Cancellation
of
Certain
Uses
of
Carbaryl
...........................................
3
Dietary
Exposure
Assessment
............................................................
4
Water
Exposure
Assessment
..............................................................
4
Occupational
and
Residential
Exposure
Assessment
.........................
4
PART
I
­
TOXICOLOGY
DATA
BASE
...................................................................................
6
Line­
by­
Line
Review
of
the
Toxicology
Data
Base
Evaluation
from
the
Human
Health
Risk
Assessment
Document
for
Carbaryl
(June
19,
2001)
........................................................................
6
PART
I
­
TOXICOLOGY
DATA
BASE
..................................................................................
20
SUPPORTING
DISCUSSIONS
.......................................................
20
PART
II
­
DIETARY
AND
WATER
EXPOSURE/
RISK
ASSESSMENT
......................................
32
Line­
by­
Line
Review
of
the
Dietary
and
Water
Exposure
Assessment
of
the
Human
Health
Risk
Assessment
Document
for
Carbaryl
(June
19,
2001)
......................................................................
32
Line­
by­
Line
Review
of
the
Supporting
Document
"Revised
Dietary
Exposure
Analysis
for
the
HED
Revised
Human
Health
Risk
Assessment
(Felicia
A.
Fort;
April
26,
2001)"
......................
40
PART
II
­
DIETARY
AND
WATER
EXPOSURE/
RISK
ASSESSMENT
......................................
44
SUPPORTING
DISCUSSIONS
.......................................................
44
PART
III
­
RESIDUE
CHEMISTRY
CONSIDERATIONS
..........................................................
51
Line­
by­
Line
Review
of
the
Residue
Chemistry
Information
Included
in
the
Human
Health
Risk
Assessment
Document
for
Carbaryl
(June
19,
2001)
......................................................................
51
Line­
by­
Line
Review
of
the
Supporting
Document
"Product
and
Residue
Chemistry
Chapters
for
the
Reregistration
Eligibility
Decision
(Felecia
Fort;
November
14,
2000)"
......................
53
Line­
by­
Line
Review
of
the
Supporting
Document
"Carbaryl:
Occupational
and
Residential
Exposure
Assessment
and
Recommendations
for
the
Reregistration
Eligibility
Decision
Document
(Jeffrey
L.
aDawson;
June
28,
2001)"
..................
59
PART
IV
­
OCCUPATIONAL
AND
RESIDENTIAL
EXPOSURE/
RISK
ASSESSMENT
...............
64
SUPPORTING
DISCUSSION
.........................................................
64
APPENDIX
I
...................................................................................................................
66
3
General
Comments
FQPA
Considerations
A
2­
Generation
reproduction
study
was
recently
completed
and
submitted
to
the
Agency.
The
No
Observable
Effect
Levels
(NOELs)
for
carbaryl
administered
in
the
diet,
in
CD®
(SD)
rats
under
the
conditions
of
this
study,
were:
°
For
parental
systemic
toxicity:
75
ppm
°
For
parental
reproductive
toxicity:
1500
ppm
°
For
offspring
toxicity:
75
ppm
In
addition,
Aventis
has
submitted
to
the
EPA
a
revised
Developmental
Neurotoxicity
Study
(DNT)
which
demonstrated
that
no
alterations
in
brain
morphometric
measurements
were
observed
in
the
offspring,
thus
no
increased
sensitivity
was
observed
in
the
pups.

Aventis
firmly
believes
that
with
the
submission
of
these
two
studies,
the
data
gap
for
the
2­
generation
reproduction
study
is
fulfilled
and
the
issue
with
the
developmental
neurotoxicity
study
is
resolved.
Thus,
the
Agency
has
the
required
data
to
justify
removing
the
extra
10X
FQPA
Safety
Factor
from
the
risk
calculations
throughout
the
Human
Health
Risk
Assessment
for
carbaryl.

Additional
UncertaintyFactors
For
the
chronic
study
in
dogs,
Aventis
does
not
agree
with
the
additional
3X
for
a
lack
of
a
NOEL
in
the
study.
According
to
the
policy
issued
in
August
2000
concerning
the
endpoint
for
selection
for
cholinesterase
inhibiting
compound,
the
Agency
stated
that
the
red
blood
cell
(RBC)
should
be
used
instead
of
the
plasma.
The
NOEL
for
RBC
cholinesterase
inhibition
was
125
ppm
(3.1
mg/
kg/
day).
Additionally,
Aventis
believes
that
the
brain
cholinesterase
inhibition
observed
at
the
125
ppm
level
was
slightly
above
background
level
and
was
not
of
toxicological
concerns
due
to
the
lack
of
clinical
signs
observed
at
this
dose
level.
Furthermore,
Aventis
performed
a
five­
week
study
in
dogs,
which
agreed
with
the
results
of
the
chronic
study.
Therefore,
it
is
our
opinion
that
all
calculations
based
on
the
chronic
dog
in
the
document
should
be
adjusted
to
reflect
the
removal
of
the
3X­
safety
factor.

Cancellation
of
Certain
Uses
of
Carbaryl
Use
on
Barley,
Oats,
Rye,
and
Cotton
Changes
should
be
made
throughout
the
Human
Health
Risk
Assessment
and
associated
supporting
documents
to
reflect
the
cancellation
of
the
use
on
barley,
oats,
rye,
and
cotton.
It
should
be
noted
that
Aventis
CropScience
labels
for
the
technical
materials
and
the
end­
use
products
containing
carbaryl
were
amended
to
delete
these
uses.
The
Agency
has
already
approved
the
labeling
changes.
Please
refer
to
Part
III
of
this
document
(Residue
Chemistry
Considerations)
for
current
labeling
information.
Corrections
on
label
acceptance
dates
are
also
provided
in
Part
III.
4
Use
on
Poultry
Aventis
CropScience
will
no
longer
support
the
use
of
carbaryl
for
direct
application
to
poultry,
as
well
as
the
poultry
quarters
treatment.
We
will
submit
a
request
for
cancellation
of
these
uses
in
accordance
with
section
6(
f)(
1)
of
the
Federal
Insecticide,
Fungicide
and
Rodenticide
Act
(FIFRA).
The
cancellation
of
this
use
causes
a
significant
improvement
in
the
risk
assessment
since
poultry
are
the
major
contributor
to
the
risk
estimate.
We
believe
that
only
the
dietary
analyses
conducted
without
poultry
should
be
included
in
the
HED
Chapter.
Analyses
including
poultry
would
be
unrealistic
and
misleading.

Dietary
Exposure
Assessment
An
industry­
sponsored
Carbamate
Market
Basket
Survey
(CMBS)
was
conducted
and
submitted
to
the
Agency
for
use
in
the
dietary
exposure
and
risk
assessment
for
carbaryl.
The
Carbamate
Market
Basket
Survey
Task
Force
met
with
EPA/
HED
before
the
start
of
the
study
to
review
the
purpose
and
design
of
the
study.
The
EPA
assured
the
task
force
that
the
study
data
would
be
used
in
risk
assessments,
taking
precedence
over
all
other
monitoring
or
field
trial
data
available
for
those
crops.
A
surrogation
plan
according
to
HED
SOP
99.3
was
also
assured.
These
data
provide
more
realistic
measures
of
residues
to
which
consumers
are
exposed
and
they
should
be
included
in
the
dietary
analyses
as
agreed
upon
by
the
CMBS
Task
Force
and
the
EPA.

Water
Exposure
Assessment
EPA
has
based
its
assumptions
about
concentrations
of
carbaryl
in
drinking
water
upon
model
simulations.
Aventis
CropScience
conducted
a
drinking
water
monitoring
program
that
provides
the
best
estimate
of
concentrations
of
carbaryl
in
drinking
water.
The
study
uses
the
sampling
design
for
acute
endpoints
recommended
in
industry/
EPA
meetings
during
1999
(weekly
sampling
during
times
of
peak
concentrations
over
a
three­
year
period).
Twenty
sites,
representative
of
the
most
vulnerable
community
water
systems
in
the
highest
use
areas
of
carbaryl,
were
selected.
These
included
16
sites
in
agricultural
areas
and
4
locations
in
urban
areas.
Samples
were
collected
from
the
inlet
and
outlet
water
at
each
sampling
interval.
Outlet
samples
were
only
analyzed
when
residues
were
present
in
the
inlet
samples.
The
analytical
method
has
a
limit
of
quantification
of
0.030
ppb
and
a
limit
of
detection
of
0.002
ppb.

The
study
provides
information
on
concentrations
of
carbaryl
in
community
water
systems
most
likely
to
contain
the
highest
concentrations
of
carbaryl.
Residues
of
carbaryl
in
other
areas
would
be
expected
to
be
lower.
The
study
should
be
used
in
the
carbaryl
exposure
calculations
as
it
provides
the
drinking
water
concentrations
needed
for
FQPA
dietary
assessments.

Occupational
and
Residential
Exposure
Assessment
Aventis
CropScience
has
conducted,
or
is
in
the
process
of
conducting,
studies
relevant
to
the
refinement
of
the
occupational
and
residential
risk
assessment.
These
studies
are
as
follows:
5
Study
Anticipated
Completion
Date

4­
Week
Repeated­
Dose
Dermal
Toxicity
Study
with
Sevin
 
XLR
Plus
in
Rats.
September
2001

4­
Week
Repeated­
Dose
Dermal
Toxicity
Study
with
Sevin
 
80S
in
Rats.
September
2001

4­
Week
Repeated­
Dose
Dermal
Toxicity
Study
with
Carbaryl
Technical
in
Rats.
September
2001

Measurement
of
Pesticide
Exposure
of
Suburban
Residents
Associated
with
the
Residential
Use
of
Carbaryl.
March
2002

Biological
monitoring
of
citrus
orchard
application
and
postapplication
apple
thinning
and
cherry
harvesting
job
functions
March
2002
6
Part
I
­
Toxicology
Data
Base
Line­
by­
Line
Review
of
the
Toxicology
Data
Base
Evaluation
from
the
Human
Health
Risk
Assessment
Document
for
Carbaryl
(June
19,
2001)

1.0
Executive
Summary
Section
Number
1.0
Page:
3
Paragraph:
4
Line:
3
EPA
comment:
In
the
developmental
neurotoxicity
study
in
the
rat,
alterations
in
brain
morphometric
measurements
were
observed
in
offspring.

Aventis'
response:
Aventis
has
submitted
to
EPA
a
revised
DNT,
which
demonstrated
no
alterations
in
brain
morphometric
measurements
taken
in
the
offspring,
thus
no
increased
sensitivity
was
observed
in
the
pups.
The
details
of
this
study
will
be
discussed
below
in
the
supporting
`Discussion
Section'.

Page:
3
Paragraph:
5
Line:
3
EPA
comment:
No
acceptable
two­
generation
reproduction
study
is
available.

Aventis'
response:
Aventis
has
submitted
to
EPA
a
new
2­
Generation
Reproduction
Study
with
carbaryl,
which
neither
demonstrated
that
carbaryl
is
a
reproductive
toxicant
nor
demonstrated
that
the
pups
were
more
sensitive
than
the
dams.
The
NOELs
for
carbaryl
administered
in
the
diet,
in
CD®
(SD)
rats
under
the
conditions
of
this
study
were:
°
For
parental
systemic
toxicity:
75
ppm
°
For
parental
reproductive
toxicity:
1500
ppm
°
For
offspring
toxicity:
75
ppm
The
details
of
this
study
will
be
discussed
below
in
the
supporting
`Discussion
Section'.

Page:
3
continued
on
page
4
Paragraph:
5
Lines:
4
­7
EPA
comment:
In
the
developmental
neurotoxicity
study,
maternal
toxicity
was
observed
at
the
same
dose
as
changes
in
brain
morphometric
measurements
in
offspring;
however,
these
brain
measurements
were
not
performed
at
the
next
lower
dose
and
consequently,
there
is
some
uncertainty
about
the
NOAEL
for
this
effect.
7
Aventis'
response:
As
stated
above,
Aventis
has
submitted
to
EPA
a
revised
DNT,
which
demonstrated
no
alterations
in
brain
morphometric
measurements
taken
in
the
offspring,
thus
no
increased
sensitivity
was
observed
in
the
pups.
The
details
of
this
study
will
be
discussed
below
in
the
supporting
`Discussion
Section'.

Page:
4
Paragraph:
5
­
continuation
from
page
3
Lines:
2
­9
EPA
comment:
The
10x
Food
Quality
Protection
Act
(FQPA)
Safety
Factor
has
been
retained
because:
1)
the
toxicology
data
base
is
incomplete;
there
is
a
data
gap
for
the
multi­
generation
reproduction
study
in
rats;
2)
an
assessment
of
susceptibility
following
pre­/
post­
natal
exposure
to
carbaryl
could
not
be
made
due
to
the
data
gap
for
the
reproduction
study;
3)
there
is
concern
for
the
results
of
the
developmental
neurotoxicity
study
(uncertainty
about
NOAEL).
Based
on
these
considerations,
the
10x
safety
factor
was
applied
to
all
population
subgroups
in
assessing
risks
from
acute
and
chronic
dietary
exposures
and
residential
exposures
of
all
durations.

Aventis'
response:
As
stated
above,
Aventis
has
submitted
to
EPA
a
revised
DNT
and
a
new
2­
Generation
Reproduction
Study
that
has
a
direct
impact
on
the
FQPA
Safety
Factor.
Aventis
firmly
believes
that
with
the
submission
of
these
two
studies,
the
data
gap
for
the
2­
generation
reproduction
study
will
be
fulfilled
and
the
NOEL
question
for
the
DNT
will
be
resolved.
Thus,
the
Agency
has
the
required
data
to
reduce
the
FQPA
Safety
Factor
from
10X
to
1X.

Page:
4
Paragraph:
1
Lines
1­
6:
EPA
comment:
Carbaryl
has
been
classified
as
a
Group
C
possible
human
carcinogen
based
on
an
increased
incidence
of
hemangiosarcomas
and
combined
hemangiomas/
hemangiosarcomas
in
CD­
1
mice
at
1000
ppm
(146
mg/
kg/
day)
and
above.
Mechanistic
metabolism
studies
and
a
study
in
heterozygous
p53­
deficient
mice
were
considered
inadequate
to
demonstrate
a
mode
of
action
for
the
vascular
tumors.
The
default
linear
low
dose
extrapolation
was
used
for
risk
assessment;
the
Q1*
is
1.19
x
10
­2
(mg/
kg/
day)
­1
based
on
the
mouse
vascular
tumors.

Aventis'
response:
Aventis
strongly
believes
that
the
use
of
carbaryl
products
presents
no
imminent
carcinogenic
risk
to
users
based
on
the
following
points:

 
Current
data
support
a
hypothesis
that
high
doses
of
carbaryl
in
life­
time
studies
produce
tumors
via
a
non­
genotoxic
mechanism,
possibly
related
to
altered
metabolism
at
these
high
doses.
8
 
At
doses
less
than
the
MTD,
there
was
an
increased
incidence
of
tumors
only
in
1
site
of
1
sex
of
1
species.

 
The
increased
incidence
of
vascular
tumors
in
the
male
mice
after
two
years
of
administration
is
of
questionable
biological
significance.

 
In
the
p53
knockout
mouse
model,
which
was
demonstrated,
to
be
sensitive
to
the
induction
of
vascular
tumors
by
a
genotoxic
reference
compounds,
carbaryl
was
found
to
be
negative.

 
The
weight­
of­
the
evidence
indicates
that
carbaryl
shows
no
potential
for
genotoxicity
 
Epidemiological
data
on
carbaryl
production
workers
show
no
increase
in
tumor
incidence.

In
the
two­
year
bioassay
conducted
in
the
CD1
mouse,
a
statistically
significantly
higher
incidence
of
vascular
tumors
was
noted
in
males
at
1,000
and
8,000
ppm.
In
females,
no
statistically
significant
change
in
the
incidence
of
vascular
tumors
was
observed.
The
overall
incidence
of
vascular
tumors
was
2,
6,
10
and
10
in
males
and
3,
3,
4
and
9
in
females
at
0,
100,
1,000
and
8,000
ppm,
respectively.

In
order
to
address
the
biological
significance
of
the
vascular
tumors,
the
p53
knockout
mouse
model
was
used
as
a
tool
to
study
chemical
carcinogenesis
(Donehower,
1996).
The
p53
mouse
is
a
genetically
manipulated
mouse
in
which
one
allele
of
the
p53
tumor
suppresser
gene
has
been
inactivated.
One
of
the
assumptions
with
the
knockout
model
is
that
a
mutation
at
the
intact
p53
allele
is
necessary
for
the
development
of
the
carcinogenic
process.
In
principle
nongenotoxic
compounds,
which
induce
tumors
by
other
mechanisms
should
not
induce
tumors
in
this
system.
The
p53
knockout
mouse
model
was
validated
by
testing
two
compounds:
urethane,
a
genotoxic
compound
known
to
produce
vascular
tumors
in
standard
carcinogenicity
bioassays,
and
d­
limonene,
which
is
neither
genotoxic
nor
carcinogenic
in
mice
but
which
is
known
to
be
carcinogenic
in
the
male
rat
by
a
well
described
non­
genotoxic
mechanism.
In
the
validation
study,
the
p53
model
proved
to
be
very
efficient
for
the
induction
of
vascular
tumors
by
genotoxic
compounds
(Bigot,
1999;
Carmichael
et
al.
1999).
Therefore,
this
model
will
produce
an
unambiguous
response
to
specifically
identify
genotoxic
compounds.

Therefore,
carbaryl
was
tested
in
the
p53
mouse.
The
objectives
of
the
studies
were
to
evaluate
if
carbaryl
would
induce
vascular
tumors
in
this
model
and
to
set
a
NOEL
in
this
sensitive
and
specific
model.
Additionally,
this
model
provides
evidence
that
the
tumors
observed
in
the
standard
carcinogenicity
bioassays
resulted
from
an
indirect
genotoxic
effect.
9
Carbaryl
was
administered
continuously
via
the
diet
to
groups
of
20
male
heterozygous
p53
knockout
mice
at
concentrations
of
0,
10,
30
100,
300,
1,000
and
4,000
ppm
for
at
least
180
days.
At
the
end
of
the
study,
all
animals
were
necropsied,
selected
organs
weighed
and
a
range
of
tissues
were
taken,
fixed
and
examined
microscopically.
Carbaryl
did
not
induce
mortalities
or
clinical
signs
related
to
the
treatment.
Only
a
slight
decrease
in
food
consumption
during
the
first
eight
weeks
was
observed
at
4,000
ppm.
This
observation
was
correlated
with
a
lower
body
weight
evolution
in
comparison
with
the
control
animals.

No
tumors
were
found
in
the
4,000
ppm
group.
In
the
other
treated
groups,
a
few
sporadic
tumors
were
found,
but
they
were
clearly
unrelated
to
treatment
and
representative
of
the
spontaneous
tumor
types
present
in
mice
of
this
age
and
strain.
In
particular,
it
should
be
noted
that
no
tumors
were
found
in
the
liver,
kidney
or
vascular
system,
which
were
seen
in
the
original
mouse
oncogenicity
study
with
carbaryl.
The
only
treatment­
related
non­
proliferative
change
observed
was
the
presence
of
globular
deposits
in
the
umbrella
cell
layer
of
the
urinary
bladder
at
100
ppm
or
more.

In
conclusion,
the
data
from
the
p53
studies,
the
following
conclusion
can
be
made:

1).
The
p53
model
was
validated
with
urethane
one
of
the
few
compound
known
to
induce
specifically
vascular
tumors.
2).
Carbaryl
was
negative,
and
also
did
not
accelerate
the
formation
of
any
other
tumor
types.
3).
There
is
a
clear
NOEL
at
4000
ppm
(approximately
716
mg/
kg
b.
w./
day)
in
the
p53
mouse
study
for
carcinogenicity.
4).
Carbaryl
is
not
a
genotoxic
carcinogen.

Thus
the
weight­
of­
the
evidence
indicates
that
carbaryl
is
not
a
genotoxic
agent
in
humans.
In
addition,
epidemiological
data
on
carbaryl
production
workers
show
no
increase
in
tumor
incidence.
The
two
epidemiology
studies
of
Aventis
CropScience
(formerly
Rhône­
Poulenc
and
Union
Carbide)
factory
workers
representing
a
sub­
population
of
the
general
public
with
the
highest
exposure
(i.
e.,
exposure
on
a
daily
basis
over
many
years).
Thus,
the
data
from
these
studies
represent
a
potential
"worst­
case"
for
chronic
carbaryl
exposure
to
humans.
The
results
from
these
studies
show
that
the
overall
mortality
experience
of
the
cohort
is
significantly
less
than
expected
when
compared
to
both
United
States
and
West
Virginia
male
death
rates.
In
general,
these
studies
support
conclusion
that
there
were
no
excess
cases
of
cancer
in
this
working
population.

A
further
discussion
of
this
study
and
the
entire
carcinogenicity
issue
will
be
discussed
below
in
the
supporting
`Discussion
Section'.
10
Page:
4
Paragraph:
2
Lines
5­
12:
EPA
comment:
The
toxicity
endpoints
selected
for
risk
assessment
are
based
on
neurotoxic
effects
of
ChEI.
The
dose
levels
used
for
the
acute
and
chronic
dietary
risk
assessment,
i.
e.
10
mg/
kg
and
3.1
mg/
kg/
day,
were
both
from
studies
(acute
neurotoxicity
study
in
the
rat
and
chronic
toxicity
study
in
the
dog,
respectively)
in
which
a
NOAEL
was
not
determined.
Therefore,
an
additional
uncertainty
factor
of
3x
was
added
to
the
customary
100x
uncertainty
factor
[10x
for
extrapolation
from
animal
studies
to
humans
and
10x
for
intraspecies
(human)
variation].
The
acute
and
chronic
reference
doses
were
0.03
mg/
kg
and
0.01
mg/
kg/
day,
respectively.

Aventis'
response:
Concerning
the
acute
neurotoxicity
study
in
rats,
Aventis
does
not
disagree
with
the
Agency's
assessment
for
this
endpoint.
However,
for
the
chronic
study
in
dogs,
Aventis
does
not
agree
with
the
additional
3X
for
a
lack
of
a
NOEL
in
the
study.
According
to
the
policy
issued
in
August
2000
concerning
the
endpoint
for
selection
for
cholinesterase
inhibiting
compound,
the
Agency
stated
that
the
red
blood
cell
(RBC)
should
be
used
instead
of
the
plasma.
The
NOEL
for
RBC
cholinesterase
inhibition
was
125
ppm
(3.1
mg/
kg/
day).
Additionally,
Aventis
believes
that
the
brain
cholinesterase
inhibition
observed
at
the
125
ppm
level
are
slightly
above
background
level
and
were
not
of
toxicological
concerns
due
to
the
lack
of
clinical
signs
observed
at
this
dose
level.
Furthermore,
Aventis
performed
a
five­
week
study
in
dogs,
which
agreed
with
the
results
of
the
chronic
study.
Therefore,
it
is
our
opinion
that
all
calculations
based
on
the
chronic
dog
in
the
document
should
be
adjusted
to
reflect
the
removal
of
the
3X
safety
factor.

Page:
4
Paragraph:
2
Lines
1­
6:
EPA
comment:
No
dermal
or
inhalation
studies
were
available.

Aventis'
response:
As
part
of
the
60­
day
public
comment
period,
Aventis
will
be
submitting
three
21­
day
dermal
studies,
which
cover
the
technical
material
and
the
formulated
products
SEVIN®
XLR,
and
SEVIN®
80S.
The
study
results
will
not
be
available
in
time
to
provide
the
information
to
the
Agency
as
part
of
the
30­
day
comment
period.
Similarly,
Aventis
will
be
submitting
comments
for
the
inhalation
requirement
as
part
of
the
60­
day
public
comment
period.
However,
it
should
be
noted
that
in
the
document
titled,
"Carbaryl:
Toxicology
Chapter
For
RED"
the
inhalation
study
was
not
listed
as
a
data
gap
on
page
25
of
that
document.
11
Page:
8
Paragraph:
3
Lines
1­
10:
EPA
comment:
HED
calculated
the
risks
of
carbaryl
exposure
in
tobacco.
Based
on
a
pyrolysis
study
submitted
by
the
registrant,
residues
of
carbaryl
total
approximately
44.58
ppm
in
tobacco
smoke
(side­
stream
and
main­
stream
combined).
Since
this
is
a
composited
sample
of
main­
stream
and
side­
stream
smoke,
it
greatly
exaggerates
the
actual
exposure
to
the
smoker,
whose
primary
route
of
exposure
is
via
mainstream
smoke.
HED
further
assumed
that
100%
inhaled
is
absorbed
(i.
e.,
that
none
of
the
residue
is
exhaled
along
with
the
smoke).
These
assumptions
result
in
an
extreme
overestimate
of
actual
likely
exposure.
Comparing
exposure
to
the
short­
term
inhalation
Lowest­
Observable­
Adverse­
Effects­
Level
(LOAEL)
of
10
mg/
kg/
day,
the
short­
term
MOE
for
carbaryl
exposure
from
the
use
of
tobacco
is
estimated
to
be
1000
for
males
and
900
for
females.
The
MOEs
are
less
than
the
residential
short­
term
inhalation
target
MOE
of
3000
and
therefore,
the
risks
exceed
HED's
level
of
concern
for
the
general
population.

Aventis'
response:
The
relative
risk
of
carbaryl,
which
is
not
a
genotoxic
agent,
as
compared
to
the
many
direct
acting
genotoxic
agents
contained
in
main­
and
side­
stream
smoke
is
insignificant
and
is
not
calculable.
The
calculation
presented
by
the
Agency
is
crude
at
best
and
does
not
have
any
impact
on
the
relative
risk
of
smoking
cigarettes.

Page:
10
Paragraph:
2
Lines
1­
18:
EPA
comment:
HED
evaluated
reports
of
human
carbaryl
poisonings
and
adverse
reactions
associated
with
its
use
from
the
following
sources:
OPP
Incident
Data
System
(IDS);
Poison
Control
Centers'
Toxic
Exposure
Surveillance
System;
California
Department
of
Pesticide
Regulation;
the
National
Pesticide
Telecommunications
Network
(NPTN);
open
literature;
and
an
unpublished
study
submitted
by
the
registrant.
The
data
from
IDS
indicated
that
a
majority
of
cases
from
carbaryl
exposure
involved
dermal
reactions.
A
number
of
other
cases
involved
asthmatics
and
people
who
experienced
hives
and
other
allergic
type
reactions.
It
is
noted
that
the
dermal
sensitization
study
in
the
guinea
pig
was
negative.
Reports
of
allergic­
type
reactions
in
humans
could
be
evidence
of
a
difference
in
species
sensitivity
or
could
be
attributable
to
inert
ingredients
in
the
marketed
formulations.
According
to
California
data,
about
half
have
the
cases
involved
skin
and
eye
effects
in
handlers.
About
a
quarter
of
the
skin
reactions
were
due
to
workers
that
were
exposed
to
residues
on
crops.
Reports
from
the
literature
are
very
limited
but
tend
to
support
the
finding
that
carbaryl
has
irritant
properties.
The
Poison
Control
Center
cases
involving
non­
occupational
adults
and
older
children
showed
an
increased
risk
in
five
of
the
six
measures
used
for
comparing
carbaryl
incidents
to
all
other
pesticides.
The
carbaryl
cases
were
almost
twice
as
likely
to
require
serious
health
care
(hospitalization
or
treatment
in
a
critical
care
unit)
and
were
two
and
a
half
times
more
likely
to
experience
major
medical
12
outcome
(life­
threatening
effects
or
significant
residual
disability)
than
other
pesticides.
This
pattern
of
increased
risk
was
not
seen
among
occupational
reports
or
in
young
children.
This
may
mean
that
careless
handling
by
nonprofessionals
is
a
particular
hazard.

Aventis'
response:
As
stated
by
the
Agency,
reports
from
the
literature
are
very
limited.
To
draw
such
conclusions
that
carbaryl
cases
were
almost
twice
as
likely
to
require
serious
health
care
(hospitalization
or
treatment
in
a
critical
care
unit)
and
were
two
and
a
half
times
more
likely
to
experience
major
medical
outcome
(life­
threatening
effects
or
significant
residual
disability)
than
other
pesticides
is
speculative
and
goes
too
far
in
assigning
increased
frequency
values
based
on
limited
and
often
unconfirmed
information.
The
reports
cited
by
the
Agency
from
poison
control
centers
are
difficult
to
interpret
because
the
relative
causative
agent
is
frequently
not
identified.
If
the
agent
is
identified
by
officials,
the
majority
of
the
time
it
is
stated
by
the
officials
that
the
causative
agent
is
merely
speculation
on
their
part
and
that
the
real
agent
is
not
readily
known
with
great
certainty.

According
to
test
conducted
by
Aventis
on
both
technical
grade
and
formulation
products,
carbaryl
is
not
a
skin
sensitizer.
Furthermore,
epidemiological
data
on
carbaryl
production
workers
showed
no
pattern
of
skin
reaction.
This
would
indicate
that
if
the
product
is
used
according
to
the
label,
the
relative
risks
of
skin
reactions
to
carbaryl
are
very
low.

Page:
10
Paragraph:
3
Lines
1­
5:
EPA
comment:
Five
case
report
studies
suggested
that
carbaryl
might
be
a
cause
of
chronic
neurological
or
psychological
problems.
Some
of
these
effects
appear
to
be
consistent
with
those
reported
from
organophosphate
poisoning.
However,
unlike
organophosphates,
no
controlled
studies
have
been
undertaken.
If
such
effects
occur
as
a
result
of
over­
exposure
to
carbaryl,
they
appear
to
be
relatively
rare.
The
effects
reported
among
the
five
case
reports
are
too
inconsistent
to
draw
any
conclusions,
but
do
suggest
the
need
for
further
study.

Aventis'
response:
Aventis
agrees
with
the
Agency
that
the
five
case
studies
cited
are
weak
at
best
and
that
the
Agency
should
not
be
drawing
any
conclusions
based
on
such
inconsistent
data.
The
discussion
of
such
data
will
lead
the
reader
to
draw
inappropriate
conclusions.
Furthermore,
epidemiological
data
on
carbaryl
production
workers
showed
no
such
pattern
and
would
indicate
that
if
the
product
is
used
according
to
the
label,
the
relative
risks
of
carbaryl
are
very
low.

Page:
11
Paragraph:
1
Lines
1­
3:
EPA
comment:
The
epidemiologic
study
submitted
by
the
registrant
compared
mortality
rates
in
plant
workers
exposed
to
carbaryl
to
the
general
population.
HED
concluded
that
13
the
sample
of
workers
was
too
small
and
the
period
of
follow­
up
too
short
to
permit
definitive
conclusions.

Aventis'
response:
Aventis
disagrees
with
the
Agency
that
the
mortality
rates
in
plant
workers
exposed
to
carbaryl
cannot
be
translated
to
the
general
population.
The
study
was
statistically
designed
to
specifically
address
this
concern
and
to
examine
other
non­
mortality
endpoints.
The
two
epidemiology
studies
of
Aventis
CropScience
(formerly
Rhône­
Poulenc)
factory
workers
representing
a
sub­
population
of
the
general
public
with
the
highest
exposure
(i.
e.,
exposure
on
a
daily
basis
over
many
years).
Thus,
the
data
from
these
studies
represent
a
potential
"worst­
case"
for
chronic
carbaryl
exposure
to
humans.
The
results
from
these
studies
show
that
the
overall
mortality
experience
of
the
cohort
is
significantly
less
than
expected
when
compared
to
both
United
States
and
West
Virginia
male
death
rates.
In
general,
these
studies
support
conclusion
that
there
were
no
excess
cases
of
cancer
in
this
working
population.

Page:
11
Paragraph:
4
Lines
2­
4:
EPA
comment:
However,
certain
key
information,
which
would
help
refine
the
risk
assessment,
is
missing.
Toxicology
data
gaps
include
a
reproduction
study,
a
21­
day
dermal
study
in
the
rat,
a
90­
day
inhalation
study
in
the
rat
and
a
micronucleus
study.

Aventis'
response:
The
Agency
has
previously
granted
a
wavier
for
the
in
vivo
cytogenicity/
micronucleus.
Furthermore,
Aventis
has
fulfilled
the
data
requirement
for
the
in
vivo
mouse
micronucleus
study.
Aventis
believes
that
repeating
the
in
vivo
mouse
micronucleus
study
is
not
appropriate
based
on
the
fact
that
the
top
dose
in
the
currently
submitted
mouse
micronucleus
was
200
mg/
kg/
day
and
was
similar
to
the
mid­
dose
of
250
mg/
kg/
day
tested
in
an
acute
gavage
study
in
mice.
At
the
250
mg/
kg
dose
level
tested
in
the
acute
study
plasma,
RBC,
and
brain
cholinesterase
inhibition
was
seen.
The
percent
cholinesterase
inhibition
observed
was
45.7,
45.7,
and
57.8,
respectively.
At
the
highest
dose
tested
(500
mg/
kg)
in
the
acute
study,
percent
cholinesterase
inhibition
for
plasma,
RBC,
and
brain
was
59.8,
57.1,
and
66.6,
respectively.
Additionally,
the
clinical
signs
observed
at
both
the
250
and
500
mg/
kg
dose
levels
were
very
similar
in
both
findings
and
incidences.
Repeating
the
mouse
micronucleus
study
at
a
slightly
higher
dose
level
would
not
result
in
an
increase
in
significant
clinical
signs.
Therefore,
the
base
results
of
the
study
concerning
clastogenic
or
aneugenic
effects
would
not
change.
Carbaryl
is
not
mutagenic
for
these
endpoints.

As
previously
indicated
in
this
document,
Aventis
will
be
submitting
three
21­
day
dermal
studies
which
cover
the
technical
material
and
for
the
formulated
products
SEVIN
XLR
and
SEVIN
80S.
Similarly,
Aventis
will
be
submitting
comments
for
the
inhalation
requirement
as
part
of
the
60­
day
public
comment
period.
14
However,
it
should
be
noted
that
in
the
document
titled,
"Carbaryl:
Toxiccology
Chapter
For
RED"
the
inhalation
study
was
not
listed
as
a
data
gap
on
page
25
of
that
document.

Section
Number
3.0
Hazard
Profile
Page:
14­
15
Paragraph:
1
Lines:
1­
10
EPA
comment:
The
Toxicology
Chapter
of
the
RED
was
prepared
by
Dr.
Virginia
Dobozy
(D240992
dated
December
13,
1999).
The
toxicology
database
is
of
good
quality;
however,
it
is
incomplete.
The
following
studies
are
required:
two­
generation
reproduction
study,
21­
day
dermal
toxicity
study
with
cholinesterase
measurements,
90­
day
inhalation
study
with
cholinesterase
measurements
and
micronucleus
study.
The
lack
of
a
reproduction
study
and
some
uncertainty
about
the
findings
in
offspring
in
the
developmental
neurotoxicity
study
are
obstacles
to
assessing
special
sensitivity
of
infants
and
children
and
as
such,
the
10x
FQPA
Safety
Factor
was
retained.
However,
the
database
provides
sufficient
information
for
selecting
toxicity
endpoints
for
risk
assessment
and
therefore,
supports
a
reregistration
eligibility
decision
for
the
currently
registered
uses.

Aventis'
response:
As
stated
previously,
Aventis
has
submitted
to
EPA
a
revised
DNT
and
a
new
2­
Generation
Reproduction
Study,
which
have
a
direct
impact
on
the
FQPA
Safety
Factor.
Aventis
firmly
believes
that
with
the
submission
of
these
two
studies,
the
data
gap
for
the
2­
generation
reproduction
study
is
fulfilled
and
the
DNT
NOEL
question
will
be
resolved.
Thus,
the
Agency
has
the
required
data
to
reduce
the
FQPA
Safety
Factor
from
10X
to
1X.
As
for
the
21­
day
dermal,
90­
day
inhalation
and
mouse
micronucleus
studies,
the
need
for
these
studies
is
addressed
in
the
previous
comments
made
by
Aventis.

Page:
17
Paragraph:
1
Lines:
1­
6
EPA
comment:
In
the
developmental
neurotoxicity
study,
clinical
signs
of
toxicity
and
plasma
and
brain
ChEI
were
seen
in
maternal
animals
at
the
same
dose
(10
mg/
kg/
day)
as
changes
in
brain
morphometric
measurements
(decreases
in
cerebellar
measurements
in
females
on
Day
11
post­
partum)
were
observed
in
offspring;
however,
brain
measurements
were
not
conducted
at
the
next
lower
dose.
The
lowest
NOAEL
after
a
single
dose
administration
in
adult
animals
was
for
maternal
animals
in
the
developmental
neurotoxicity
study,
i.
e.,
1
mg/
kg/
day.

Aventis'
response:
Aventis
has
submitted
to
EPA
a
revised
DNT
Study,
which
demonstrated
that
no
alterations
in
brain
morphometric
measurements
were
observed
in
the
offspring,
thus
no
increased
sensitivity
was
observed
in
the
pups.
The
details
of
this
study
15
will
be
discussed
below
in
the
supporting
`Discussion
Section'.

Page:
17
Paragraph:
2
Lines:
1­
3
EPA
comment:
No
subchronic
studies
in
the
rat
or
dog
are
available,
except
for
the
subchronic
neurotoxicity
study
in
rats.
The
chronic
toxicity
data
showed
that,
in
dogs,
decreases
in
plasma,
RBC
and
brain
ChE
were
observed
at
10
mg/
kg/
day;
clinical
signs
of
toxicity
were
also
observed
in
both
sexes.

Aventis'
response:
For
the
chronic
study
in
dogs,
Aventis
does
not
agree
with
the
additional
3X
for
a
lack
of
a
NOEL
in
the
study.
According
to
the
policy
issued
in
August
2000
concerning
the
endpoint
for
selection
for
cholinesterase
inhibiting
compound,
the
Agency
stated
that
the
RBC
should
be
used
instead
of
the
plasma.
The
NOEL
for
RBC
cholinesterase
inhibition
was
125
ppm
(3.1
mg/
kg/
day).
Additionally,
Aventis
believes
that
the
brain
cholinesterase
inhibition
observed
at
the
125
ppm
level
is
slightly
above
background
level
and
was
not
of
toxicological
concerns
due
to
the
lack
of
clinical
signs
observed
at
this
dose
level.
Furthermore,
Aventis
performed
a
five­
week
study
in
dogs,
which
agreed
with
the
results
of
the
chronic
study.
Therefore,
it
is
our
opinion
that
all
calculations
based
on
the
chronic
dog
in
the
document
should
be
adjusted
to
reflect
the
removal
of
the
3X
safety
factor.

Page:
17
Paragraph:
3
Lines:
1­
21
EPA
comment:
The
Health
Effects
Division's
(HED)
Cancer
Peer
Review
Committee
(CPRC)(
12/
8/
93)
classified
carbaryl
as
a
Group
C
­
possible
human
carcinogen
based
on
an
increased
incidence
of
hemangiosarcomas
and
combined
hemangiomas/
hemangiosarcomas
in
male
mice.
Both
the
low
dose
extrapolation
(Q1*)
and
margin
of
exposure
(MOE)
approaches
were
proposed
for
risk
assessment.
In
addition,
an
RfD
approach
would
be
provided
to
assess
the
most
sensitive
non­
cancer
health
endpoint
for
comparison
to
the
linear
and
MOE
approaches.
The
CPRC
requested
additional
metabolism
studies,
which
could
1)
direct
the
selection
of
the
more
appropriate
quantitative
approach;
and
2)
provide
insight
into
the
significance
of
the
tumors
seen
only
at
excessively
toxic
doses.
Additional
metabolism
studies,
including
mechanistic
studies,
were
submitted
subsequent
to
the
1993
meeting.
A
subgroup
of
the
Cancer
Assessment
Review
Committee
(CARC)
met
on
September
3,
1998,
to
review
the
metabolism
studies
and
concluded
that
the
data
from
all
available
metabolism
studies
were
not
adequate
to
support
a
nonlinear
mode
of
action,
as
described
in
the
1996
EPA
Proposed
Guidelines
for
Carcinogen
Risk
Assessment,
and
therefore
recommended
that
the
default
linear
approach
should
be
used
for
the
cancer
risk
assessment.
The
revised
Q1*,
based
on
the
CD­
1
mouse
dietary
study
with
¾
Interspecies
Scaling
Factor,
is
1.19
x
10
­2
(mg/
kg/
day)
­1
in
human
equivalents.
Subsequently,
the
registrant
submitted
a
special
study
in
genetically
modified
mice.
Carbaryl
was
administered
to
heterozygous
p53­
deficient
(knockout)
male
mice
in
the
diet
at
concentrations
of
up
to
4000
ppm
(716.6
mg/
kg/
day)
for
six
16
months.
There
was
no
evidence
of
neoplastic
or
preneoplastic
changes
in
the
vascular
tissues
of
any
organ.
A
model
validation
study
demonstrated
that
vascular
tumors
occur
in
heterozygous
p53
deficient
mice
within
six
months
of
administration
of
a
known
genotoxic
carcinogen
(urethane).

Aventis'
response:
Aventis
strongly
believes
that
the
use
of
carbaryl
products
presents
no
imminent
carcinogenic
risk
to
users
based
on
the
following
points:

 
Current
data
support
a
hypothesis
that
high
doses
of
carbaryl
in
life­
time
studies
produce
tumors
via
a
non­
genotoxic
mechanism,
possibly
related
to
altered
metabolism
at
these
high
doses.

 
At
doses
less
than
the
MTD,
there
was
an
increased
incidence
of
tumors
only
in
1
site
of
1
sex
of
1
species.

 
The
increased
incidence
of
vascular
tumors
in
the
male
mice
after
two
years
of
administration
is
of
questionable
biological
significance.

 
In
the
p53
knockout
mouse
model,
which
was
demonstrated,
to
be
sensitive
to
the
induction
of
vascular
tumors
by
a
genotoxic
reference
compounds,
carbaryl
was
found
to
be
negative.

 
The
weight­
of­
the
evidence
indicates
that
carbaryl
shows
no
potential
for
genotoxicity
 
Epidemiological
data
on
carbaryl
production
workers
show
no
increase
in
tumor
incidence.

As
background,
in
the
two­
year
bioassay
conducted
in
the
CD1
mouse,
a
statistically
significantly
higher
incidence
of
vascular
tumors
was
noted
in
males
at
1,000
and
8,000
ppm.
In
females,
no
statistically
significant
change
in
the
incidence
of
vascular
tumors
was
observed.
The
overall
incidence
of
vascular
tumors
was
2,
6,
10
and
10
in
males
and
3,
3,
4
and
9
in
females
at
0,
100,
1,000
and
8,000
ppm
respectively.

In
order
to
address
the
biological
significance
of
the
vascular
tumors,
the
p53
knockout
mouse
model
was
used
as
potential
tool
to
study
chemical
carcinogenesis
(Donehower,
1996).
The
p53
mouse
is
a
genetically
manipulated
mouse
in
which
one
allele
of
the
p53
tumor
suppresser
gene
has
been
inactivated.
One
of
the
assumptions
with
the
knockout
model
is
that
a
mutation
at
the
intact
p53
allele
is
necessary
for
the
development
of
the
carcinogenic
process.
In
principle
non­
genotoxic
compounds,
which
induce
tumors
by
other
mechanisms
should
not
induce
tumors
in
this
system.
The
p53
knockout
mouse
model
was
validated
by
testing
two
compounds:
urethane,
a
genotoxic
compound
known
to
produce
vascular
tumors
in
standard
carcinogenicity
bioassays,
and
d­
limonene,
17
which
is
neither
genotoxic
nor
carcinogenic
in
mice
but
which
is
known
to
be
carcinogenic
in
the
male
rat
by
a
well
described
non­
genotoxic
mechanism.
In
the
validation
study,
the
p53
model
proved
to
be
very
efficient
for
the
induction
of
vascular
tumors
by
genotoxic
compounds
(Bigot,
1999;
Carmichael
et
al.
1999).
Therefore,
this
model
will
produce
an
unambiguous
response
to
specifically
identify
genotoxic
compounds.

Therefore,
carbaryl
was
tested
in
the
p53
mouse.
The
objectives
of
the
studies
were
to
evaluate
if
carbaryl
would
induce
vascular
tumors
in
this
model
and
to
set
a
NOEL
in
this
sensitive
and
specific
model.
Additionally,
this
model
provides
evidence
that
the
tumors
observed
in
the
standard
carcinogenicity
bioassays
resulted
from
an
indirect
geneotoxic
effect
Carbaryl
was
administered
continuously
via
the
diet
to
groups
of
20
male
heterozygous
p53
knockout
mice
at
concentrations
of
0,
10,
30
100,
300,
1,000
and
4,000
ppm
for
at
least
180
days.
At
the
end
of
the
study,
all
animals
were
necropsied,
selected
organs
weighed
and
a
range
of
tissues
were
taken,
fixed
and
examined
microscopically.
Carbaryl
did
not
induce
mortalities
or
clinical
signs
related
to
the
treatment.
Only
a
slight
decrease
in
food
consumption
during
the
first
eight
weeks
was
observed
at
4,000
ppm.
This
observation
was
correlated
with
a
lower
body
weight
evolution
in
comparison
with
the
control
animals.

No
tumors
were
found
in
the
4,000
ppm
group.
In
the
other
treated
groups,
a
few
sporadic
tumors
were
found,
but
they
were
clearly
unrelated
to
treatment
and
representative
of
the
spontaneous
tumor
types
present
in
mice
of
this
age
and
strain.
In
particular,
it
should
be
noted
that
no
tumors
were
found
in
the
liver,
kidney
or
vascular
system,
which
were
seen
in
the
original
mouse
oncogenicity
study
with
carbaryl.
The
only
treatment­
related
non­
proliferative
change
observed
was
the
presence
of
globular
deposits
in
the
umbrella
cell
layer
of
the
urinary
bladder
at
100
ppm
or
more.

In
conclusion,
the
data
from
the
p53
studies,
the
following
conclusion
can
be
made:

1).
The
p53
model
was
validated
with
urethane
one
of
the
few
compound
known
to
induce
specifically
vascular
tumors.
2).
Carbaryl
was
negative,
and
also
did
not
accelerate
the
formation
of
any
other
tumor
types.
3).
There
is
a
clear
NOEL
at
4000
ppm
in
the
p53
mouse
study
for
carcinogenicity.
4).
Carbaryl
is
not
a
genotoxic
carcinogen.

Thus
the
weight­
of­
the
evidence
indicates
that
carbaryl
is
not
a
genotoxic
agent
in
humans.
In
addition,
epidemiological
data
on
carbaryl
production
workers
show
no
increase
in
tumor
incidence.
The
two
epidemiology
studies
of
Aventis
CropScience
(formerly
Rhône­
Poulenc)
factory
workers
representing
a
sub­
18
population
of
the
general
public
with
the
highest
exposure
(i.
e.,
exposure
on
a
daily
basis
over
many
years).
Thus,
the
data
from
these
studies
represent
a
potential
"worst­
case"
for
chronic
carbaryl
exposure
to
humans.
The
results
from
these
studies
show
that
the
overall
mortality
experience
of
the
cohort
is
significantly
less
than
expected
when
compared
to
both
United
States
and
West
Virginia
male
death
rates.
In
general,
these
studies
support
conclusion
that
there
were
no
excess
cases
of
cancer
in
this
working
population.

Page:
18
Paragraph:
1
Lines:
1­
8
EPA
comment:
Carbaryl
was
not
mutagenic
in
the
Salmonella
typhimurium
assay
or
genotoxic
in
the
Unscheduled
DNA
Synthesis
assay.
In
a
chromosomal
aberration
test
using
Chinese
Hamster
Ovary
(CHO)
cells,
carbaryl
was
clastogenic
in
the
presence
of
S9
activation.
The
CPRC
(1993
meeting)
required
an
in
vivo
cytogenetics
study
in
rodents
to
provide
insight
into
the
structural
and/
or
numerical
aberrations
in
the
study.
The
mouse
micronucleus
study
submitted
to
satisfy
this
requirement
was
deemed
unacceptable.
HED
concluded
that,
based
on
a
Weight
of
the
Evidence
assessment,
the
database
does
not
support
that
carbaryl
acts
as
a
DNA­
reactive
mutagen.
However,
this
alone
is
not
sufficient
to
demonstrate
a
mode
of
action
for
establishing
a
threshold
cancer
risk
assessment.

Aventis'
response:
The
Agency
has
previously
granted
a
wavier
for
the
in
vivo
cytogenicity/
micronucleus.
Furthermore,
Aventis
has
fulfilled
the
data
requirement
for
the
in
vivo
mouse
micronucleus
study.
Aventis
believes
that
repeating
the
in
vivo
mouse
micronucleus
study
is
not
appropriate
based
the
fact
that
the
top
dose
in
the
currently
submitted
mouse
micronucleus
was
200
mg/
kg/
day
and
was
similar
to
the
mid­
dose
of
250
mg/
kg/
day
tested
in
an
acute
gavage
study
in
mice.
At
the
250
mg/
kg/
day
dose
level
tested
in
the
acute
study
plasma,
RBC,
and
brain
cholinesterase
inhibition
was
seen.
The
percent
cholinesterase
inhibition
observed
was
45.7,
45.7,
and
57.8,
respectively.
At
the
highest
dose
tested
(500
mg/
kg)
in
the
acute
study,
percent
cholinesterase
inhibition
for
plasma,
RBC,
and
brain
were
59.8,
57.1,
and
66.6,
respectively.
Additionally,
the
clinical
signs
observed
at
both
the
250
and
500
mg/
kg
were
very
similar
in
both
findings
and
incidences.
Repeating
the
mouse
micronucleus
study
at
a
slightly
higher
dose
level
would
not
result
in
an
increase
in
significant
clinical
signs.
Therefore,
the
base
results
of
the
study
concerning
clastogenic
or
aneugenic
effects
would
not
change.
Carbaryl
is
not
mutagenic
for
these
endpoints.

Page:
18
Paragraph:
6
Lines:
1­
8
EPA
comment:
The
FQPA
Safety
Committee
concluded
at
meetings
on
November
29,
1999
and
April
16,
2001,
that
the
10x
safety
factor
should
be
retained
because:
1)
the
toxicology
data
base
is
incomplete;
there
is
a
data
gap
for
the
multi­
generation
reproduction
study
in
rats;
2)
an
assessment
of
susceptibility
following
pre­/
post­
19
natal
exposure
to
carbaryl
could
not
be
made
due
to
the
data
gap
for
the
reproduction
study;
3)
there
is
concern
for
the
results
of
the
developmental
neurotoxicity
study
(uncertainty
about
NOAEL/
LOAEL
for
brain
morphometric
alterations).
The
Committee
concluded
the
10x
safety
factor
should
be
applied
to
all
population
subgroups
when
assessing
acute
and
chronic
dietary
exposures
and
residential
exposures
of
all
durations.

Aventis'
response:
As
stated
above,
Aventis
has
submitted
to
EPA
a
revised
DNT
Study
and
a
new
2­
Generation
Reproduction
Study
which
will
have
a
direct
impact
on
the
FQPA
Safety
Factor.
Aventis
firmly
believes
that
with
the
submission
of
these
two
studies,
the
data
gap
for
the
2­
generation
reproduction
study
will
be
fulfilled
and
clarification
of
the
DNT
NOEL
issue
will
be
provided.
Thus
the
Agency
has
the
required
data
to
reduce
the
FQPA
Safety
Factor
from
10X
to
1X.
20
Part
I
­
Toxicology
Data
Base
SUPPORTING
DISCUSSIONS
Supplemental
Developmental
Neurotoxicity
Study
Findings
The
document
submitted
to
EPA
describes
the
supplemental
histomorphometric
evaluation
performed
on
the
cerebellum
of
rat
pups
and
adults
from
the
carbaryl
developmental
neurotoxicity
study
(Robinson
and
Broxup
2001).
This
additional
work
was
conducted
in
response
to
the
EPA
review
of
the
study
final
report.
In
that
review,
EPA
indicated
that
the
bilateral
decrease
in
the
length
of
the
cerebellum
accompanied
by
a
non­
statistically
significant
5%
decrease
in
cerebellar
weights
in
the
day
11
females
and
the
bilateral
increase
in
the
width
of
the
cerebellum
in
the
day
70
female
animals
at
the
highest
tested
dose
(10
mg/
kg/
day)
may
possibly
be
treatment
related.
Further,
some
forebrain
measurements
may
have
also
been
affected.
In
order
to
clarify
the
treatment
relationship
of
these
findings,
the
Agency
recommended
that
additional
morphometric
measurements
be
performed
to
support
the
NOEL
and
that
the
thickness
of
the
cellular
layers
in
the
cerebellum
be
more
fully
described.

The
submitted
report
addresses
the
EPA
comments
by
presenting
additional
evaluations
of
the
high
dose
animals,
only.
The
evaluations
clearly
demonstrate
that
there
are
no
treatment­
related
morphometric
findings
in
the
cerebellum
of
male
and
female
pups
and
adults.
These
findings
are
in
agreement
with
the
results
from
the
high
dose
group
animals
in
the
study
final
report
that
found
no
treatment­
related
changes
in:
1)
Brain
weights
in
male
and
female
pups;
2)
Cerebellar
weights
in
male
and
female
pups;
3)
Terminal
body
weights
in
male
and
female
pups;
4)
Brain
weights
in
male
and
female
adults;
5)
Terminal
body
weights
in
male
and
female
adults;
6)
FOB
measurements
and
motor
activity
in
male
and
female
pups;
7)
Motor
activity,
auditory
startle
response,
passive
avoidance,
and
water
maze
measurements
in
male
and
female
adults.

The
evaluations
clearly
demonstrate
there
are
no
treatment
related
morphometric
alterations
of
the
cerebellum
in
male
and
female
pups
or
adults
treated
with
carbaryl
at
a
dose
of
10
mg/
kg/
day,
from
Day
6
of
gestation
to
Day
10
post­
partum,
inclusively.
In
addition,
in
the
forebrain
no
measurable
bilateral
differences
are
detected
between
control
and
carbaryl
exposed
pups
and
adults.

In
conclusion,
carbaryl
did
not
induce
any
morphologic
or
morphometric
changes
in
either
the
cerebellum
or
the
forebrain
in
animals
receiving
a
high
dose
of
10
mg/
kg/
day.
As
described,
the
new
morphometric
results
revealed
no
changes
at
the
high
dose,
are
in
agreement
with
the
lack
of
treatment­
related
effects
on
terminal
body,
brain,
and
cerebellar
weights,
as
well
as
behavioral
and
motor
activity
tests
in
male
and
female
pups
and
adults.
21
2­
Generation
Reproduction
Study
Findings
A
2­
Generation
reproduction
study
with
carbaryl
technical
was
completed
at
Research
Triangle
Institute
located
in
North
Carolina,
USA,
in
order
to
provide
information
on
the
influence
of
carbaryl
technical
on
the
reproductive
performance
in
rats
(Tyl
et
al.
2001).
The
purpose
of
this
study
was
to
evaluate
the
potential
of
carbaryl,
administered
in
the
feed
to
CD
rats,
to
produce
alterations
in
parental
fertility,
maternal
pregnancy
and
lactation,
and
growth
and
development
of
the
offspring
for
two
generations,
one
litter
per
generation.
This
study
was
performed
in
compliance
with
U.
S.
EPA
FIFRA
GLP
Standards
(U.
S.
EPA,
1989),
the
U.
S.
EPA
OPPTS
Testing
Guidelines
(U.
S.
EPA,
1998),
and
the
OECD
testing
guidelines
for
a
two­
generation
reproductive
toxicity
study
(OECD,
1983).

Male
and
female
CD®
(Sprague­
Dawley)
rats
(the
FO
generation)
were
administered
carbaryl
(1­
naphthyl
methylcarbamate;
CAS
No.
63­
25­
2)
in
the
feed
at
0,
75,
300,
and
1500
ppm,
available
ad
libitum,
30
animals/
sex/
dose,
for
ten
weeks.
Body
weights
and
feed
consumption
were
recorded
weekly,
and
clinical
signs
were
recorded
at
least
once
daily.
Vaginal
cytology
was
evaluated
for
the
last
three
weeks
of
the
prebreed
period.
Animals
were
then
randomly
mated
within
treatment
groups
for
a
two­
week
mating
period
to
produce
the
Fl
generation,
with
exposure
continuing.
FO
males
were
necropsied
after
the
delivery
period,
with
histological
evaluation
of
reproductive
and
other
organs
and
andrological
assessments
(reproductive
organ
weights,
epididymal
sperm
number,
motility
and
morphology,
testicular
homogenization­
resistant
spermatid
head
counts,
daily
sperm
production,
and
efficiency
of
daily
sperm
production).
Fl
litters
were
culled
to
ten
pups
on
postnatal
day
(pnd)
4
and
weaned
on
pnd
21.
At
weaning,
up
to
three
weanlings/
sex/
litter
were
necropsied,
and
30/
sex/
dose
were
selected
as
Fl
parents
of
the
F2
generation.
FO
females
were
then
necropsied
with
organ
weights,
stage
of
estrus
at
necropsy,
enumeration
of
ovarian
primordial
follicles,
and
histopathology
of
reproductive
and
other
selected
organs.
Selected
Fl
weanlings,
30/
sex/
dose,
were
administered
carbaryl
in
the
diet
for
a
ten­
week
prebreed
exposure
period,
with
acquisition
of
vaginal
patency
in
females
and
preputial
separation
in
males
assessed,
and
vaginal
cytology
for
estrous
cyclicity
in
Fl
selected
females
evaluated
during
the
last
three
weeks
of
the
prebreed
exposure
period.
They
were
mated
for
a
two­
week
period,
as
described
above.
At
weaning
of
F2
litters,
up
to
three
weanlings/
sex/
litter
were
necropsied.
Fl
males
were
necropsied
after
the
delivery
period,
with
histopathology
and
andrological
assessments
(as
described
above).
At
weaning
of
the
F2
litters,
parental
Fl
females
were
then
necropsied
with
histopathology,
as
described
above,
and
F2
weanlings,
up
to
three/
sex/
litter,
were
necropsied.

Dietary
exposure
to
carbaryl
for
two
generations,
one
litter
per
generation,
at
0,
75,
300,
and
1500
ppm,
resulted
in:
decreased
body
weights
and
decreased
weight
gains
in
P0
and
P1
parental
males
and
females,
accompanied
by
decreased
feed
consumption
at
1500
ppm,
and
slightly
reduced
body
weights
and
feed
consumption
at
300
ppm;
Fl
and
P2
offspring
toxicity
(reduced
body
weights
during
lactation
beginning
on
post­
natal
day
(pnd)
4
through
lactation
and
continuing
in
selected
P1
offspring
through
acquisition
of
puberty)
at
1500
ppm;
increases
in
P1
pup
mortality
at
1500
ppm
and
in
P2
pup
mortality
22
at
300
ppm
and
1500
ppm
during
lactation,
especially
early
(pnd
0­
4),
accompanied
by
reduced
maternal
body
weights
in
FO
and
P1
dams
at
1500
ppm
and
300
ppm,
and
delayed
vaginal
opening
and
preputial
separation
in
selected
Fl
offspring
at
1500
ppm
considered
an
indirect
effect
related
to
body
weight
decreases.

In
conclusion,
the
NOELs
for
carbaryl
administered
in
the
diet,
in
CD®
(SD)
rats
under
the
conditions
of
this
study,
were:
°
parental
systemic
toxicity:
75
ppm
°
parental
reproductive
toxicity:
1500
ppm
°
offspring
toxicity:
75
ppm
EPIDEMIOLOGIC
DATA
­
Epidemiological
data
on
carbaryl:
standardized
mortality
ratio
analysis
of
employees
exposed
to
carbaryl
at
the
Rhône­
Poulenc
Institute,
West
Virginia
plant.

The
National
Institute
of
Occupational
Safety
and
Health
(NIOSH)
and
the
Union
Carbide
Corporation
(UCC)
on
chemical
production
plants
in
the
Kanawha
Valley
(KV)
of
West
Virginia
originally
initiated
an
epidemiological
study.
One
of
the
plants
included
in
this
extensive
epidemiology
study
was
the
facility
at
which
carbaryl
is
produced
in
Institute,
West
Virginia.
The
plant
was
acquired
by
Rhône­
Poulenc
(now
Aventis
CropScience)
from
UCC
in
December
1986.

The
study
(Pastides
1993)
reported
here
investigated
the
total
and
cause­
specific
mortality
experience
of
employees
exposed
to
carbaryl
at
the
production
plant.
No
additional
information
other
than
what
was
available
in
the
original
database
was
collected
about
the
health
or
mortality
experience
of
individuals
employed
at
this
plant
for
this
study.
The
central
results
of
this
report
were
presented
in
terms
of
Standardized
Mortality
Ratios
(SMR's)
and
95%
confidence
intervals
(95%
CI's).

Carbaryl
technical
is
manufactured,
packaged,
and
shipped
from
the
Aventis
CropScience
plant
in
Institute,
West
Virginia.
Besides
carbaryl,
the
plant
also
produces
other
agricultural
products.
In
the
past,
numerous
other
chemical
compounds
were
produced
there,
including
butanol,
styrene,
acetaldehyde,
toluene,
fluorocarbons
and
others.

Carbaryl
is
produced
in
crystalline
form,
as
well
as
in
an
oil­
based
solution
(41%
carbaryl).
Final
processing
includes
crystallizing,
drying,
bin
storage,
and
packing.
Exposure
to
carbaryl
would
occur
among
three
basic
categories
of
employees:
those
in
the
manufacturing
unit,
those
in
maintenance,
and
those
in
packaging
and
distribution.

For
employees
of
the
carbaryl
unit,
UCC
tracked
through
1988
the
vital
status,
and
cause
of
death
information,
of
all
individuals
who
were
first
hired
between
the
start­
up
of
the
carbaryl
unit
in
1960
through
1978.
Employees
hired
after
1978
were
not
part
of
the
NIOSH/
UCC
KV
study.
23
Using
the
KV
database,
UCC
conducted
SMR
analyses
of
the
employees
from
the
entire
Institute
plant
in
the
past.
However,
no
separate
examination
of
data
from
employees
specifically
exposed
to
carbaryl
was
made.
In
the
overall
Institute
study,
no
excess
risk
in
total
mortality,
total
cancer
mortality,
or
site­
specific
cancer
mortality,
was
observed.

In
the
carbaryl
specific
study,
a
total
of
522
employees
were
identified
within
the
UCC
database
as
belonging
to
one
of
the
three­
carbaryl
exposure
groups;
(158­
Carbaryl
Production,
229­
Packing
and
Distribution,
327­
Maintenance)

Because
of
some
overlap
between
the
groups
and
because
some
individuals
did
not
meet
the
NIOSH/
UCC
cohort
entry
criteria,
the
final
number
of
employees
available
for
the
SMR
analysis
was
488.
These
488
employees
contributed
7,531.5
person­
years
to
the
mortality
analysis.
This
number
represents
the
combined
number
of
years
in
which
these
employees
were
followed
through
1988.

As
of
the
end
of
1988,
twenty­
five
deaths
were
identified
from
this
cohort
of
employees.
Elevated
SMR's,
reflecting
an
observed
number
of
deaths
greater
than
the
number
expected
were
seen
for
cancer
of
the
pancreas,
cancer
unspecified,
and
cancer
of
the
brain
and
other
parts
of
the
nervous
system.
In
the
former
two
categories,
the
excess
was
slight
and
based
on
only
a
single
death;
furthermore
the
wide
confidence
intervals
suggest
a
relatively
imprecise
SMR
estimate
and
one
that
is
well
within
the
range
of
chance
variation.
In
the
case
of
brain
and
nervous
system
cancer,
the
higher
SMR
suggested
a
possibility
of
some
association
with
work
in
the
carbaryl
unit,
yet
the
very
wide
confidence
interval
indicates
that
the
SMR
estimate
has
low
precision.
In
other
words,
the
estimate
is
unstable,
is
within
the
range
of
chance
variation,
and
reflects
the
small
sample
size
on
which
it
is
based.
Furthermore,
review
of
the
death
certificates
of
these
two
individuals
revealed
that
these
tumors
were
of
different
histologic
origin;
one
was
reported
as
an
astrocytoma,
the
other
as
a
glioblastoma
multiforme.
This
information
reduces
the
plausibility
that
both
malignancies
were
caused
by
the
same
exposure.

The
next
phase
of
the
carbaryl
plant
worker
epidemiology
study
has
been
completed
(Pastides
and
Zorn,
1997),
and
the
results
are
based
on
vital
status
of
employees
through
1994.
The
new
cohort
consisted
of
817
employees
(488
from
the
previous
analysis
and
329
who
were
hired
after
1978).
Taking
into
account
restriction
criteria,
as
described
in
the
report,
the
restricted
cohort
that
reflects
the
number
of
employees
who
worked
across
departments
is
599.
As
previously
shown,
the
overall
mortality
experience
of
the
cohort
is
significantly
less
than
expected
when
compared
to
both
United
States
and
West
Virginia
male
death
rates.
In
general,
the
follow­
up
study
supported
the
results
from
the
initial
study
that
there
were
no
excess
cases
of
cancer
in
this
working
population.
24
Carcinogenicity
Issue
–
Historical
Perspective
and
as
presented
to
JMPR
and
submitted
to
EPA
in
2000
A.
Combined
Oncogenicity/
Chronic
Toxicity
Studies
1)
Carbaryl
produced
tumors
primarily
at
the
highest
dose
tested.

The
dietary
concentrations
in
the
rat
study
were
0,
250,
1500,
and
7500
ppm
(approximately
13,
75,
and
375
mg/
kg
b.
w.);
for
the
mouse
study
the
concentrations
of
technical
carbaryl
were
0,
100,
1000,
and
8000
ppm
(approximately
12,
143,
and
1143
mg/
kg).
The
highest
doses
were
chosen
to
satisfy
the
U.
S.
EPA
requirement
for
an
MTD,
while
the
lower
doses
were
selected
for
determining
the
NOEL
and
to
produce
moderate
toxicity
in
tissues
and
on
cholinesterase.
Considering
an
approximate
acute
oral
LD
50
in
rats
and
mice
of
250
mg/
kg,
these
doses
would
therefore
be
equivalent
to
0.06,
0.6
and
4.6
times
the
acute
oral
LD
50
for
mice
and
0.05,
0.3
and
1.5
times
the
acute
oral
LD
50
for
rats.
(The
high
doses
for
the
mouse
and
rat,
respectively,
also
would
be
equivalent
on
a
body
weight
basis
to
a
70­
kg
person
consuming
approximately
80
grams
and
26
grams
of
technical
carbaryl
per
day
for
a
lifetime.)

These
MTD
levels
caused
a
significant
reduction
in
body
weight
and
body
weight
gain
in
both
species,
especially
early
in
the
study.
Several
other
parameters,
such
as
cholinesterase,
were
also
significantly
affected
throughout
the
studies
at
the
top
doses.
The
U.
S.
EPA
in
their
review
of
the
studies
agreed
that
the
highest
doses
in
the
mouse
and
rat
study
exceeded
the
MTD.
As
a
result,
these
MTD
levels
were
deemed
inappropriate
for
chronic
testing.

Therefore,
the
only
relevant
findings
for
establishing
a
carcinogenic
classification
are
the
vascular
system
tumors
in
male
mice.

2)
The
data
supporting
the
oncogenic
potential
of
carbaryl
is
equivocal
since
tumors
occurred
only
at
one
site
of
one
sex
of
one
species
when
irrelevant
data
at
doses
exceeding
the
MTD
are
excluded
from
consideration.

Rat
Study
In
the
rat
carcinogenicity
study,
carbaryl
produced
tumors
in
both
sexes
(thyroid
and
bladder
in
males;
liver
and
bladder
in
females)
but
only
at
the
MTD
(375
mg/
kg).
At
the
low
and
intermediate
doses
(13
and
75
mg/
kg,
respectively)
there
was
no
indication
of
an
increased
incidence
of
tumors
at
any
of
the
sites
for
which
increases
occurred
at
the
top
dose.
Thus,
at
levels
less
than
the
MTD,
carbaryl
has
not
produced
tumors
in
rats,
whether
considering
the
current
study
or
a
previous
study
in
which
another
rat
strain
was
tested.
25
Mouse
Study
In
the
mouse
carcinogenicity
study,
carbaryl
produced
tumors
(liver
and
vascular)
in
females
only
at
the
highest
dose
tested
(1143
mg/
kg).
At
the
low
and
intermediate
doses
(12
and
143
mg/
kg,
respectively)
there
was
no
indication
of
an
increased
incidence
of
tumors
at
any
of
the
sites
for
which
increases
occurred
at
the
top
dose.
Thus,
at
doses
less
than
the
MTD,
carbaryl
did
not
produce
tumors
in
female
mice
in
the
current
study
or
in
an
older
study.

In
male
mice,
carbaryl
produced
tumors
in
the
vascular
system
and
kidney
at
the
highest
dose
tested
(1143
mg/
kg
b.
w.).
At
the
low
and
intermediate
doses
(12
and
143
mg/
kg)
there
was
no
indication
of
an
increased
incidence
in
kidney
tumors.
At
the
low
and
intermediate
doses
there
was
an
apparent
increase
in
the
incidence
of
vascular
tumors
as
follows:

Dose
Male
Mouse
Vascular
Tumors
(mg/
kg)
Benign
Malignant
Total
0
122
12
1
6
7
143
1
9
10
1143
3
7
10
However,
despite
a
nearly
10­
fold
increase
in
dose
increments,
the
total
tumors
observed
at
the
intermediate
and
high
doses
were
the
same.
The
number
of
malignant
tumors
actually
was
slightly
lower
at
the
high
dose
than
at
the
intermediate
dose.
Thus,
these
results
show
there
is
no
clear
linear
dose­
response
relationship
from
the
intermediate
dose
to
the
high
dose,
despite
a
nearly
10­
fold
difference
in
dose
increment.

With
regard
to
statistical
significance,
the
performing
laboratory
indicated
that
statistical
significance
was
just
achieved
at
the
p<
0.05
levels
for
the
top
dose,
while
the
intermediate
dose
results
were
just
above
this
level
of
statistical
significance
(this
result
was
due
to
the
actuarial­
type
statistical
analysis
used
for
tumor
incidence
and
onset).
However,
the
U.
S.
EPA
and
U.
K.
authorities
have
performed
separate
statistical
analyses
and
the
tumor
incidence
at
the
intermediate
dose
also
was
found
to
fall
within
statistical
significance.
Based
on
the
results
from
the
original
study
report,
a
NOEL
could
be
established
based
on
the
lowest
dose
tested
at
approximately
12
mg/
kg
b.
w.
However,
arguments
are
made
in
succeeding
sections
that
18­
month
historical
control
data
from
the
performing
laboratory
are
inadequate
to
determine
if
the
incidence
of
tumors
at
any
of
these
doses
is
biologically
significant.

As
previously
stated,
vascular
tumors
were
produced
in
the
female
mice
but
only
at
the
top
dose
and
with
a
very
similar
incidence
(N
=
9)
as
observed
in
the
males
at
the
top
dose.
In
most
other
aspects
of
the
study,
e.
g.,
body
weight
effects,
clinical
signs,
clinical
26
pathology,
etc.,
the
females
reacted
very
similar
to
males.
However,
there
was
no
indication
of
an
increased
incidence
of
vascular
tumors
in
females
at
the
low
or
intermediate
doses
as
was
observed
in
the
male
mice.
Thus,
these
facts
raise
questions
regarding
the
biological
significance
of
the
vascular
tumors,
which
occurred
in
the
male
mice.

3)
Adequate
historical
control
data
were
not
available
from
the
laboratory
conducting
the
chronic
toxicity
and
oncogenicity
studies.

Historical
data
from
the
laboratory
at
which
the
mouse
oncogenicity
study
was
conducted
was
only
available
for
studies
conducted
up
to
18
months.
Aventis
CropScience
was
not
certain
if
these
data
accurately
reflected
the
incidence
of
vascular
tumors
in
24­
month
CD­
1
mice.
Additional
information
was
obtained
from
other
sources
and
summarized
(Klonne,
1995)
to
compare
historical
control
data
from
several
sources
and
at
several
study
intervals
to
that
of
the
data
for
the
carbaryl
study.
The
following
key
observations
were
made:

 
An
increase
in
the
spontaneous
vascular
tumor
incidence
appears
to
occur
from
18
to
24
months
of
age
in
mice.

 
An
increase
in
the
spontaneous
vascular
tumor
incidence
over
the
last
10
years
may
be
occurring
in
CD­
1
mice.

 
The
carbaryl
study
(a
2­
year
study)
should
be
compared
to
historical
control
data
from
2­
year
old
mice
generated
during
the
last
10
years.

In
general,
considering
the
most
relevant
historical
control
data
outside
of
the
performing
laboratory,
the
incidences
of
vascular
tumors
in
major
organs
(spleen/
liver)
in
male
mice
at
the
low
and
middle
doses
fall
within
the
historical
control
ranges.

4)
The
lack
of
significant
histopathological
findings
at
the
one­
year
interim
sacrifice
of
both
the
mouse
and
rat
studies
did
not
correlate
with
the
detection
of
tumors
in
certain
tissues
at
the
end
of
the
study.

To
further
investigate
the
mechanism
by
which
carbaryl
induced
multiple
tumors
in
the
rat
and
the
mouse
after
two
years
of
exposure,
a
decision
was
made
to
re­
evaluate
the
histological
slides
of
target
organs
from
the
interim
sacrifice
after
one
year
of
treatment.
The
purpose
of
this
work
was
to
identify
if
subtle
changes
present
at
one
year
could
explain
the
appearance
of
tumors
seen
at
two
years.

Histopathological
examinations
were
conducted
on
the
target
organs
(liver,
kidney,
thyroid
gland,
urinary
bladder)
from
the
control
and
high
dose
groups
in
rats
and/
or
mice.
The
review
was
conducted
independently
by
two
of
Aventis
CropScience's
pathologists
(Debruyne
and
Irisarri,
1996).

In
the
rat,
at
the
end
of
a
52­
week
exposure
period
to
carbaryl
technical
by
the
dietary
27
route,
the
re­
evaluation
of
the
histological
slides
revealed
the
presence
of
microscopic
changes
not
previously
reported
in
the
bladder
(transitional
epithelial
hyperplasia
in
both
male
and
females),
kidney
(pelvic
urothelial
hyperplasia
in
males),
and
thyroid
(thyroid
follicular
hypertrophy
in
males)
and
liver
(hepatocellular
hypertrophy
in
males
and
females).

In
the
mouse,
no
microscopic
changes
were
detected
at
the
end
of
the
52­
week
exposure
period.

These
new
findings
prompted
the
Aventis
CropScience
to
conduct
further
studies
to
confirm
cell
proliferation
in
the
various
tissues
(associated
with
positive
tumor
formation)
and
to
re­
examine
the
histological
slides
from
the
chronic
toxicity
and
oncogenicity
studies.

5)
Oncogenicity
and
chronic
toxicity
slide
review
Aventis
CropScience
then
commissioned
independent
pathologists
to
re­
examine
the
histological
slides
at
the
1­
year
and
2­
year
sacrifices
from
the
chronic
toxicity
and
oncogenicity
studies
to
determine
what
other
(if
any)
discrepancies
exist
outside
of
those
determined
by
the
Aventis
CropScience's
internal
review.
This
peer
review
study
was
conducted
according
to
US.
EPA
Peer
Review
policy
guidelines.
The
results
of
this
review
showed
no
differences
from
the
original
study
pathologists
review.

B.
Carcinogenic
Mechanism
1)
Subsequent
to
the
two­
year
studies,
Aventis
CropScience
proactively
performed
in
vivo
genotoxicity
studies.

In
response
to
the
results
of
the
mouse
oncogenicity
study,
Aventis
CropScience
proactively
conducted
an
in
vivo
DNA
adduct
study
in
mice
at
a
concentration
similar
to
the
top
dose
used
in
the
two­
year
carcinogenicity
study
(i.
e.,
8000
ppm)
to
determine
if
carbaryl
caused
any
genotoxic
effects
(Sagelsdorff,
P.
1994).
Results
indicated
that
carbaryl
did
not
interact
with
the
DNA
in
mice,
even
at
the
excessive
dose
used
in
the
study.
Additionally,
an
in
vivo
rat
bone
marrow
chromosomal
aberration
study
was
conducted
and
showed
that
carbaryl
did
not
produce
chromosomal
aberrations
at
doses
up
to
approximately
50%
of
the
acute
oral
LD
50
(McEnaney,
1993).
Although
there
was
no
indication
of
genotoxic
effects
in
the
rat
chromosomal
aberration
and
mouse
DNA
adduct
studies,
Aventis
CropScience
recently
conducted
an
in
vivo
micronucleus
study
in
mice
to
determine
if
carbaryl
produced
aneuploidy
in
this
species.
Results
from
the
mouse
micronucleus
assay
again
showed
no
genotoxic
effects
(Marshall,
1996).

Thus,
results
from
these
in
vivo
studies
indicate
that
the
mechanism
of
action
of
carbaryl
in
the
production
of
tumors
in
the
two­
year
oncogenicity
studies
does
not
occur
via
a
direct
genotoxic
effect.
28
2)
The
weight­
of­
the­
evidence
approach
indicates
that
carbaryl
shows
little
potential
as
a
genotoxin.

There
have
been
nearly
50
literature
and
Aventis
CropScience
reports
on
the
genotoxicity
of
carbaryl.
The
number
of
test
systems
and
endpoints
is
extensive.
In
neither
bacterial
nor
mammalian
cell
cultures
has
carbaryl
demonstrated
any
significant
mutagenic
potential.
In
various
in
vitro
DNA
damage
and
repair
assays
there
is
no
convincing
evidence
to
suggest
that
carbaryl
produces
DNA
damage.
While
carbaryl
has
shown
some
clastogenic
potential
in
vitro,
these
changes
occur
at
or
near
cytotoxic
levels.
In
addition,
studies
performed
in
vivo,
the
mouse
DNA
adduct
and
rat
chromosomal
aberration
(previously
discussed),
a
mouse
dominant
lethal
study,
and
an
evaluation
for
mouse
micronucleated
polychromatic
erythrocytes
all
have
been
negative.

The
absence
of
tumor
induction
in
a
6­
month
carcinogenicity
study
conducted
in
the
p53
knockout
mouse
demonstrated
that
the
tumors
observed
in
the
standard
two­
year
bioassays
in
rats
and
mice
are
not
linked
to
an
indirect
genotoxic
mechanism.

3)
It
is
possible
that
altered
metabolism
from
unrealistically
high
doses
of
carbaryl
plays
a
significant
role
in
the
formation
of
these
tumors.

The
results
from
the
metabolism
study
indicated
that
the
carbaryl
was
almost
completely
absorbed
and
metabolized
(approximately
90%)
and
that
there
was
essentially
no
difference
in
the
metabolism
between
the
sexes
or
in
the
low
versus
the
high
dose.
Likewise,
there
was
no
difference
in
the
metabolism
between
the
single
versus
multiple
doses.

The
data
also
indicated
that
the
low
dose
was
almost
completely
eliminated
in
12
hours
and
the
high
dose
was
almost
completely
eliminated
in
24
hours.
Two
metabolites,
5,6­
dihydro­
5,
6­
dihydroxy
carbaryl
and
3,4­
dihydro­
3,
4­
dihydroxy
carbaryl
were
identified
to
be
approximately
8%
and
1%,
respectively,
of
the
total
dose
and
were
found
to
be
primarily
conjugated
to
glucuronide.
These
metabolites
are
very
likely
result
from
the
metabolism
of
epoxide
intermediates.
Other
metabolites,
which
were
likely
formed
from
the
epoxide
intermediates,
were
identified
as
5­
hydroxycarbaryl
(13%)
and
4­
hydroxycarbaryl
(6%)
and
conjugated
carbaryl
(3%).

Epoxide
intermediates
have
been
proposed
to
be
the
proximate
carcinogen
for
several
classes
of
carcinogens.
Depending
on
such
factors
as
the
stability
of
the
epoxide
intermediate,
the
ability
of
the
cell
to
detoxify
the
epoxide
(related
to
such
factors
as
the
available
glutathione
stores,
epoxide
hydrase
activity,
etc.),
the
excretion
pattern
of
the
compound,
etc.,
these
reactive
intermediates
may
be
routinely
handled
by
the
body
just
like
other
endogenous
epoxides
or
could
be
available
to
react
with
cellular
components.
Thus,
it
is
possible
that
excessively
high
doses
of
carbaryl
could
alter
the
normal
metabolism,
distribution,
and/
or
excretion
pattern
in
many
different
ways,
e.
g.,
saturate
the
normal
metabolic
pathways
with
a
shift
of
carbaryl
metabolism
through
the
epoxide
29
intermediates,
deplete
the
available
glutathione
stores
available
for
conjugation,
increase
the
half­
life
of
the
epoxide
intermediates,
increase
tissue
concentrations
of
carbaryl
and
its
metabolites
due
to
diminished
ability
for
excretion,
etc.
Additionally,
saturation
of
many
enzyme
systems
with
large
carbaryl
substrate
concentrations
could
allow
accumulation
of
endogenous
chemicals
and
by­
products
of
normal
metabolism
that
would
otherwise
be
detoxified
and
excreted.

4)
Vascular
tumors.

In
the
two­
year
bioassay
conducted
in
the
CD1
mouse,
a
statistically
significantly
higher
incidence
of
vascular
tumors
was
noted
in
males
at
1,000
and
8,000
ppm.
In
females,
no
statistically
significant
change
in
the
incidence
of
vascular
tumors
was
observed.

A
search
of
the
literature
indicated
there
is
very
few
chemical
agents
known
to
induce
vascular
tumors
in
humans,
among
them
utherane
and
vinyl
chloride
were
identified
(Creech,
J.
L.
Jr
et
Johnson,
M.
N.
1974).
Angiosarcoma
of
liver
in
the
manufacture
of
polyvinyl
chloride.
(J.
Occup.
Med.
16:
150­
151;
Marion,
M.
J.,
De
Vivo,
I.,
Smith,
S.,
Luo,
J.
C.
and
Brandt­
Rauf,
P.
W.
1996).
The
molecular
epidemiology
of
occupational
carcinogenesis
in
vinyl
chloride
exposed
workers
(Int.
Arch.
Occup.
Environ.
Health,
68:
394­
398)
and
the
mechanism
underlying
the
formation
of
vascular
tumors
is
through
the
formation
of
etheno­
adducts,
formed
maybe
by
reactive
oxygen
species
(Barbin,
A.
2000.
Etheno­
adduct­
forming
chemicals:
from
mutagenicity
testing
to
tumor
mutation
spectra.
Mutation
Res.
462:
55­
69:
Nair,
J.,
Barbin,
A.,
Velic,
I.
and
Bartsch,
H.
1998).
Etheno
DNA­
base
adducts
from
endogenous
reactive
species
(Mutation
Res.
424:
59­
69).
All
those
compounds
would
have
been
found
to
induce
vascular
tumors
in
the
p53
model.
Thus,
the
study
using
the
genetically
modified
heterozygous
p53
knockout
mouse
was
designed
to
address
the
potential
role
of
an
indirect
genotoxic
mechanism
in
the
induction
of
vascular
tumors
in
mice.

Carbaryl
was
administered
continuously
via
the
diet
to
groups
of
20
male
heterozygous
p53
knockout
mice
at
concentrations
of
0,
10,
30
100,
300,
1,000
and
4,000
ppm
for
at
least
180
days.
No
treatment
related
tumors
were
found
even
in
the
highest
dose
ppm
group.
In
particular,
it
should
be
noted
that
no
tumors
were
found
in
the
liver,
kidney
or
vascular
system.
Under
the
conditions
of
this
study,
the
NOEL
is
4,000
ppm
(approximately
716
mg/
kg
b.
w./
day)
for
neoplastic
changes.

C.
Epidemiological/
Worker
Exposure
Information
Epidemiological
evaluations
of
carbaryl
production
workers,
the
population
of
exposed
persons
with
the
highest
and
most
consistent
carbaryl
exposure,
show
no
indication
of
effects
on
tumor
incidence.

The
two
epidemiology
studies
of
Aventis
CropScience
factory
workers
representing
a
sub­
population
of
the
general
public
with
the
highest
exposure
(i.
e.,
exposure
on
a
daily
basis
over
many
years).
Thus,
the
data
from
these
studies
represent
a
potential
worstcase
for
chronic
carbaryl
exposure
to
humans.
The
results
from
these
studies
show
that
30
the
overall
mortality
experience
of
the
cohort
is
significantly
less
than
expected
when
compared
to
both
United
States
and
West
Virginia
male
death
rates.
In
general,
these
studies
support
conclusion
that
there
were
no
excess
cases
of
cancer
in
this
working
population.

SUMMARY
Aventis
CropScience
strongly
believes
that
the
use
of
carbaryl
products
present
no
imminent
carcinogenic
risk
to
users
based
on
the
following
points:

 
Current
data
support
a
hypothesis
that
high
doses
of
carbaryl
in
life­
time
studies
produce
tumors
via
a
non­
genotoxic
mechanism,
possibly
related
to
altered
metabolism
at
these
high
doses
 
At
doses
less
than
the
MTD,
there
was
an
increased
incidence
of
tumors
only
in
1
site
of
1
sex
of
1
species.

 
The
increased
incidence
of
vascular
tumors
in
the
male
mice
after
two
years
of
administration
is
of
questionable
biological
significance
 
In
the
p53
knockout
mouse
model,
which
was
demonstrated,
to
be
sensitive
to
the
induction
of
vascular
tumors
by
a
genotoxic
reference
compound,
carbaryl
was
found
to
be
negative
 
The
weight­
of­
the
evidence
indicates
that
carbaryl
shows
little
potential
for
genotoxicity
 
Epidemiological
data
on
carbaryl
production
workers
show
no
increase
in
tumor
incidence
31
REFERENCES
1.
Bigot,
D.
1999.
Validation
on
Transgenic
Mice
–
p53
Knockout
Mice
–
to
Predict
Rodent
Carcinogenicity.
Non­
Guideline
Study.
Rhône­
Poulenc
Agro.
Study
No.
SA
97040.
November
10,
1999.
460pp.
MRID
45281802.
2.
Carmichael
NGC,
Debruyne
ELM
and
Bigot­
Lasserre
D.
2000
The
p53
heterozygous
knockout
mouse
as
a
model
for
chemical
carcinogenesis
in
vascular
tissue.
Envi.
Health
Perspective.
108:
61­
65.
3.
Debruyne,
E.
and
Irisarrri,
E.
(1996).
Carbaryl
technical
­
chronic
toxicity
study
in
the
rat
(HWA
Study
No.
656­
139)
and
the
mouse
(HWA
Study
No.
656­
138):
evaluation
of
histological
slides.
Rhône­
Poulenc
agrochimie
report
No.
R&
D/
CRSA/
TOX­
HPA­
4.
Unpublished
report.
MRID
45365503.
4.
Donehower
LA:
The
p53­
deficient
mouse:
a
model
for
basic
and
applied
cancer
studies.
Semin.
Cancer
Biol.,
7,
269­
278
(1996).
5.
Klonne,
D.
R.
1995.
Carbaryl­
mouse
historical
control
data
­
position
paper,
November
1995.
Rhône­
Poulenc.
74
pp.
MRID
45365501.
6.
Marshall,
R.
1996.
Carbaryl:
induction
of
micronuclei
in
the
bone
marrow
of
treated
mice.
CH
Study
No.
198/
89­
1052.
Corning
Hazelton.
Unpublished
report.
MRID
44069301.
7.
McEnaney,
S.
1993.
Study
to
evaluate
the
chromosome
damaging
potential
of
carbaryl
technical
by
its
effects
on
the
bone
marrow
cells
of
treated
rats.
Hazelton
Microtest.
Hazleton
U.
K.
Study
No.
198/
64.
Unpublished
report.
MRID
43039301.
8.
Pastides,
H.
1993.
Standardized
mortality
ratio
analysis
of
employees
exposed
to
carbaryl
at
the
Rhône­
Poulenc
Institute,
West
Virginia
Plant.
Unpublished
report.
MRID
42901501.
9.
Pastides,
H.,
and
M.
Zorn.
1997.
An
evaluation
of
the
mortality
experience
of
carbaryl
unit
employees
at
the
Rhône­
Poulenc
Institute,
West
Virginia
Plant.
Unpublished
report.
MRID
44349901
10.
Sagelsdorff,
P.
1994.
Investigation
of
the
potential
of
protein­
and
DNA­
binding
of
carbaryl.
CIBA­
GEIGY
Limited
Toxicology
Services/
Cell
biology.
Study
No.
CB93/
52,
unpublished
report.
MRID
43282201.
11.
Robinson,
K.
and
Broxup,
B.
2001
A
Developmental
Neurotoxicity
Study
of
Orally
Administered
Carbaryl,
Technical
Grade,
in
the
Rat.
Lab.
I.
D.
Number
97391.
Performed
by
ClinTrials
BioResearch
Ltd.
34
pages.
Submitted
to
EPA
in
July
2001.
12.
Tyl,
R.
W.,
Myers,
C.
B.,
and
Marr,
M.
C.
2001
Two­
Generation
Reproductive
Toxicity
Evaluation
of
Carbaryl
(RPA007744)
Administered
in
the
Feed
to
CD
(Sprague­
Dawley)
Rats.
RTI
I.
D.
Number
65C­
07407­
400.
Performed
by
Research
Triangle
Institute.
906
pages.
Submitted
to
EPA
in
June
2001.
32
Part
II
­
Dietary
and
Water
Exposure/
Risk
Assessment
Line­
by­
Line
Review
of
the
Dietary
and
Water
Exposure
Assessment
of
the
Human
Health
Risk
Assessment
Document
for
Carbaryl
(June
19,
2001)

1.0
Executive
Summary
Page:
3
Paragraph:
5
Line:
3
EPA
Comment:
No
acceptable
two­
year
reproduction
study
is
available.

Aventis
Response:
A
study
has
been
completed
and
submitted.
The
data
gap
to
remove
the
10X
FQPA
safety
factor
has
been
filled.

Page:
6
Paragraph:
2
Line:
1
EPA
Comment:
Monitoring
data
for
carbaryl
residues
in
ground
and
surface
water
are
available
but
they
are
of
limited
utility
in
developing
estimated
environmental
concentrations
for
the
aggregate
dietary
(food
and
water)
risk
assessment.

Aventis
Response:
Monitoring
data
submitted
by
Aventis
is
directly
applicable
to
drinking
water
residue
use
in
the
aggregate
(food
and
water)
assessment.
Worst­
case
community
water
supply
systems,
targeted
for
maximum
carbaryl
use,
were
monitored
for
three
years
on
a
weekly
basis
during
peak
carbaryl
use
times.

4.0
Exposure
Assessment
and
Characterization
4.2.1
Residue
Profile
Page:
25
Paragraph:
2
Line:
1
EPA
Comment:
HED
conducts
dietary
risk
assessments
using
the
DEEM™
which
incorporated
consumption
data
generated
in
USDA's
Continuing
Surveys
of
Food
Intakes
by
Individuals
(CSFII),
1989­
1992.

Aventis
Response:
The
CSFII
data
for
the
years
1994­
1996
have
been
available
for
several
years
now.
These
data
should
be
used
as
it
reflects
the
most
recent
eating
patterns
and
habits
of
the
U.
S.
population
that
is
currently
available.
33
Page:
26
Paragraph:
3
Lines:
1­
5
EPA
Comment:
Most
of
the
processing
factors
were
obtained
from
processing
studies
submitted
by
the
registrant
and
compiled
in
a
memo
entitled
"Carbaryl
Anticipated
Residues
for
Carcinogenic
Dietary
Risk
Assessment."
S.
Hummel,
12/
3/
93.

Aventis
Response:
The
studies
in
this
memo
are
older
processing
studies
submitted
in
the
1980's
or
even
earlier.
New
residue
processing
studies
were
done
in
the
early­
to
midnineties
in
conjunction
with
the
Residue
Chemistry
DCI.
These
studies
are
listed
with
MRID
numbers
in
EPA's
Product
and
Residue
Chemistry
documents.
The
factors
are
listed
in
the
`Discussion
Section'
below.
These
factors
should
be
used
in
the
dietary
risk
assessment
as
they
represent
newer,
GLP
data.

Page:
27
Paragraph:
1
Line:
5
EPA
Comment:
As
discussed
in
section
3.2,
the
10X
safety
factor
is
retained
for
carbaryl.

Aventis
Response:
The
data
gaps
and
concerns
for
the
FQPA
safety
factor
have
been
addressed
by
Aventis
with
the
submission
of
the
2­
generation
rat
reproduction
study
and
the
additional
data
generated
for
the
developmental
neurotoxicity
study.
Based
upon
the
outcome
of
the
studies
provided,
the
10X
safety
factor
is
no
longer
justified
and
should
be
removed.
There
are
no
populations
of
concern
in
this
assessment
when
the
corrected
RfD
is
used
in
the
risk
calculation
(excluding
poultry
and
using
the
CMBS
data).

4.2.2
Acute
Dietary
Exposure
Assessment
Page:
29
Table
6:
EPA
Comment:
The
top
heading
of
the
table
is
labeled
"Acute­
All
Commodities
at
the
99.9
th
percentile
of
exposure
(Market
Basket
Survey
Data
Used
in
Place
of
PDP/
FDA
data.)

Aventis
Response:
This
top
table
heading
should
be
labeled:
"Acute­
All
Commodities
at
the
99.9
th
Percentile
of
Exposure
(Market
Basket
Survey
Data
Not
Included).

4.2.5
Characterization/
Uncertainties
of
the
Risk
Estimates
34
Page:
31
Paragraph:
1
Line:
1
EPA
Comment:
Cooking
factors
were
available
for
potatoes
only.

Aventis
Response:
A
survey
of
the
literature
data
revealed
numerous
cooking
and
washing
studies
for
carbaryl.
See
`Discussion
Section'
for
a
list
of
these
studies
and
a
table
of
processing/
washing/
cooking
factors
that
can
be
derived
for
carbaryl
from
these
literature
studies.

4.3
Water
Exposure/
Risk
Pathway
Page:
31
Paragraph:
5
Line:
4
EPA
comment:
Some
non­
targeted
monitoring
data
are
available
but
they
are
of
limited
utility
in
developing
estimated
environmental
concentrations
(EECs)
for
ecological
and
human
health
risk
assessment.

Aventis'
response:
Aventis
believes
that
the
highest
estimated
EECs
of
relevance
for
ecological
risk
assessment
are
not
relevant
for
estimating
human
health
risks
due
to
the
lack
of
proximity
of
drinking
water
sources
to
likely
areas
of
highest
ecological
risk.
The
drinking
water
monitoring
program
conducted
by
the
registrant
provides
a
real
world
assessment
of
the
potential
for
human
exposure
to
carbaryl
in
drinking
water
derived
from
surface
water.
Drinking
water
concentrations
derived
from
PRZM/
EXAMS
greatly
overestimate
the
potential
exposure
to
carbaryl
in
drinking
water,
generally
by
several
orders
of
magnitude.

Monitoring
Data
Page:
32
Paragraph:
4
Line:
2
(and
elsewhere)
EPA
comment:
USGS
NAQWA
(sic)
program
Aventis'
response:
The
correct
abbreviation
for
the
USGS
water
monitoring
program
is
NAWQA.

Page:
32
Paragraph:
5
Line:
2
EPA
comment:
Because
of
limitation
in
the
analytical
methods
used
there
is
some
question
as
to
the
accuracy
of
carbaryl
analysis.
35
Aventis'
response:
This
generalized
statement
needs
to
be
qualified
or
deleted.
Whereas
the
authors
of
reports
written
as
part
of
the
NAWQA
program
have
been
clear
about
the
potential
limitations
of
the
quantitative
nature
of
the
carbaryl
data
in
the
database,
they
have
also
been
clear
about
the
validity
of
the
qualitative
nature
of
the
data.
The
use
of
the
multi­
residue
method
in
the
NAWQA
program
does
have
some
limitations
as
a
result
of
the
large
numbers
of
diverse
pesticides
and
degradation
products
that
they
are
monitoring.
However,
the
QC/
QA
data
generated
as
part
of
the
NAWQA
program
(described
in
the
discussion
section
on
surface
water
at
the
end
of
the
EFED
response)
demonstrates
the
validity
of
the
detections
of
carbaryl
in
the
studies.
The
monitoring
study
conducted
by
the
registrant,
and
reported
in
this
section,
does
not
have
the
same
potential
limitations
in
the
analytical
method
since
the
method
is
looking
specifically
for
only
carbaryl.
Therefore,
the
analytical
method
used
by
the
registrant
does
not
raise
questions
about
the
accuracy
of
the
carbaryl
analysis.

Page:
32
Paragraph:
5
Line:
3
EPA
comment:
Poor
analytical
methods
probably
have
resulted
in
lower
detection
rates
and
lower
concentrations
than
actually
present.

Aventis'
response:
This
generalized
statement
should
be
deleted
for
reasons
provided
above
and
in
the
discussion
section
of
the
EFED
response.

Page:
33
Paragraph:
2
Line:
6
EPA
comment:
The
data
do
not
give
a
good
indication
of
the
effectiveness
of
treatment
because
samples
existing
and
entering
the
treatment
plant
were
different.
In
several
cases,
finished
water
had
higher
concentrations
than
raw
water
and
finished
water
had
detectable
carbaryl
when
the
raw
did
not.

Aventis'
response:
Obtaining
matched
samples
is
not
possible
because
of
varying
residence
time
through
the
treatment
system,
as
well
as
mixing
that
occurs.
The
minimal
amount
of
carbaryl
residues
found
also
made
observations
regarding
treatment
less
definitive.
However,
an
examination
of
the
data
does
indicate
that
carbaryl
concentrations
were
generally
lower
in
finished
water
than
in
raw
water
samples
collected
at
the
same
time.
These
data
do
include
several
instances
where
carbaryl
was
detected
over
a
few
weeks
in
the
raw
water
so
the
effect
of
sample
time
is
less
important.
The
statement
regarding
several
cases
of
finished
water
having
higher
concentrations
than
raw
water
is
misleading
and
certainly
does
not
consider
the
analytical
uncertainty
for
concentrations
below
the
level
of
quantification
and
near
the
level
of
detection.
There
were
only
two
cases
when
finished
water
was
greater
than
raw
water
when
the
concentrations
in
finished
water
were
greater
than
0.01
ppb
(only
one­
third
of
the
quantification
limit).
One
36
case
was
when
the
raw
water
was
0.009
ppb
and
the
finished
water
was
0.011
ppb.
These
two
analyses
are
essentially
equivalent,
especially
considering
that
they
are
only
about
a
third
of
the
quantification
limit.
The
other
case
was
at
the
Deerfield
community
water
system.
This
drinking
water
facility
uses
a
small
river
without
a
reservoir
as
a
source
for
a
small
Community
Water
System.
Farms
are
located
immediately
upstream
of
the
facility.
The
intake
is
also
not
continuous
(shut
down
over
weekends).
Therefore,
getting
a
matching
sample
is
quite
difficult,
especially
for
a
short
duration
spike
as
a
result
of
spray
drift,
summer
thunderstorm,
or
perhaps
a
spill
that
almost
immediately
enters
the
river
as
a
runoff
event.
The
rarity
of
this
event
is
demonstrated
by
the
absence
of
residues
of
this
magnitude
the
next
year
(2000).
Samples
collected
through
this
time
of
the
year
in
2001
also
do
not
indicate
a
similar
event.
Although
the
data
from
this
site
cannot
be
used
to
determine
the
peak
concentration,
the
data
provide
a
distribution
of
residues
through
the
three­
year
period
which
will
define
up
to
the
99
th
percentile
concentration
of
the
distribution.

The
Deerfield,
Michigan
community
water
system
is
one
of
the
systems
in
which
the
greatest
variability
of
residues
would
be
expected.
Most
of
the
other
community
water
systems
are
located
on
larger
rivers,
lakes,
or
reservoirs.

Because
the
design
of
study
called
for
analysis
of
finished
water
only
when
there
were
residues
in
the
raw
water,
there
was
only
one
finished
sample
analyzed
when
the
raw
water
contained
no
residues.
This
sample
was
collected
at
the
Deerfield
community
water
system
at
the
sampling
interval
after
the
finding
of
0.16
ppb
in
the
Deerfield
system.
The
residue
level
in
this
sample
was
0.004
ppb.
The
difference
between
0.004
ppb
and
non­
detect
is
insignificant,
and
if
real
can
probably
be
attributed
to
water
at
much
higher
concentrations
remaining
in
the
system
from
the
previous
week.

Page:
33
Paragraph:
2
Line:
10
EPA
comment:
This
illustrates
that
carbaryl
contamination
is
transient,
and
that
it
is
unlikely
that
any
sampling
would
catch
the
actual
peak
concentration
Aventis'
response:
The
role
of
a
peak
concentration
is
in
dietary
exposure
assessment
is
undergoing
re­
examination
within
EPA.
The
current
policy
of
EPA
appears
to
define
a
certain
percentile
as
an
appropriate
value
for
use
in
screening
assessments,
but
the
exact
percentile
to
be
used
is
being
defined
by
EPA
management.
For
more
comprehensive
assessments,
a
distribution
of
values
is
preferred.

The
peak
concentration
in
this
study
was
measured
at
a
community
water
system
on
a
small
river.
The
registrant
agrees
that
the
sampling
schedule
was
not
adequate
to
determine
the
true
peak
in
such
systems.
Most
of
the
other
community
water
systems
are
located
on
larger
rivers,
lakes,
or
reservoirs.
Therefore,
the
peak
values
are
not
likely
to
be
an
order
of
magnitude
greater
than
37
the
amounts
detected
in
this
monitoring
program.

Page:
33
Paragraph:
2
Line:
13
EPA
comment:
Non­
targeted
monitoring,
such
as
the
NAWQA
program
has
shown
that
much
higher
concentrations
occur.

Aventis'
response:
The
main
reason
why
the
drinking
water
monitoring
study
did
not
show
residues
as
high
as
in
the
NAWQA
program
is
the
location
of
the
sampling
points.
Drinking
water
supplies
tend
to
be
located
on
larger
surface
water
bodies
than
NAWQA
sampling
points
(or
in
other
words,
the
intakes
for
community
water
systems
tend
to
be
downstream
of
NAWQA
sampling
points).
This
additional
time
allows
for
additional
degradation
and
dilution
to
occur.
Finding
the
highest
concentration
at
the
Deerfield,
Michigan
system
is
not
surprising
since
this
intake
is
on
one
of
the
smallest
surface
water
bodies
included
in
the
monitoring
study.

Page:
33
Paragraph:
2
Line:
14
EPA
comment:
This
study,
while
useful,
does
not
provide
sufficient
information
to
allow
estimation
of
actual
peak
and
mean
concentrations
that
actually
occur
in
all
areas
or
the
effect
of
treatment.

Aventis'
response:
Because
most
of
the
samples
did
not
contain
carbaryl
residues,
accurate
estimates
of
the
actual
peak
and
mean
concentrations
can
not
be
obtained.
However,
the
distributions
obtained
from
all
sites
can
be
used
to
define
up
to
the
99
th
percentile
concentration.
The
average
cannot
be
accurately
determined;
however,
the
timeweighted
average
is
only
slightly
above
the
limit
of
detection
(and
certainly
less
than
0.01
ppb)
at
all
20
sites.
The
study
provides
information
on
concentrations
of
carbaryl
in
community
water
systems
most
likely
to
contain
the
highest
concentrations
of
carbaryl.
Residues
of
carbaryl
in
other
areas
would
be
expected
to
be
lower.
The
study
does
not
provide
information
on
concentrations
in
smaller
surface
water
bodies
or
in
areas
where
surface
water
is
not
used
for
drinking
water.
Because
drinking
water
concentrations
are
what
is
needed
for
FQPA
dietary
calculations,
this
information
is
suitable
for
use
in
dietary
exposure
assessments.

Page:
34
Table
8
EPA
comment:
Carbaryl
EEC
Values
Aventis'
response:
Aventis
has
provided
a
more
detailed
response
to
this
same
table
that
is
presented
as
Table
6
in
the
draft
EFED
chapter.
An
electronic
copy
of
the
EPA
Memorandum
on
"Refined
Estimated
Environmental
Concentrations
for
38
Carbaryl"
(DP
Bar
Code
D267276,
authored
by
E.
Laurence
Libelo,
July
23,
2001,
and
sent
to
Anthony
E.
Britten
and
Virginia
Dobozy)
was
provided
to
Aventis
and
contains
the
PRZM
model
inputs
that
were
used
to
generate
the
EECs
shown
in
this
table.
An
abbreviated
version
of
our
response
to
the
information
from
EFED
is
included
below.

It
would
be
useful
to
add
another
column
to
Table
8
to
specify
which
method
of
application
was
used
to
generate
the
EECs.
It
would
also
be
of
benefit
for
the
Agency
to
state
which
of
the
carbaryl
labels
were
used
to
develop
the
"maximum"
label
application
rate
scenarios.

The
model
parameters
listed
in
the
Memorandum
show
that
the
"average"
scenarios
for
citrus
and
apples
were
conducted
using
aerial
applications.
Few
applications
to
these
crops
are
made
aerially.
Therefore,
the
model
results
overestimate
the
contributions
from
spray
drift
since
the
"average"
applications
to
these
crops
are
made
using
ground
airblast
equipment
with
lower
spray
drift
inputs.

The
"maximum
label
rate"
application
scenario
for
apples
that
is
allowed
by
the
Sevin
brand
XLR
PLUS
label
(E.
P.
A.
Reg.
No
264­
333),
the
Sevin
brand
80WSP
and
CHIPCO
Sevin
brand
80WSP
labels
(E.
P.
A.
Reg.
No
264­
526)
and
the
CHIPCO
Sevin
brand
SL
label
(E.
P.
A.
Reg.
No
264­
335)
is
5
applications
at
3­
lb.
ai/
A/
application
made
every
14
days.
The
scenario
used
in
the
model
applies
less
than
the
maximum
amount
of
product
allowed
by
the
labels.
In
addition,
application
timing
was
used
in
the
modeling
for
the
Index
Reservoir
scenario
(applications
made
by
air
every
4
days)
that
would
be
a
violation
of
the
Aventis
labels
that
restrict
applications
to
a
minimum
of
every
14
days.

The
"average"
scenario
for
sweet
corn
in
Ohio
should
be
3
applications
at
1.1
lb
ai/
A/
application
(as
noted
in
the
memo,
"Average
application
rate
from
Quantitative
Usage
Analysis
for
Carbaryl,
prepared
July
21,
1998
by
Frank
Hernandez,
OPP/
BEAD")
and
not
the
2
applications
at
3.4
lb
ai/
A/
application
as
listed
in
the
table.
The
PRZM
input
file
shows
the
correct
inputs
of
3
applications
at
1.1­
lb
ai/
A/
application.
Likewise,
the
"average"
scenario
for
sugar
beets
in
Minnesota
should
be
1
application
at
1.3­
lb
ai/
A/
application.
The
"Citrus"
scenario
would
be
more
appropriately
labeled
Oranges.
For
the
average
scenario,
the
3.4
lb.
ai/
A/
application
rate
listed
in
Table
8
is
for
oranges
which
is
the
highest
"average"
application
rate
for
any
type
of
citrus.
Therefore,
this
"average"
scenario
for
oranges
is
at
the
high
end
for
all
citrus
and
overestimates
the
EECs
for
use
in
the
other
citrus
crops.

4.3.1
DWLOCs
for
Acute
Dietary
Exposure.
39
Page:
35
Paragraph:
1
Line:
7
EPA
Comment:
Therefore,
there
is
no
allowable
contribution
for
water
to
the
risk
cup.

Aventis
Response:
Based
on
a
corrected
acute
reference
dose
for
carbaryl
(no
10X
FQPA
safety
factor)
and
the
exposure
assessment
that
uses
the
Carbamate
Market
Basket
Data,
DWLOCs
can
be
calculated.
(see
`Discussion
Section'
for
calculations.)

4.3.2
DWLOCs
for
Chronic
Dietary
Exposure
Page
36
Table
9
EPA
Comment:
DWLOC(
chronic)
calculations
in
Table.

Aventis
Response:
Based
on
a
corrected
chronic
reference
dose
for
carbaryl
(no
10X
FQPA
safety
factor)
the
DWLOCs
should
be
calculated
as
shown
in
the
`Discussion
Section'.
Based
on
the
corrected
values,
there
are
no
population
subgroups
of
concern.

8.0
Data
Needs/
Label
Requirements
Product
Chemistry
Data
Gaps
Page:
97
Paragraph:
2
Line
1
EPA
Comment:
Additional
data
are
required
depicting
carbaryl
residue
in/
on
cotton
gin
byproducts.

Aventis
Response:
Aventis
is
not
supporting
the
cotton
use
and
has
removed
this
crop
from
the
labels.
Aventis
CropScience
requested
cancellation
of
this
use
in
a
letter
to
Mr.
George
Tompkins,
RD,
on
January
22,
1999.
A
Federal
Register
Notice
dated
April
14,
1999
announced
receipt
by
the
Agency
of
an
application
from
RhonePoulenc
Ag
Company
(now
Aventis
CropScience)
to
cancel
the
use
of
carbaryl
products
on
cotton
40
Line­
by­
Line
Review
of
the
Supporting
Document
"Revised
Dietary
Exposure
Analysis
for
the
HED
Revised
Human
Health
Risk
Assessment
(Felicia
A.
Fort;
April
26,
2001)"

Conclusions/
Summary
Page:
2
Paragraph:
1
Line:
8
EPA
comment:
At
the
present
time,
information
from
the
industry­
sponsored
Carbamate
Market
Basket
Survey
has
not
been
completely
approved
for
use
in
dietary
risk
assessments.

Aventis
Response:
The
CMBS
protocols
were
presented
to
EPA
for
comment
before
the
studies
were
started.
These
studies
were
specifically
conducted
to
obtain
more
realistic
residues
for
consumer's
"at
the
plate"
to
be
used
in
dietary
assessments.
As
agreed
with
EPA
before
this
study
was
started,
the
CMBS
data
should
take
precedence
over
others
for
use
in
dietary
assessment
(after
their
quality
review
has
been
completed).

Chronic
Page:
2
Paragraph:
2
Line:
8
EPA
comment:
When
poultry
is
not
considered
in
the
risk
estimate,
the
dietary
exposure
is
<
6%
of
the
cPAD
for
all
population
subgroups.

Aventis
Response:
When
a
corrected
cRfD
is
used
(no
FQPA
10X)
and
poultry
is
not
included
in
the
risk
estimate,
the
dietary
exposure
is
<
0.6%
of
the
cRfD
for
all
population
subgroups.

Acute
Page:
3
Continuation
from
Paragraph:
4
of
Page
2
Line:
3
EPA
Comment:
Subsequent
dietary
analyses
and
all
additional
sensitivity
analyses
were
conducted
without
poultry;
the
risk
estimates
were
still
of
concern
for
all
population
subgroups
with
the
all
infants
population
subgroup
consuming
260%
of
the
aPAD
when
CMBS
data
were
used.

Aventis
Response:
When
a
corrected
aRfD
is
used
(no
FQPA
10X)
,
there
are
no
concerns
for
all
population
subgroups.
The
all
infants
population
subgroup
consumes
26%
of
the
aRfD
when
CMBS
data
are
used.
41
Residue
Data
Page:
5;
Paragraph:
2;
Line:
4
EPA
comment:
FDA
monitoring
data
were
used
for
…
cherries,
raspberry,
blueberry,
raspberry,
asparagus,
…
Aventis
response:
Second
listing
of
"raspberry"
can
be
deleted.

Page:
5;
Paragraph:
2;
Line:
9
EPA
comment:
Field
trial
data
were
used
for
the
commodities,
garden
beets,
turnips,
mustards,
dried
beans,
almonds,
pecans,
walnuts,
field
corn
grain,
rice,
flax
seed,
okra,
olive,
peanuts,
pistachio,
and
sunflower.

Aventis
response:
According
to
Table
6a,
the
following
commodities
should
be
added
to
this
list:
sugar
beets
and
dried
peas
Page:
5;
Paragraph:
2;
Line:
11
EPA
comment:
For
oysters,
the
tolerance
of
2
ppm
was
used
in
the
assessment.

Aventis
response:
According
to
Table
6a,
the
tolerance
of
0.2
ppm
was
also
used
for
Dill
(fresh)
in
the
assessment.

Processing
Factors
Page
6
Paragraph:
1
Table
2
EPA
comment:
Most
of
the
carbaryl
processing
factors
were
obtained
from
processing
studies
submitted
by
the
registrant
and
compiled
in
a
memo
entitled
"Carbaryl
Anticipated
Residues
for
Carcinogenic
Dietary
Risk
Assessment",
S.
Hummel,
12/
3/
93.

Aventis
Response:
As
previously
state,
the
studies
in
this
memo
were
older
processing
studies
submitted
in
the
1980's
or
even
earlier.
New
residue
processing
studies
were
done
in
the
early­
to
mid­
nineties
in
conjunction
with
the
Residue
Chemistry
DCI.
These
studies
are
listed
with
MRID
numbers
in
EPA's
Product
and
Residue
Chemistry
documents.
The
factors
are
listed
in
the
Discussion
section.
These
factors
should
be
used
in
the
dietary
risk
assessment
as
they
represent
newer,
GLP
data.
42
Consumption
data
Page:
7
Paragraph:
1
Line:
1
EPA
Comment:
HED
conducts
dietary
risk
assessments
using
the
DEEM™
which
incorporated
consumption
data
generated
in
USDA's
Continuing
Surveys
of
Food
Intakes
by
Individuals
(CSFII),
1989­
1992.

Aventis
response:
As
previously
stated
the
CSFII
data
for
the
years
1994­
1996
have
been
available
for
several
years
now.
These
data
should
be
used
as
it
reflects
the
most
recent
eating
patterns
and
habits
of
the
U.
S.
population
that
is
currently
available.

Results
Page:
7
Paragraph:
2
Line
2
EPA
comment:
Additional
cooking
and
processing
studies
would
allow
further
refinement.

Aventis
response:
A
literature
search
resulted
in
many
literature
references
to
cooking,
washing
and
peeling
studies
for
carbaryl
treated
commodities.
Average
cooking,
washing
factors
can
be
derived
from
these
literature
studies.
See
the
`Discussion
Section'
for
a
list
of
the
references
and
the
average
calculated
factors.

Characterization/
Uncertainties
of
the
Risk
Estimates.

Page:
9
Paragraph:
2
Line:
1
EPA
comment:
Additional
cooking
factors
could
further
reduce
the
risk
estimates.

Aventis
response:
Numerous
literature
studies
have
been
conducted
on
other
commodities,
see
the
`Discussion
Section'.

Attachment
1:
Anticipated
Residues
Summary
and
Residue
Distribution
Files
Table
6a:
Summary
of
Anticipated
Residues
for
Carbaryl
(Market
Basket
Data
not
included)
and
Table
6b:
Summary
of
Market
Basket
Survey
Data.

Aventis
Comments:
EPA/
BEAD
reports
a
maximum
of
84%
crop
treated
for
cranberries
and
an
average
of
39%
crop
treated
for
cranberries.
Three
years
of
data
were
obtained
from
the
Cranberry
Institute
(memo
from
Gary
Deziel,
Manager
of
Research
and
Communication,
Cranberry
Institute.)
These
data
show
maximum
acres
treated
43
for
1992,
39%;
1996,
37%;
and
1998,
36%.
Based
on
these
data
and
private
conversations,
the
maximum
percent
crop
treated
for
cranberries
is
39%
with
an
average
of
37%.

Barley,
oats,
rye
and
cotton/
cottonseed
are
not
supported
by
Aventis
and
should
not
be
included
in
the
risk
assessment.
A
letter
date
May
7,
1999
to
Ms.
Kathryn
Boyle
confirmed
Rhône­
Poulenc's
(now
Aventis
CropScience)
decision
not
to
support
the
registration
of
carbaryl
products
on
the
commodities
barley,
oats,
and
rye.
These
uses
have
been
deleted
from
our
technical
and
end­
use
product
labels.

Residue
Data
Sources:
The
following
data
sources
are
more
appropriate
to
use
for
the
crops
listed
in
the
table,
according
to
priority
rules
for
use
of
residue
data
(Market
Basket
(MBS)>
PDP>
FDA>
Field
Trial
(FT)>
tolerance)
and
the
Translation
of
Monitoring
Data
HED
SOP
99.3
(March
26,
1999).

Food
EPA
Used
More
Appropriate
Source
Brussels
Sprouts
Cabbage
FDA
Lettuce
MBS
Cabbage
Cabbage
FDA
Lettuce
MBS
Collards
Mustard
FT
Spinach
PDP
Eggplant
Pepper
FDA
Tomato
MBS
Endive
Leaf
lettuce
FDA
Spinach
PDP
Kale
Mustard
FT
Spinach
PDP
Kohlrabi
Cabbage
FDA
Broccoli
MBS
Mustard
Greens
Mustard
FT
Spinach
PDP
Paprika
H.
Pepper
FDA
Tomato
MBS
Chili
Pepper
H.
Pepper
FDA
Tomato
MBS
Other
Pepper
H.
Pepper
FDA
Tomato
MBS
Sweet
Pepper
S.
Pepper
FDA
Tomato
MBS
Pimentos
H.
Pepper
FDA
Tomato
MBS
Strawberries
Strawberry
FDA
Strawberry
PDP
Strawb.
Juice
Strawberry
FDA
Strawberry
PDP
Swiss
Chard
Celery
PDP
Spinach
PDP
Page
41
Residue
Distribution
Files
RDF#
68
Almonds
and
RDF#
69
Chestnuts
appear
to
have
used
the
almond
hull
field
trial
data
rather
than
the
almond
nutmeat
field
trial
(mostly
NDs)
data
for
the
residue
values.

Page
44
Residue
Distribution
Files
RDF#
86
to
RDF#
92.
Meat
and
milk.
There
is
no
clear
indication
or
discussion
anywhere
in
the
document
regarding
the
source
of
these
residue
data
or
how
the
values
were
derived
(e.
g.,
derived
from
theoretical
animal
diets
or
monitoring
data).
(see
Discussion
below).
44
Part
II
­
Dietary
and
Water
Exposure/
Risk
Assessment
SUPPORTING
DISCUSSIONS
Surface
Water
Concentrations
In
section
5.0,
page
33
(Table
6),
EPA
has
based
its
assumptions
about
concentrations
of
carbaryl
in
drinking
water
upon
model
simulations.
The
data
from
the
registrant
drinking
water
monitoring
program
provide
the
best
estimate
of
concentrations
of
carbaryl
in
drinking
water.
This
study
uses
the
sampling
design
for
acute
endpoints
recommended
in
industry/
EPA
meetings
during
1999
(weekly
sampling
during
times
of
peak
concentrations
over
a
three­
year
period).
Twenty
sites
representing
the
highest
carbaryl
use
areas
were
selected
based
on
the
information
provided
in
Appendix
I.
Included
are
16
sites
in
agricultural
areas
and
4
locations
in
urban
areas.
Samples
were
collected
from
the
inlet
and
outlet
water
at
each
sampling
interval.
Outlet
samples
were
only
analyzed
when
residues
were
present
in
the
inlet
samples.
The
analytical
method
had
a
limit
of
quantification
of
0.030
ppb
and
a
limit
of
detection
of
0.002
ppb.

Error!
Reference
source
not
found.
summarizes
the
results
of
the
monitoring
at
each
of
the
20
community
water
systems.
The
maximum
concentration
observed
was
0.16
ppb
(average
of
four
samples,
the
highest
was
0.18
ppb)
in
a
finished
water
sample
from
the
Deerfield
community
water
system
located
on
the
River
Raisin
in
Lenawee
County,
Michigan.
There
were
only
five
other
samples
above
the
limit
of
quantification
of
0.030
ppb.
One
was
a
raw
water
sample
containing
0.31
ppb
from
the
Little
Potato
Slough
Mutual
community
water
system
near
Lodi
in
San
Joaquin
County,
California
(the
source
is
the
Little
Potato
Slough).
The
corresponding
finished
water
sample
was
0.007
ppb.
A
second
one
was
a
raw
water
sample
in
Brockton,
MA
which
contained
0.031
ppb.
No
detectable
residues
were
found
in
the
corresponding
finish
water
sample.
The
last
three
samples
were
from
the
Shades
Mountain
plant
of
the
Birmingham
community
water
system
on
the
Cahaba
River
in
Jefferson
County,
Alabama
.
Two
were
raw
and
finished
samples
of
0.038
and
0.032
ppb
at
the
same
sampling
interval
in
2001.
The
other
sample
was
0.035
ppb
in
the
raw
water
in
a
2000
sample
(the
corresponding
finished
sample
did
not
contain
carbaryl
residues.
All
residues
were
transient
so
the
time
weighted
average
concentration
of
carbaryl
in
each
of
the
years
was
0.005
ppb
or
less
at
all
20
community
water
systems.
45
Table
1.Summary
of
results
from
the
carbaryl
drinking
water
monitoring
study.

Site
Major
Uses
Maximum
Concentration
(ppt)
TWA
Conc.
(ppt)*
in
Outlet
Water
Inlet
Water
Outlet
Water
1999
2000
2001**
1999
2000
2001**
1999
2000
Manatee,
FL
citrus
9
3
ND
11
ND
NA
1
1
West
Sacramento,
CA
orchards,
nuts
3
24
ND
3
10
NA
1
1
Lodi,
CA
orchards,
nuts
12
31
ND
4
7
NA
1
1
Riverside,
CA
grapes,
tree
crops
8ND
ND
ND
NANA
1
1
Lake
Elsinore,
CA
citrus
ND
3
6
NA
NA
Analysis
Pending
1
1
Corona,
CA
citrus
ND
ND
ND
NA
NA
NA
1
1
Beaumont,
TX
various
agricultural
ND
ND
ND
NA
NA
NA
1
1
Point
Comfort,
TX
rice,
tree
crops
18
5ND
ND
NDNA
1
1
Penn
Yan,
NY
grapes,
apples
ND
23
ND
NA
ND
NA
1
1
Westfield,
NY
grapes,
apples
21
5
ND
ND
9
NA
1
1
Jefferson,
OR
vegetables,
strawberries
ND
10
ND
NA
ND
NA
1
1
Coweta,
OK
pecans
4
ND
***
ND
NA
***
1
1
Pasco,
WA
apples,
potatoes
2
3
ND
ND
ND
NA
1
1
Manson,
WA
apples
ND
ND
ND
NA
NA
NA
1
1
Deerfield,
MI
vegetables
10
4
ND
160
ND
NA
5
1
Brockton,
MA
cranberries
31
27
ND
ND
3
NA
1
1
East
Point,
GA
home
and
garden
18
18
4
3
8
ND
1
1
Midlothian,
TX
home
and
garden
14
ND
14
ND
NA
ND
1
1
Cary,
NC
home
and
garden
4ND
ND
ND
NANA
1
1
Birmingham,
AL
home
and
garden
23
35
38
ND
ND
32
1
1
*
Annual
Time
Weighted
Concentration,
outlet
values
substituted
for
inlet
values
when
available;
values
below
the
detection
limit
were
considered
to
be
half
the
detection
limit.
**
Results
represent
one
to
six
months
of
sampling
into
the
third
year
program.
***
No
results
available
for
the
third
year
of
sampling.
46
ND
Not
detected.
NA
No
outlet
samples
analyzed
due
to
carbaryl
residues
not
being
detected
in
inlet
samples.

Appropriateness
of
the
Carbamate
Market
Basket
Survey
Data
The
Agency
questions
the
appropriateness
of
the
use
of
these
data
in
several
discussions
of
the
dietary
risk
assessment.
The
CMBS
Task
Force
met
with
EPA/
HED
before
the
start
of
the
study
to
review
the
purpose
of
the
study
and
protocols
for
the
study.
The
EPA
assured
the
task
force
that
the
study
data
would
be
used
in
risk
assessments,
taking
precedence
over
all
other
monitoring
or
field
trial
data
available
for
those
crops.
A
surrogation
plan
according
to
HED
SOP
99.3
was
also
assured.
The
specific
target
of
this
study
was
to
measure
more
realistic
residues
that
consumer's
are
exposed
to
"at
the
plate".
The
study
protocol
was
designed
to
mimic
the
typical
consumer
shopping
at
his/
her
local
grocery
store
and
the
preparation
he
would
do
at
home
before
consuming
or
further
cooking/
preparing
the
food.
The
PDP
data
is
taken
from
distribution
centers
before
the
grocery
stores
are
reached
and
some
typical
preparation
is
done.
The
CMBS
data
would
therefore
be
expected
to
have
somewhat
lower
and
more
realistic
measures
of
residues
consumers
are
exposed
to.
There
is
likely
to
be
even
further
reduction
of
residues
before
the
food
is
"put
on
the
plate"
by
storing,
cooking,
slicing,
peeling
or
other
preparation.
It
would
be
very
difficult
to
justify
within
the
industry,
further
generation
of
these
much
needed
data
for
other
crops/
compounds
if
EPA
decided
at
this
late
date
that
the
data
could
not
be
used.

Secondary
Residues
Although
residue
distribution
files
are
presented
by
the
Agency
in
Attachment
1
of
the
Dietary
Assessment
Support
Document
for
meat
and
milk
commodities,
there
is
no
place
in
the
document
where
these
residue
numbers
and
residue
distribution
files
are
discussed.
Aventis
neither
can
determine
the
source
of
the
residue
data
was
(monitoring
or
theoretical
animal
diets)
nor
how
the
data
were
translated
or
calculated.
We
request
that
a
more
detailed
description
of
this
process
be
provided
to
us.

Processing
Factors
EPA
uses
processing
factors
listed
in
a
1993
memo
by
S.
Hummel.
Examination
of
this
document
shows
that
the
source
of
these
processing
factors
is
studies
that
were
conducted
in
the
1980s
or
even
earlier.
As
part
of
the
Residue
Chemistry
DCI
for
carbaryl
in
the
early
1990s,
numerous
residue­
processing
studies
were
conducted
along
with
residue
(RAC)
field
trials.
These
studies
are
referenced
in
the
Product
and
Residue
Chemistry
Support
Document
prepared
by
the
EPA.
These
processing
studies
and
the
derived
factors,
which
were
conducted
under
GLP
conditions
to
more
recent
specifications,
are
the
factors
that
should
be
used
in
the
current
dietary
risk
assessment.
A
table
of
these
factors
is
listed
here.
47
Food
Process
Average
PF
Citrus
Dried
Pulp
1.46
Juice
0.
06
Oil
13.92
Molasses
0.47
Field
Corn
Small
grits
0.25
Meal
0.25
Flour
0.25
Starch
0.25
Crude
Oil
3.
38
Refined
Oil
0.
25
Grapes
Pasteurized
Juice
0.
24
Wet
Pomace
1.37
Dry
Pomace
3.85
Processed
Raisins
1.
37
Unprocessed
Raisins
2.
17
Raisin
Waste
4.
88
Olives
Olive
oil
0.
81
Peanuts
Meal
0.29
Refined
Oil
0.
29
Pome
Juice
0.
40
Wet
Pomace
1.26
Dry
Pomace
3.70
Potatoes
Wet
Peel
1.00
Dry
Peel
0.75
Flakes
0.75
Chips
0.
75
Rice
Polished
Rice
0.03
Hulls
2.37
Bran
0.36
Soybeans
Hulls
0.35
Meal
0.22
Crude
Oil
2.
71
Refined
Oil
0.
005
Sunflowers
Hulls
0.35
Meal
0.03
Crude
Oil
0.
18
Refined
Oil
0.
03
Tomatoes
Juice
0.
52
Wet
Pomace
1.74
Dry
Pomace
2.89
Puree
1.
26
Paste
2.
01
Wheat
Middlings
0.42
Shorts
0.83
Asp
Grain
Fractions
11.79
LG
Flour
0.08
Patent
Flour
0.10
Wheat
Germ
0.65
Wheat
Bran
1.03
48
Cooking
Washing
Factors
EPA
states
several
times
that
cooking/
washing
studies
would
further
reduce
the
anticipated
residues
used
in
the
assessment
and
reduce
the
risk
estimates.
Aventis
conducted
a
general
literature
search
for
studies
such
as
this.
A
table
of
factors
is
listed
followed
by
the
references
found
in
the
general
literature.
Average
cooking,
washing
and
canning
factors
could
be
derived
from
these
references
and
applied
to
the
risk
assessment
as
supported
by
these
literature
studies.

Food
Process
Reduction
Reference
Broccoli
Cooking/
washing
55%
8
Cabbage
Heads
Cooking
90%
2
Cabbage
Heads
Washing
75%
2
Cauliflower
Cooking/
washing
94%
4
Grapes
Washing
49%;
85%
7
Green
Beans
Canning
100%
11
Green
Beans
Cooking/
blanching
81%
11
Green
Beans
Washing
52%
11
Okra
Cooking
42%;
25%
1,14
Okra
Cooking/
steaming
82%
1,14
Okra
Washing
80%;
66%;
70%
1,14
Onions
Washing
89%;
98%;
100%
9
Orchard
Fruit
Washing
50%
12
Peas
Cooking/
boiling
85%
3
Peas
Washing
70%
3
Spinach
Canning
99.5%
10
Spinach
Washing
70%
10
Tomatoes
Peeling/
washing
99%
5,6
Tomatoes
Puree/
catsup
98%
5,6
Tomatoes
Washing
66%;
68%,
84%
5,6
1.
Indian
Journal
of
Plant
Protection.
1996,
24,
86­
89.
2.
Pest
Management
and
Econ.
Zoology.
1994,
2,
131­
134.
3.
Plant
Protection
Bulletin.
1988,
40,
12­
13.
4.
Beitrage
zur
Trop.
Land.
Veter.
1982,
20,
89­
95.
5.
Indian
Journal
of
Entomology.
1978,
40,
187­
190.
6.
Indian
Journal
of
Entomology.
1973,
34,
31­
34.
7.
Indian
Journal
of
Ag
Sciences.
1978,
48,
179­
183.
8.
J.
Ag.
Food
Chem.
1969,
15,
215­
216.
9.
J.
Food
Science
Technology.
1978,
15,
215­
216.
10.
J.
Ag.
Food
Chem.
1968,
16,
967­
973.
11.
J.
Ag.
Food
Chem.
1968,
16,
962­
966.
12.
J.
Assoc.
Off.
Anal.
Che,.
1989,
72,
533­
535.
13.
Env.
Health
Criteria
1994,
153,
358pp.
49
14.
Indian
Journal
of
Ag
Sciences.
1976,
45,
139­
144.
15.
Indian
Journal
of
Env.
Health
Acute
DWLOC
EPA
indicates
there
is
no
room
in
the
risk
cup
for
water
and
DWLOCs
cannot
be
calculated.
If
a
corrected
acute
RfD
is
used
(no
FQPA
10X),
acute
DWLOCs
can
be
calculated
as
follows:

US
Population:
Acute
RfD=
0.03
mg/
kg
Exposure=
0.005937
mg/
kg
(from
EPA
assessment
without
poultry
and
with
CMBS
data)
70
kg.
Man,
2­
liter
consumption
Acute
DWLOC
is
842ppb
Children
1­
6:
Acute
RfD=
0.03
mg/
kg
Exposure=
0.008363
mg/
kg
(from
EPA
assessment
without
poultry
and
with
CMBS
data)
10kg
child,
1
liter
consumption
Acute
DWLOC
is
216ppb
Aventis
monitoring
programs
of
raw
surface
drinking
water
have
shown
0.16
ppb
as
the
maximum
encountered.
NAWQA
data
shows
5
ppb
as
the
maximum
encountered
in
ground
or
surface
water.
These
monitoring
numbers
would
indicate
there
is
no
concern
for
dietary
and
drinking
water
exposure
in
an
aggregate
situation.

Chronic
DWLOCs
EPA
indicates
there
is
some
concern
for
subpopulations
for
drinking
water
in
the
chronic
scenarios.
If
a
corrected
chronic
RfD
is
used
(no
FQPA
10X)
chronic
DWLOCs
can
be
calculated
as
follows:

US
Population:
Chronic
RfD=
0.01
mg/
kg
Exposure
=
0.000037
mg/
kg
(from
EPA
assessment
without
poultry)
70kg
man,
2
liter
consumption
Chronic
DWLOC
is
349
ppb
Children
1­
6:
Chronic
RfD=
0.01
mg/
kg
Exposure=
0.000062
mg/
kg
(from
EPA
assessment
without
poultry)
10kg
child,
1
liter
consumption
Chronic
DWLOC
is
99
ppb
50
Aventis
and
NAWQA
monitoring
data
averages
are
significantly
less
than
the
maximum
values
of
0.16
ppb
and
5
ppb
numbers
cited
above.
There
are
no
populations
of
concern
for
aggregate
exposure
to
food
and
water
for
carbaryl.
51
Part
III
­
Residue
Chemistry
Considerations
Line­
by­
Line
Review
of
the
Residue
Chemistry
Information
Included
in
the
Human
Health
Risk
Assessment
Document
for
Carbaryl
(June
19,
2001)

1.08
Data
Needs/
Label
Requirements
Product
Chemistry
Data
Gaps
Page
95
Paragraph
3
Line
1­
3
EPA
comment:
A
review
of
the
labels
and
supporting
residue
data
indicate
that
several
label
amendments
are
required.
Details
are
provided
in
the
Product
and
Residue
Chemistry
Chapters
(DP
Barcode:
D238151)
dated
October
17,
2000.

Aventis
response:
Most
of
these
label
amendments
have
already
been
made
and
approved
by
the
EPA.
See
detailed
information
in
the
review
of
the
Residue
Chemistry
Chapter
below.

Page
96
Paragraph
7
Lines
1­
2
EPA
comment:
Adequate
residue
data
are
available
on
olives
provided
that
use
directions
for
olives
are
amended
to
remove
the
statement
allowing
the
use
of
summer
oil
as
an
adjuvant.

Aventis
response:
The
statement
allowing
the
use
of
summer
oil
as
an
adjuvant
has
already
been
deleted
from
the
use
directions
for
olives.

Page:
97
Paragraph:
2
Line
1
EPA
comment:
Additional
data
are
required
depicting
carbaryl
residue
in/
on
cotton
gin
byproducts.

Aventis
response:
Aventis
is
not
supporting
the
cotton
use
and
has
removed
this
crop
from
the
labels.
Aventis
CropScience
requested
cancellation
of
this
use
in
a
letter
to
Mr.
George
Tompkins,
RD,
on
January
22,
1999.
A
Federal
Register
Notice
dated
April
14,
1999
announced
receipt
by
the
Agency
of
an
application
from
RhonePoulenc
Ag
Company
(now
Aventis
CropScience)
to
cancel
the
use
of
carbaryl
products
on
cotton.
52
Page:
97
Paragraph:
3
Lines
1­
4
EPA
comment:
The
registrant
does
not
intend
to
support
carbaryl
on
avocados,
barley,
maple
sap,
oats,
rye,
and
sweet
sorghum;
however,
IR­
4
has
indicated
(correspondence
from
K.
Dorschner,
IR­
4
Project,
9/
15/
94)
that
they
may
fulfill
the
residue
data
requirements
for
some
of
these
commodities.
These
data
have
not
been
submitted.

Aventis
response:
IR­
4
has
not
generated
the
residue
data
necessary
to
support
these
uses.
Aventis
CropScience
will
not
support
these
uses.

Page:
97
Paragraph:
4
Lines
1­
2
EPA
comment:
The
reregistration
requirements
for
magnitude
of
the
residue
in
livestock
commodities
are
not
fulfilled.
Additional
data
are
required
to
support
dermal
and
poultry
house
uses.

Aventis
response:
Aventis
CropScience
will
neither
support
dermal
nor
poultry
house
uses
of
carbaryl.
A
request
for
cancellation
of
these
uses
in
accordance
with
section
6(
f)(
1)
of
the
Federal
Insecticide,
Fungicide
and
Rodenticide
Act
(FIFRA)
will
be
submitted
shortly
to
the
Agency.
53
Line­
by­
Line
Review
of
the
Supporting
Document
"Product
and
Residue
Chemistry
Chapters
for
the
Reregistration
Eligibility
Decision
(Felecia
Fort;
November
14,
2000)"

Cover
Memo
Residue
Chemistry
Page
2
Paragraph
3
1
st
Bullet
Pont
EPA
comment:
A
review
of
the
labels
and
supporting
residue
data
indicate
that
several
label
amendments
are
required.

Aventis
response:
Most
of
these
label
amendments
have
already
been
made
and
approved
by
the
EPA.

Page
3
6
th
Bullet
Point
Lines
1­
2
EPA
comment:
Adequate
residue
data
are
available
on
olives
provided
that
use
directions
for
olives
are
amended
to
remove
the
statement
allowing
the
use
of
summer
oil
as
an
adjuvant.

Aventis
response:
The
statement
allowing
the
use
of
summer
oil
as
an
adjuvant
has
already
been
deleted
from
the
use
directions
for
olives.

Page:
4
1
st
Bullet
Point
Line
1
EPA
comment:
Additional
data
are
required
depicting
carbaryl
residue
in/
on
cotton
gin
byproducts.

Aventis
response:
Aventis
is
not
supporting
the
cotton
use
and
has
removed
this
crop
from
the
labels.
Aventis
CropScience
requested
cancellation
of
this
use
in
a
letter
to
Mr.
George
Tompkins,
RD,
on
January
22,
1999.
A
Federal
Register
Notice
dated
April
14,
1999
announced
receipt
by
the
Agency
of
an
application
from
RhonePoulenc
Ag
Company
(now
Aventis
CropScience)
to
cancel
the
use
of
carbaryl
products
on
cotton.

Page:
4
2
nd
Bullet
Point
Lines
1­
4
EPA
comment:
The
registrant
does
not
intend
to
support
carbaryl
on
avocados,
barley,
maple
sap,
oats,
rye,
and
sweet
sorghum;
however,
IR­
4
has
indicated
(Correspondence
from
K.
Dorschner,
IR­
4
Project,
9/
15/
94)
that
they
may
fulfill
the
residue
data
requirements
for
some
of
these
commodities.
These
data
have
not
been
submitted.
54
Aventis
response:
IR­
4
has
not
generated
the
residue
data
necessary
to
support
these
uses.
Aventis
CropScience
will
not
support
these
uses.

Page:
4
3
rd
Bullet
Point
Lines
1­
2
EPA
comment:
The
reregistration
requirements
for
magnitude
of
the
residue
in
livestock
commodities
are
not
fulfilled.
Additional
data
are
required
to
support
dermal
and
poultry
house
uses.

Aventis
response:
Aventis
CropScience
will
neither
support
dermal
nor
poultry
house
uses
of
carbaryl.
A
request
for
cancellation
of
these
uses
in
accordance
with
section
6(
f)(
1)
of
the
Federal
Insecticide,
Fungicide
and
Rodenticide
Act
(FIFRA)
will
be
submitted
shortly
to
the
Agency.

Residue
Chemistry
Chapter
of
the
Reregistration
Eligibility
Decision
(RED)
Document
Residue
Chemistry
Considerations
Summary
of
Science
Findings
Page
3
Table
A1.
Carbaryl
EPs
with
Food/
Feed
Uses
Registered
to
Aventis
Ag
Company
EPA
comment:
Label
acceptance
dates
for
Carbaryl
EPs
are
listed
Aventis
response:
Label
acceptance
dates
listed
for
certain
products
are
incorrect.
Acceptance
dates
of
the
most
recently
EPA
approved
labels
for
SEVIN®
80S,
SEVIN®
5%
Bait,
SEVIN®
XLR
PLUS,
SEVIN®
RP2,
SEVIN®
RP4,
SEVIN®
4F,
and
SEVIN®
80WSP
are
listed
below.

EPA
Reg.
No.
Corrected
Label
Acceptance
Date
Product
Name
264­
316
1/
13/
2000
Notification
7/
26/
2001
SEVIN®
brand
80S
Carbaryl
Insecticide
264­
320
6/
11/
1999
SEVIN®
brand
5%
Bait
Carbaryl
Insecticide
264­
333
2/
8/
2001
SEVIN®
brand
XLR
Carbaryl
Insecticide
(Alternate
brand
name:
SEVIN®
brand
XLR
PLUS
Carbaryl
Insecticide)
264­
334
Notification
10/
29/
1999
SEVIN®
brand
RP2
Carbaryl
Insecticide
264­
335
10/
18/
2000
SEVIN®
brand
RP4
Carbaryl
Insecticide
264­
349
1/
13/
2000
SEVIN®
brand
4F
Carbaryl
Insecticide
55
264­
526
4/
12/
2000
SEVIN®
brand
80
WSP
Carbaryl
Insecticide
The
following
SLN
Nos
are
no
longer
active:

SEVIN®
brand
50W:
CA830007,
NC820007
SEVIN®
brand
80S:
CA8300007,
WA900013
SEVIN®
brand
XLR:
NC960003,
OH960003,
OR950006,
PA960002,
VA950001,
WA940021
SEVIN®
brand
4F:
FL890037
Page
3
Paragraph
2
Lines
1­
2
EPA
comment:
Use
directions
for
sugar
beets
on
all
labels
should
be
amended
to
specify
a
maximum
of
two
applications
at
a
maximum
single
application
rate
of
1.5­
lb
ai/
A
and
a
28­
day
PHI.

Aventis
response:
This
change
has
been
made
on
all
Aventis
CropScience
product
labels.

Page
3
Paragraph
3
Lines
1­
2
EPA
comment:
Use
directions
for
sweet
corn
on
the
G
formulations
should
be
amended
to
specify
a
48­
day
PHI
for
stover.

Aventis
response:
This
change
has
been
made.

Page
3
Paragraph
4
Lines
1­
2
EPA
comment:
As
the
registrant
is
no
longer
supporting
uses
on
avocado,
the
SLN
labeling
(CA83007)
for
use
of
carbary
(sic)
on
avocados
in
CA
should
be
revoked.

Aventis
response:
This
SLN
label
has
been
canceled.

Page
3
Paragraph
5
Line
1
EPA
comment:
Use
directions
for
cotton
on
all
labels
should
be
amended
to
remove
the
14­
day
PHI
for
forage.

Aventis
response:
The
use
of
carbaryl
products
on
cotton
has
been
canceled.
56
Page
4
Paragraph
1
Lines
1­
3
EPA
comment:
Based
on
acceptable
residue
data
on
okra
from
IR­
4,
the
registrant
should
amend
use
directions
on
FIC
and
WP
labels
to
specify
a
maximum
or
four
applications
per
season
at
1.5
lb
ai/
A/
application
at
a
minimum
re­
treatment
interval
(RTI)
of
6
days
and
a
minimum
PHI
of
3
days.

Aventis
response:
The
current
labels
allow
the
application
of
1
to
1.5
lb.
ai/
A
on
a
6­
to
8­
day
interval.
A
maximum
of
6­
lb
ai/
A
may
be
applied
per
season
with
a
PHI
of
3
days.

Page
4
Paragraph
2
Lines
1­
2
EPA
comment:
Use
directions
for
olives
on
all
labels
should
be
amended
to
remove
the
statement
allowing
the
use
of
a
tank
mix
with
summer
oil.

Aventis
response:
This
change
has
been
made
on
all
Aventis
CropScience
product
labels.

Page
4
Paragraph
3
Lines
1­
3
EPA
comment:
Based
on
acceptable
residue
data
on
prickly
pear
cactus
from
IR­
4,
the
registrant
should
amend
use
directions
on
FIC
and
WP
labels
to
specify
a
maximum
of
three
applications
per
season
at
2
lb
ai/
A/
application
at
a
minimum
RTI
of
7
days
and
a
minimum
PHI
of
3
days.

Aventis
response:
The
current
labels
allow
the
application
of
2
lb
ai/
A
on
a
7
to
10
day
interval.
A
maximum
of
6
lb
ai/
A
may
be
applied
per
season
with
a
PHI
of
3
days.

GLN
860.1300:
Nature
of
the
Residue
­
Livestock
Page
5
Paragraph
4
Lines
2­
6
EPA
comment:
The
registrant
has
stated
its
intention
to
support
dermal
uses
on
poultry
and
carbaryl
uses
in
poultry
houses
and,
according
to
REFS,
these
uses
are
on
carbaryl
labels.
If
the
registrant
intends
to
support
these
uses
of
carbaryl
on
poultry,
tolerances,
supported
by
adequate
metabolism
and
magnitude
of
the
residue
data,
will
be
required,
at
levels
appropriate
for
these
uses.

Aventis
response:
Aventis
CropScience
will
no
longer
support
the
use
of
carbaryl
for
direct
application
to
poultry,
as
well
as
the
poultry
quarters
treatment.
We
will
submit
a
request
for
cancellation
of
these
uses
in
accordance
with
section
6(
f)(
1)
of
the
Federal
Insecticide,
Fungicide
and
Rodenticide
Act
(FIFRA).
57
GLN
860.1500:
Crop
Field
Trials
Page
9
Paragraph
2
Lines
1­
3
EPA
comment:
Adequate
data
are
available
to
reassess
the
tolerances
for
residues
of
carbaryl
in/
on
sugar
beet
roots
and
tops
provided
that
use
directions
on
five
currently
approved
labels
are
modified
to
allow
a
maximum
of
two
applications
per
season
at
1.5
lb
ai/
A/
application
and
a
PHI
of
28
days
Aventis
response:
The
labels
were
amended
to
conform
to
these
requirements.

Page
10
Paragraph
3
Lines
1­
2
EPA
comment:
Adequate
data
are
available
on
olives
provided
that
use
directions
for
olives
are
amended
to
remove
the
statement
allowing
the
use
of
summer
oil
as
an
adjuvant.

Aventis
response:
The
labels
were
amended
to
conform
to
this
requirement.

Page:
10
Paragraph
5
Lines
2­
3
EPA
comment:
Additional
data
are
required
depicting
carbaryl
residue
in/
on
cotton
gin
byproducts.

Aventis
response:
Aventis
is
not
supporting
the
cotton
use
and
has
removed
this
crop
from
the
labels.

Page:
10
Paragraph
6
Lines
1­
4
EPA
comment:
The
registrant
does
not
intend
to
support
carbaryl
on
avocados,
barley,
maple
sap,
oats,
rye,
and
sweet
sorghum;
however,
IR­
4
has
indicated
(correspondence
from
K.
Dorschner,
IR­
4
Project,
9/
15/
94)
that
they
may
fulfill
the
residue
data
requirements
for
some
of
these
commodities.
These
data
have
not
been
submitted.

Aventis
response:
IR­
4
has
not
generated
the
residue
data
necessary
to
support
these
uses.
Aventis
CropScience
will
not
support
these
uses.

Table
A2.
Food/
Feed
Use
Patterns
on
EP
Labels
Subject
to
Reregistration
for
Carbaryl
(Case
0080)

Page:
16
EPA
comment:
58
Maximum
Single
Application
Rate,
ai:
3.0
lb/
A
&
4.0
lb/
A
(CA
only).

Aventis
response:
Maximum
Single
Application
Rate,
ai:
3.0
lb/
A
&
4.0
lb/
A
(CA
only).
An
additional
application
at
the
dormant
or
delayed
dormant
timing
may
be
made
at
a
maximum
rate
of
5
lb/
A.

Page:
21
Site:
Beet,
Sugar
EPA
comment:
The
maximum
seasonal
rate
for
sugar
beets
is
4.0­
lb
ai/
A.

Aventis
response:
The
rate
for
the
80
WSP
is
4.0­
lb
ai/
A.
however,
the
maximum
seasonal
rate
for
flowable
formulations
to
sugar
beets
is
3.0
lb
ai/
A.

Page:
23
Site:
Broccoli,
Brussels
sprouts,
etc.
EPA
comment:
The
preharvest
interval
for
broccoli,
Brussel
sprouts,
cabbage,
cauliflower,
Chinese
cabbage,
collards,
kale,
kohlrabi,
and
mustard
greens
is
3
days.

Aventis
response:
The
preharvest
interval
for
broccoli,
Brussel
sprouts,
cabbage,
cauliflower,
and
kohlrabi
is
3
days.
The
preharvest
interval
for
Chinese
cabbage,
collards,
kale,
and
mustard
greens
is
14
days.

Page:
26
Site:
Citrus
fruits
EPA
comment:
The
maximum
seasonal
rate
to
Citrus
Fruits
in
CA
is
20.0
lbs/
A
Aventis
response:
The
maximum
seasonal
rate
to
Citrus
Fruits
in
CA
is
16.0
lbs/
A
59
Part
IV
­
Occupational
and
Residential
Exposure/
Risk
Assessment
Line­
by­
Line
Review
of
the
Supporting
Document
"Carbaryl:
Occupational
and
Residential
Exposure
Assessment
and
Recommendations
for
the
Reregistration
Eligibility
Decision
Document
(Jeffrey
L.
aDawson;
June
28,
2001)"

1.0
Occupational
and
Residential
Exposure/
Risk
Assessment
1.3
Summary
of
Hazard
Concerns
Page:
12
Table
1:
EPA
comment:
UF
column
for
short­
term
dermal,
inhalation
&
non­
dietary
ingestion
reads
"3000
for
residential
and
300
for
occupational"

Aventis
response:
This
box
should
read,
"3000
for
residential
and
300
for
occupational,+
dermal
absorption
factor"
for
consistency
purposes.
However,
for
the
reasons
previously
stated
the
extra
10X
and
3X
MOS
factors
should
be
removed
due
to
the
recently
submitted
toxicological
studies.

EPA
comment:
UF
column
for
intermediate­
term
dermal,
inhalation
&
non­
dietary
ingestion
reads
"1000
for
residential
and
100
for
occupational"
and
the
Type
of
Exposure
column
refers
only
to
dermal
exposure.

Aventis
response:
The
UF
box
should
read,
"1000
for
residential
and
100
for
occupational,+
dermal
absorption
factor"
for
consistency
purposes.
The
1000X
MOS
factor
should
be
reduced
as
the
extra
10X
MOS
is
no
longer
necessary.
In
addition,
the
Type
of
Exposure
box
should
read
"Intermediate­
term
dermal,
inhalation,
&
non­
dietary
ingestion"
for
accuracy.

2.0
Occupational
Exposures
and
Risk
2.1.2
Data
and
Assumptions
for
Handler
Exposure
Scenarios
Occupational
Handler
Exposure
Studies
Page:
34
Paragraph:
3
Note
to
Chemical
Review
Managers:
EPA
comment:
"
There
are
no
data
compensation
issues
associated
with
the
use
of
…the
propoxur
trigger
sprayer
study
has
a
signed
PHED
data
waiver
but
just
has
not
been
included
into
PHED
at
this
time.
60
Aventis
response:
Although
Bayer
has
submitted
a
data
compensation
waiver
for
the
inclusion
of
the
propoxur
study
in
PHED,
the
data
compensation
waiver
is
applicable
only
to
the
data
cited
once
in
PHED.
Specifically,
the
first
paragraph
of
the
data
compensation
waiver
form
states,
`The
following
pertains
only
to
those
data
taken
from
company
studies
and
incorporated
into
the
generic
exposure
monitoring
database'.
In
addition,
the
29
December
1986
memorandum
from
Douglas
Campt,
Director
of
OPP
to
pesticide
registrants
regarding
the
formation
of
PHED,
specifically
states
on
page
2,
bullet
point
1
Those
companies
who
submit
data
for
inclusion
in
the
data
base
will
waive
their
compensation
rights
under
FIFRA
to
the
data
when
referenced
generically
as
part
of
the
data
base.
However,
existing
compensation
rights
will
apply
when
a
company's
data
is
cited
specifically.

Aventis
concludes
that
the
specific
referencing
of
the
Bayer
propoxur
exposure
study
rather
than
a
PHED
generic
study
number
does
not
eliminate
the
data
compensation
rights
of
Bayer.
Aventis
believes
that
the
Health
Effects
Division
should
seek
specific
guidance
from
the
Office
of
General
Counsel
regarding
the
data
compensation
status
of
the
propoxur
study
for
use
in
the
carbaryl
RED.

2.0
Occupational
Exposures
and
Risk
2.1.2
Data
and
Assumptions
for
Handler
Exposure
Scenarios
Page:
32
Paragraph:
First
bullet
Line
2:
EPA
comment:
If
additional
information
such
as
average
or
typical
rates
are
available,
these
values
are
used
as
well
in
order
to
allow
risk
managers
to
make
a
more
informed
risk
management
decision.
Average
application
rates
were
available
from
the
SMART
meeting
and
BEAD's
QUA.
These
data
indicate
that
in
most
cases,
average
application
rates
differ
from
maximum
application
rates
on
average
by
a
factor
of
two.
For
example,
when
interpreting
the
results
of
the
cancer
assessment,
the
small
differences
generally
seen
in
the
available
rates
should
be
considered
along
with
the
overall
magnitude
of
the
cancer
risk
results.

Aventis
response:
Aventis
believes
that
the
average
or
typical
application
rates
provided
in
the
document
should
be
used
in
the
determination
of
intermediate­
term,
chronic,
and
cancer
risk
assessments.
The
use
of
the
label
maximum
application
rates
for
each
application
becomes
extremely
unlikely
for
intermediate
term
grower
exposure
assessments
of
10
applications,
commercial
intermediate­
term
assessments
of
30
applications,
and
chronic
or
cancer
risk
assessments.
Although
the
Agency
may
believe
it
is
obligated
to
look
at
the
maximum
application
rates
for
exposure
durations
longer
than
the
short­
term
assessments,
it
should
as
it
states
provide
61
estimates
of
risk
based
on
average
application
rates
for
the
intermediate­
term
and
longer
risk
assessments.
Appendix
C
provides
and
utilizes
only
label
maximum
application
rates
for
the
estimation
of
intermediate­
term
and
chronic
MOEs
in
Tables
3,
4,
5,
6,
8,
9,
10,
and
11.
The
risk
assessment
must
provide
MOEs
based
on
average
application
rates
to
allow
risk
managers
to
make
a
more
informed
risk
management
decision.

2.0
Occupational
Exposures
and
Risk
2.1.2
Data
and
Assumptions
for
Handler
Exposure
Scenarios
Page:
42
Table
9:
EPA
comment:
LCO
Turfgun
exposure
estimates
obtained
from
ORETF
study
OMA002
are
based
on
the
wettable
powder
formulation
data.
The
dermal
exposure
estimates
are
0.99
mg/
lb
a.
i.,
0.72
mg/
lb
a.
i.,
and
0.38
mg/
lb
a.
i.
for
single
layer
no
gloves,
single
layer
gloves,
and
double
layer
gloves
clothing
scenarios,
respectively.
Inhalation
exposure
is
0.062
mg/
lb
a.
i.
The
Table
9
footnote
states
WP
formulation
used
for
turfgun
assessment
as
the
unit
exposures
for
this
scenario
were
slightly
higher
than
for
the
other
formulations
(i.
e.,
well
within
a
factor
of
2).

Aventis
response:
Sevin
is
sold
to
the
professional
lawn
care
market
as
either
an
80S
in
water
soluble
packaging
or
as
the
SL
liquid
formulation.
Therefore,
the
use
of
the
wettable
powder
formulation
is
not
appropriate.
Therefore,
the
dermal
exposure
for
the
single
layer
of
clothing
and
glove
scenario
should
be
0.66
mg/
lb
a.
i.
based
on
all
formulations
and
the
inhalation
exposure
should
be
0.007
mg/
lb
a.
i.
based
on
the
higher
water­
soluble
packaging
estimate
compared
to
the
flowable
liquid
inhalation
exposure
estimate
of
0.002
mg/
lb
a.
i.

2.3
Occupational
Risk
Characterization
2.3.1
Handler
Characterization
Page:
103
Paragraph:
2
Line:
19
EPA
comment:
Where
available,
average
use
rates
were
also
used
to
provide
for
a
more
informed
risk
management
decision
Aventis
response:
While
Aventis
agrees
that
average
use
rates
should
be
used
for
intermediate­
term
and
longer
repeated
exposure
risk
assessments,
the
use
of
average
application
rates
in
the
exposure
assessments
does
not
appear
evident.
The
MOEs
presented
for
intermediate­
term
and
longer
risk
are
based
on
label
maximum
application
rates.
The
position
of
Aventis
is
that
handler
MOEs
for
intermediate­
term
and
62
longer
risk
assessments
should
be
based
on
average
application
rates
and
not
label
maximum
rates
as
presented
in
the
assessment.

3.0
Residential
and
Other
Non­
Occupational
Exposures
and
Risks
3.1.1
Handler
Exposure
Scenarios
Page:
116
Table
27:
EPA
comment:
Table
27
states
that
Hose­
end
sprayer
data
for
mix
your
own
(not
the
locking/
no
contact
package)
considered.

Aventis
response:
The
dermal
and
inhalation
exposure
estimates
of
2.61
mg/
lb
a.
i.
and
0.010
mg/
lb
a.
i.,
respectively,
for
the
ready­
to­
use
(RTU)
hose
end
sprayer
need
to
be
included
in
the
handler
exposure
scenario
estimations
of
homeowner
exposure
during
application
to
lawns.
The
use
of
the
product
in
this
packaging
results
in
exposure
estimates
that
are
different
than
the
dial
type
sprayer
(DTS).
Because
the
RTU
sprayer
is
a
significant
method
of
lawn
application,
the
consideration
of
the
potential
exposure
and
risk
for
lawn
applications
with
the
RTU
sprayer
need
to
be
presented
and
discussed
in
the
risk
assessment.
The
difference
in
exposure
potential
between
the
DTS
and
RTU
hose­
end
sprayers
may
provide
important
information
for
risk
management
decisions
that
are
currently
lacking.

3.0
Residential
and
Other
Non­
Occupational
Exposures
and
Risks
3.1.3
Residential
Handler
Exposure
and
Non­
Cancer
Risk
Estimates
Page:
122
Table
28:
EPA
comment:
Scenario
3,
Fire
Ant
MOEs
are
based
on
the
use
of
100
gallons
or
0.75
lb
a.
i./
event
for
the
garden
vegetable
hose­
end
spray
scenario
and
5
gallons
or
0.0375
lb
a.
i./
event
for
garden
low
pressure
handwand
applications.

Aventis
response:
Aventis
does
not
understand
why
there
is
a
20­
fold
difference
in
the
amount
of
active
ingredient
handled
between
the
hose­
end
sprayer
and
hand
wand
application
of
carbaryl
to
control
fire
ants
at
the
same
site,
gardens.
Aventis
believes
that
the
use
of
0.0375
lb
a.
i./
event
is
reasonable
for
a
vegetable
garden
and
that
the
use
of
100
gallons
of
spray
or
0.75
lb
a.
i./
A
is
more
representative
of
a
lawn
spot
application
to
control
fire
ants.
Scenario
3,
Fire
ant
vegetable
garden
use
should
be
reduced
to
0.075
lb
a.
i./
event,
which
is
similar
to
the
other
scenario
3
vegetable
garden
uses
and
consistent
with
the
fire
ant
control
in
vegetable
gardens
using
the
low
pressure
hand
wand
equipment.
Any
estimation
of
fire
ant
63
control
on
the
lawn
should
be
based
on
exposure
data
from
the
ORETF
hose­
end
sprayer
exposure
study
(OMA004)
and
not
from
the
carbaryl
vegetable
garden
exposure
study
(MRID
44459801).

3.0
Residential
and
Other
Non­
Occupational
Exposures
and
Risks
3.1.4
Residential
Handler
Exposure
and
Risk
Estimates
for
Cancer
Page:
128
Table
29:
EPA
comment:
Presentation
of
cancer
risks
to
three
significant
figures.

Aventis
response:
Aventis
does
not
believe
that
the
expression
of
cancer
risks
to
three
significant
figures
accurately
portrays
the
lack
of
precision
in
the
risk
assessment.
Cancer
risks
should
be
presented
as
either
one
or
two
significant
figures,
consistent
with
HED
policy.
The
use
of
three
significant
figures
is
leading
to
the
portrayal
of
cancer
risks
such
as
1.08
x
10
­6
for
fire
ants
or
1.24
x
10
­6
for
vegetable
garden
dusts
as
being
in
excess
of
the
1
x
10
­6
guideline.
Both
of
these
risks
are
more
correctly
presented
as
1
x
10
­6
to
present
the
lack
of
precision
in
the
cancer
risk
assessment.

Appendix
F:
Carbaryl
Residential
Handler
Exposure
Data
Table
2.
Input
Parameters
for
Carbaryl
Homeowner
Handler
Exposure
and
Risk
Calculations
EPA
comment:
Inhalation
unit
exposures
are
estimated
based
on
a
breathing
volume
of
29
l/
min.

Aventis
response:
The
recommended
breathing
volume
for
adult
short­
term
exposure
is
1.0
m
3
/hr
or
17
l/
min
for
light
activities
(U.
S.
EPA
Exposure
Factors
Handbook,
Volume
1
General
Factors,
EPA/
600/
P­
95/
002Fa).
In
discussions
between
the
ORETF
and
the
regulatory
advisory
board
to
ORETF,
it
was
agreed
that
application
of
home
pesticide
products
constituted
light
activities
and
that
the
17
l/
min
breathing
rate
will
be
used.
The
29
l/
min
rate
used
by
HED
is
the
PHED
default
and
represents
agricultural
work
practices
that
are
more
strenuous
than
home
pesticide
application.
Therefore,
all
residential
inhalation
exposure
estimates
should
be
based
on
17
l/
min
breathing
volume
or
59%
of
the
inhalation
exposure
estimates
presented
in
the
HED
assessment.
64
Part
IV
­
Occupational
and
Residential
Exposure/
Risk
Assessment
SUPPORTING
DISCUSSION
Occupational
and
Residential
Exposure
Assessment
The
assessment
of
occupational
and
residential
exposure
and
risk
prepared
by
Mr.
Jeff
Dawson
was
a
comprehensive
assessment
of
excellent
quality.
This
is
reflected
by
the
relatively
minimal
number
of
factual
error
comments
provided
by
Aventis
CropScience.
Based
on
the
quality
of
the
initial
assessment,
the
major
effort
regarding
the
assessment
of
occupational
and
residential
exposure
to
carbaryl
will
involve
refinement
of
the
risk
assessment.

Aventis
CropScience
has
conducted,
or
is
in
the
process
of
conducting,
studies
relevant
to
the
refinement
of
the
occupational
and
residential
risk
assessment.
These
studies
are
as
follows:

 
4­
Week
Repeated­
Dose
Dermal
Toxicity
Study
with
Sevin
XLR
Plus
in
Rats.
The
study
is
completed
and
a
draft
report
is
currently
in
review
by
Aventis.
The
results
of
this
study
are
intended
to
replace
the
current
dermal
exposure
toxicity
endpoints
based
on
oral
toxicity
used
for
short­
term
and
intermediate­
term
occupational
and
residential
handler
exposure
when
handling
liquid
formulations
of
carbaryl.
Expected
completion
date:
September
2001
 
4­
Week
Repeated­
Dose
Dermal
Toxicity
Study
with
Sevin
80S
in
Rats.
The
study
is
completed
and
a
draft
report
is
currently
in
review
by
Aventis.
The
results
of
this
study
are
intended
to
replace
the
current
dermal
exposure
toxicity
endpoints
based
on
oral
toxicity
used
for
short­
term
and
intermediate­
term
occupational
and
residential
handler
exposure
when
handling
solid
formulations
of
carbaryl.
Expected
completion
date:
September
2001
 
4­
Week
Repeated­
Dose
Dermal
Toxicity
Study
with
Carbaryl
Technical
in
Rats.
The
study
is
completed
and
a
draft
report
is
currently
in
review
by
Aventis.
The
results
of
this
study
are
intended
to
replace
the
current
dermal
exposure
toxicity
endpoints
based
on
oral
toxicity
used
for
short­
term
and
intermediate­
term
occupational
and
residential
postapplication
dermal
exposure.
Expected
completion
date:
September
2001
 
Measurement
of
Pesticide
Exposure
of
Suburban
Residents
Associated
with
the
Residential
Use
of
Carbaryl.
This
is
an
ongoing
biological
monitoring
study
of
10
families
in
California
and
10
families
in
Missouri.
The
study
will
monitor
the
absorbed
dose
of
carbaryl
resulting
from
a
lawn
broadcast
application
and
subsequent
vegetable
garden
application
of
carbaryl
by
an
adult
family
member.
Postapplication
absorbed
dose
of
all
adults
and
children
aged
four
to
seventeen
will
be
followed
over
a
three­
day
period.
The
participants'
postapplication
activities
will
be
recorded
but
not
controlled
to
establish
a
range
of
absorbed
65
doses.
The
data
will
be
used
along
with
the
residential
exposure
assessment
to
improve
understanding
of
residential
exposure
to
carbaryl.
Expected
completion
date:
March
2002
 
University
of
California
­
Riverside
biological
monitoring
of
citrus
orchard
application
and
postapplication
apple
thinning
and
cherry
harvesting
job
functions
is
completed
in
the
field.
The
results
from
these
studies
will
be
used
to
refine
the
PHED
and
ARTF­
based
exposure
estimates.
These
are
high
exposure
potential
work
functions
and
the
biological
monitoring
data
are
intended
to
refine
the
assessments
of
these
work
functions.
Expected
completion
date:
March
2002
66
APPENDIX
I
Business
Confidential
Surface
Water
Monitoring
for
Residues
of
Carbaryl
in
High
Use
Areas
of
the
United
States
(Stone
Environmental,
Inc.
Report
#99­
1005­
F)
(hard
copy
provided)
1
CARBARYL
PC
Code
No.
056801;
Case
0080
Review
of
the
Draft
Environmental
Fate
and
Ecological
Risk
Assessment
for
the
Reregistration
of
Carbaryl
August
6,
2001
Aventis
CropScience
P.
O.
Box
12014,
2
T.
W.
Alexander
Drive
Research
Triangle
Park,
NC
27709
2
CARBARYL
PC
Code
No.
056801;
Case
0080
Review
of
the
Draft
Environmental
Fate
and
Ecological
Risk
Assessment
for
the
Reregistration
of
Carbaryl
TABLE
OF
CONTENTS
TABLE
OF
CONTENTS
...............................................................................................................
2
GENERAL
COMMENTS
....................................................................................................................
5
TRANSMITTAL
DOCUMENT
............................................................................................................
9
Data
Gaps
...............................................................................................................................
9
Environmental
Fate
and
Transport
...............................................................................
9
Water
Resources
...........................................................................................................
9
Ecological
Effects
Data
requirement
..........................................................................
10
Label
Information
.......................................................................................................
11
DRAFT
RED
DOCUMENT
..............................................................................................................
13
1.0
Summary
and
Environmental
Risk
Conclusions
...........................................................
13
Risk
to
Terrestrial
Organisms
.....................................................................................
13
Fate
and
Water
Assessment
........................................................................................
13
2.0
Introduction
....................................................................................................................
14
3.0
Integrated
Risk
Characterization
...................................................................................
15
Introduction
.................................................................................................................
15
Aquatic
Organisms
......................................................................................................
15
Terrestrial
Organisms
.................................................................................................
16
Endocrine
Disruption
Concerns
..................................................................................
18
Uncertainties
...............................................................................................................
19
4.0
Environmental
Fate
Assessment
....................................................................................
20
Exposure
Characterization
..........................................................................................
20
Persistence
....................................................................................................
23
Mobility
........................................................................................................
24
Field
Dissipation
...........................................................................................
25
Foliar
Dissipation
.........................................................................................
27
Atmospheric
Transport
.................................................................................
28
1­
Naphthol
Fate
and
Transport
.....................................................................
28
Aquatic
Exposure
Assessment
....................................................................................
30
Surface
Water
...............................................................................................
30
Estimated
Environmental
Concentrations
for
Terrestrial
Ecological
Risk
Assessment32
5.0
Drinking
Water
Assessment
..........................................................................................
32
Water
Resources
Assessment
.....................................................................................
32
Drinking
Water
Exposure
Assessment
.........................................................
33
Water
Treatment
Effects
..............................................................................
37
Ground
Water
Resources
..............................................................................
38
3
Surface
Water
Resources
..............................................................................
39
6.0
Hazard
and
Risk
Assessment
for
Aquatic
Organisms
..................................................
45
Hazard
assessment
for
Aquatic
organisms
.................................................................
45
Estuarine/
Marine
Fish
..................................................................................
45
Aquatic
Plants
...............................................................................................
45
Risk
Assessment
for
Aquatic
Organisms
....................................................................
45
Estuarine/
Marine
Fish
..................................................................................
46
7.0
Hazard
and
Risk
Assessment
for
Terrestrial
Organisms
..............................................
46
Hazard
Assessment
for
Terrestrial
Organisms
...........................................................
46
Mammalian
...................................................................................................
46
Risk
Assessment
for
Terrestrial
Organisms
................................................................
46
Avian
Risk
....................................................................................................
46
Mammalian
Risk
..........................................................................................
47
Reproduction
Effects
....................................................................................
48
9.0
References
(non­
MRID)
...............................................................................................
50
Appendix
A:
Environmental
Fate
Study
Reviews
(DERs)
.................................................
50
Appendix
B:
Refined
Water
Memo
....................................................................................
50
Appendix
C:
Ecological
Risk
Assessment
..........................................................................
51
Toxicity
Endpoints
Used
in
the
Risk
Assessment
......................................................
51
Avian
Acute
and
Chronic
Risk
...................................................................................
51
Risk
from
Exposure
to
Non­
granular
Products
...........................................................
52
Risk
from
Exposure
to
Granular
Products
..................................................................
52
Aquatic
Plants
.............................................................................................................
52
Appendix
D:
Toxicity
Assessment
......................................................................................
52
Birds,
Chronic
Toxicity
..............................................................................................
53
Mammals,
Acute
and
Chronic
....................................................................................
53
Freshwater
Fish,
Acute
...............................................................................................
54
Freshwater
Invertebrates,
Acute
.................................................................................
54
Estuarine
and
Marine
Invertebrates,
Acute
.................................................................
54
DISCUSSION
.............................................................................................................................
56
1.
Surface
Water
Concentrations
........................................................................................
56
Summary
of
Registrant
Surface
Water/
Drinking
Water
Monitoring
Program
...........
56
Summary
of
Surface
Water
Data
from
the
NAWQA
Program
..................................
58
Summary
of
Carbaryl
Analytical
Methods
used
in
the
NAWQA
Program
................
60
Gas
Chromatography/
Mass
Spectroscopy
Method
......................................
60
High­
Performance
Liquid
Chromatography/
Photodiode­
Array
Method
.....
64
Summary
of
Surface
Water
Data
from
the
California
DPR
Surface
Water
Database
64
2.
Ground
Water
Concentrations
........................................................................................
66
Summary
of
Ground
Water
Data
from
the
NAWQA
Program
..................................
66
REFERENCES
............................................................................................................................
68
4
TABLE
OF
CONTENTS
(continued)

Tables
Table
1.
Summary
Of
Results
From
The
Carbaryl
Drinking
Water
Monitoring
Study.
57
Table
2.
Carbaryl
Detections
Reported
In
Pesticides
In
Streams
Update
(Larson,
2001)
...........................................................................................................
58
Table
3.
Frequency
Of
Carbaryl
Detections
By
GC/
MS
In
Different
Concentration
Ranges
Reported
In
The
NAWQA
Database
As
Of
July
16,
2001
............
59
Table
4.
Frequency
Of
Carbaryl
Detections
By
LC/
PDA
In
Different
Concentration
Ranges
Reported
In
The
NAWQA
Database
As
Of
July
16,
2001
............
59
Table
5.
Recovery
And
Precision
For
Multiple
Determinations
Of
Carbaryl
In
GC/
MS
Method
For
Carbaryl
Spiked
In
Different
Water
Samples
............
61
Table
6.
Percent
Recoveries
Of
Carbaryl
Detected
By
The
NAWQA
GC/
MS
Method
In
Laboratory
Control
Spikes
And
Field
Matrix
Spikes
At
A
Spiking
Level
Of
0.1
µ
g/
L
.................................................................................................
63
Table
7.
Recovery
And
Precision
For
Multiple
Determinations
Of
Carbaryl
In
LC/
PDA
Method
For
Carbaryl
Spiked
In
Different
Water
Samples
..........
64
Table
8.
Carbaryl
Detections
Reported
In
California
DPR
Surface
Water
Monitoring
Database
......................................................................................................
65
Table
9.
Limits
Of
Quantification
For
Carbaryl
Analytical
Methods
Reported
In
California
DPR
Surface
Water
Monitoring
Database
.................................
65
Table
10.
Carbaryl
Detections
Reported
In
Pesticides
In
Ground
Water
Update
(Kolpin,
2001)
.............................................................................................
67
Confidential
Business
Attachments
APPENDIX
1
..............................................................................................................................
70
APPENDIX
2
..............................................................................................................................
71
The
last
page
of
this
report
is
number
70
\#
"0"
70.
5
General
Comments
The
EFED
draft
chapter
of
the
carbaryl
RED
is
very
thorough
using
a
wealth
of
references.
The
use
of
published
literature
over
submitted
data
is
significant.
The
quality
of
the
published
literature
is
at
times
at
least
questionable
and
other
times
does
not
fulfill
the
requirements
set
by
EPA
for
studies
submitted
by
the
registrant
(e.
g.
thorough
description
of
test
conditions,
clear
identification
of
the
test
material,
analytical
verification,
GLP
etc.).
Data
of
such
poor
quality
should
not
be
used
as
key
information
in
the
risk
assessment.
For
the
30­
day
response
not
all
literature
references
could
be
verified
or
the
quality
ascertained.

There
is
a
high
level
of
redundancy
in
the
document
making
it
difficult
to
read.
Reducing
repetitions
to
a
minimum
would
facilitate
the
reading.

We
believe
it
is
inappropriate
to
include
DERs
in
the
RED
Chapters.
A
summary
of
study
findings
is
already
presented
in
the
document.
DERs
should
be
made
available
to
the
public
through
the
regular
procedure
under
the
Freedom
of
Information
Act
after
they
have
been
reviewed
and
cleared
for
confidential
business
information.

The
use
of
carbaryl
on
barley,
oats,
rye,
cotton,
and
livestock
are
cancelled
.
It
should
be
noted
that
Aventis
CropScience
labels
for
the
technical
materials
and
the
end­
use
products
containing
carbaryl
were
amended
to
delete
these
uses.
The
Agency
has
already
approved
the
labeling
changes
(please
refer
to
HED
response
document,
Section
III
for
details).

Aventis
CropScience
will
no
longer
support
the
use
of
carbaryl
on
poultry
(direct
application
and
poultry
quarters
treatment).
We
will
shortly
submit
a
request
for
cancellation
of
these
uses
in
accordance
with
section
6(
f)(
1)
of
the
Federal
Insecticide,
Fungicide
and
Rodenticide
Act
(FIFRA)
(please
refer
to
HED
response
document,
Section
III
for
details).

Aventis
CropScience
is
in
the
process
of
conducting,
or
has
scheduled,
studies
relevant
to
the
refinement
of
the
environmental
risk
assessments
for
carbaryl
and
the
major
degradate
1­
naphthol.
These
studies
are
as
follows:

 
Rate
and
Route
of
Aerobic
Degradation
in
Soils.
These
studies
have
been
initiated
with
parent
carbaryl
applied
to
four
diverse
U.
S.
soils.
The
data
are
intended
to
provide
additional
half­
life
determinations
for
parent
carbaryl
and
the
major
degradate
1­
naphthol.
Expected
completion
date:
March
2002
 
Aerobic
Aquatic
Metabolism
in
Two
Water/
Sediment
Systems.
These
studies
have
been
initiated
with
parent
carbaryl
applied
to
two
distinct
U.
S.
water/
sediment
systems.
The
data
are
intended
to
provide
additional
half­
life
determinations
for
parent
carbaryl
and
the
major
degradate
1­
naphthol.
In
addition,
further
identification
of
additional
degradation
products
is
anticipated.
Expected
completion
date:
March
2002
 
Adsorption
and
Desorption
of
1­
Napthol
to
five
soils.
This
study
has
been
scheduled
to
evaluate
the
adsorption
and
desorption
of
the
major
carbaryl
degradate
to
five
soils/
sediment.
The
data
are
intended
to
provide
information
necessary
to
evaluate
the
6
environmental
risks
from
1­
naphthol
in
standard
models.
Expected
completion
date:
March
2002
For
the
reregistration
process
in
the
EU,
Aventis
CropScience
is
in
the
process
of
conducting,
or
has
scheduled,
studies
relevant
to
the
refinement
of
the
ecotoxicological
risk
assessments
for
carbaryl
and
the
major
degradate
1­
naphthol.
These
studies
are
as
follows:

 
Studies
with
Carbaryl:
Acute
oral
LD50
in
mallard
ducks
Dynamic
acute
LC50
in
bluegill
sunfish
Acute
toxicity
in
Daphnia
Acute
toxicity
in
Chironomus
riparius
Toxicity
in
Selenastrum
capricornutum
Acute
oral
and
contact
toxicity
in
honeybees
14­
d
toxicity
in
earthworms
Effects
on
soil
microorganisms
(nitrification/
carbon
cycle)
Effect
on
sewage
treatment
 
Studies
with
1­
naphthol
Early
life­
stage
study
in
fathead
minnow
Acute
toxicity
in
Daphnia
Acute
toxicity
in
Daphnia
in
presence
of
sediment
Chronic
toxicity
in
Daphnia
14­
d
toxicity
in
earthworms
 
Formulated
Product
Vegetative
Vigor
Toxicity
in
Selenastrum
capricornutum
Acute
oral
and
contact
toxicity
in
honeybees
Effect
on
non­
target
arthropods
14­
d
toxicity
in
earthworms
Effects
on
soil
microorganisms
(nitrification/
carbon
cycle)

Ecotoxicological
Risk
Assessments
Aventis
has
pointed
out
several
errors
in
the
PRZM
input
parameters
(see
comments
made
to
Tables
5
and
6
of
the
draft
RED),
overly
conservative
estimates
of
foliar
dissipation
half­
lives
and
changes
in
ecotoxicology
study
endpoints.
This
indicates
that
a
re­
calculation
of
the
EECs
and
risk
quotients
are
warranted
in
a
number
of
instances.

Endocrine
Disruption
Reports
in
the
open
literature
on
the
reproductive
effects
of
carbaryl
in
wild
mammals
are
at
best
ambivalent.
The
recently
submitted
2­
generation
study
in
rats
demonstrates
the
absence
of
7
reproductive
effects.
As
EPA
pointed
out,
findings
reported
in
the
literature
were
made
at
concentrations
well
above
the
highest
peak
concentration
modeled.
Therefore
these
findings
are
irrelevant
for
a
risk
assessment
and
at
the
current
stage
of
discussion
about
endocrine
disruption.
If
the
concern
about
the
endocrine
potential
of
carbaryl
persists,
the
issue
should
be
revisited
once
the
Agency's
endocrine
disrupter
screening
and
testing
program
as
well
as
a
policy
on
how
to
incorporate
positive
findings
into
an
ecological
risk
assessment
have
been
fully
developed.

Mobility
The
classification
of
carbaryl
as
mobile
to
very
mobile
is
inconsistent
with
measured
Koc
values
of
177
to
249
(MRID
43259301).
According
to
the
widely
used
classification
scheme
of
McCall,
et
al.
(1980)
wherein
Koc
values
between
150
and
500
denote
medium
mobility
in
soil,
carbaryl
would
be
classified
as
having
medium
mobility
in
most
soils.
This
classification
of
medium
mobility
is
further
supported
by
the
acceptable
column
leaching
study
(MRID
43320701)
in
which
aged
carbaryl
residues
were
only
slightly
mobile
in
a
number
of
soils.
The
mobility
of
carbaryl
would
be
expected
to
be
higher
in
sandy
soils
or
in
soils
of
low
organic
matter.

1­
Napthol
Fate
and
Transport
The
Agency
is
requiring
additional
information
on
the
persistence
and
mobility
of
1­
naphthol,
a
major
environmental
degradate
of
carbaryl.
However,
a
half­
life
for
1­
naphthol
of
less
than
1
day
can
be
calculated
from
the
carbaryl
aerobic
soil
metabolism
study
(MRID
42785101).
The
data
from
this
study
demonstrate
that
under
aerobic
soil
conditions
the
formation
and
decline
of
1­
naphthol,
starting
from
parent
carbaryl,
is
complete
in
less
than
14
days.
This
half­
life
can
be
used
for
preliminary
environmental
fate
modeling
to
estimate
EECs
for
1­
naphthol.

The
EPA
suggested
that
1­
naphthol
is
not
strongly
sorbed
to
soil.
Additional
information
available
in
the
literature
demonstrates
that
the
sorption
of
1­
naphthol
to
soil
is
stronger
than
that
seen
for
carbaryl
itself.
Hassett
et
al.
(1981)
has
demonstrated
that
the
sorption
of
1­
naphthol
was
the
result
of
sorption
to
organic
carbon
resulting
in
Koc
values
between
431
and
15,618.
These
data
indicate
that
1­
naphthol
is
less
mobile
and
less
susceptible
to
leaching
than
carbaryl
itself,
and
they
demonstrate
that
at
least
a
portion
of
the
1­
naphthol
residue
is
tightly
sorbed
to
soil
constituents.
(A
copy
of
this
article
is
being
submitted
with
the
response
to
the
draft
RED.)
To
meet
the
requirement
for
information
on
the
adsorption
and
desorption
of
1­
naphthol
by
the
Agency,
the
registrant
is
conducting
an
adsorption/
desorption
study
to
meet
the
163­
1
guideline.
Study
results
should
be
available
for
submission
to
the
Agency
in
the
first
quarter
of
the
calendar
year
2002.

Surface
Water/
Drinking
Water
Aventis
disagrees
with
EPA
that
the
modeling
simulations
provide
a
conservative,
though
not
unreasonable,
estimate
on
possible
concentrations
in
drinking
water.
Drinking
water
concentrations
derived
from
PRZM/
EXAMS
greatly
overestimate
the
potential
exposure
to
carbaryl
in
drinking
water,
generally
by
several
orders
of
magnitude.
Results
from
the
drinking
8
water
monitoring
program
conducted
by
the
registrant
provides
a
`real
world'
assessment
of
the
potential
for
human
exposure
to
carbaryl
in
drinking
water
derived
from
surface
water.

Ground
Water
EPA
summarized
information
on
the
detection
of
carbaryl
in
groundwater
from
the
EPA
Pesticides
in
Groundwater
Database,
the
EPA
STORET
database
and
the
NAWQA
database.
Each
of
the
databases
shows
a
pattern
of
very
low
levels
of
carbaryl
detection
in
few
groundwater
resources.
These
analyses
confirm
several
statements
made
by
the
Agency
that
carbaryl
have
limited
potential
to
impact
groundwater
resources.
However,
on
page
2
of
the
Memorandum
issued
June
28,
2001,
in
conjunction
with
the
EFED
RED
chapter
for
carbaryl,
EPA
is
requiring
additional
information
on
"Surface
and
groundwater
monitoring
in
urban
and
suburban
use
areas
(non­
guideline)."
Based
on
the
characteristics
of
carbaryl
and
the
available
data
demonstrating
limited
impact
of
carbaryl
on
ground
water
resources,
additional
studies
to
evaluate
the
potential
for
carbaryl
to
contaminate
groundwater
are
unnecessary
and
unwarranted.
9
Line­
by­
Line
Review
of
the
Carbaryl
EFED
RED
Chapter
Transmittal
Document
Data
Gaps
Environmental
Fate
and
Transport
Page:
2
Paragraph:
1
Line:
1
EPA
comment:
Fate
information
on
the
degradation
product
1­
naphthol
is
required.

1.
Mobility
–
adsorption
and
desorption
studies
for
the
1­
naphthol
degradate
(163­
1)
2.
Persistence
–
aerobic
soil
metabolism
study
on
1­
naphthol
Aventis'
response:
Literature
data
(Hassett
et
al.
1981)
on
the
adsorption
of
1­
naphthol
are
provided
in
this
response.
Aventis
is
in
the
process
of
conducting
an
additional
adsorption/
desorption
study
on
1­
naphthol
and
intends
to
submit
study
data
to
EPA
by
March
2002.

The
degradation
of
1­
naphthol
under
aerobic
soil
conditions
has
been
widely
reported
in
the
literature.
Several
citations
are
included
in
the
EPA
draft
RED.
The
half­
life
of
1­
naphthol
estimated
from
the
acceptable
aerobic
soil
persistence
study
on
carbaryl
(MRID
42785101)
is
less
than
1
day.
Aventis
is
conducting
additional
laboratory
aerobic
soil
degradation
studies
on
carbaryl
that
will
be
used
to
provide
additional
determinations
of
the
half­
life
for
the
degradate
1­
naphthol
and
satisfy
the
Agency's
requirement
for
data
on
the
persistence
of
1­
naphthol.
Aventis
intends
to
submit
these
study
data
to
EPA
by
March
2002.

Water
Resources
Page:
2
Paragraph:
3
and
4
EPA
comment:
"EFED
believes
that
adequate
data
are
available
to
support
the
conclusions
reached
for
carbaryl's
impact
on
surface
water
and
groundwater
quality
with
the
exceptions
noted
below.
Additional
information
is
needed
to
characterize
the
impact
of
the
degradate
1­
naphthol
[in]
groundwater
and
surface
water.
 
Surface
and
groundwater
monitoring
in
urban
and
suburban
use
areas
(non­
guideline)"
are
required.
10
Aventis'
response:
The
surface
water­
monitoring
program
conducted
by
Aventis
includes
monitoring
in
urban
and
suburban
use
areas.
Aventis
believes
that
the
need
for
information
on
the
degradate
1­
naphthol
will
be
satisfied
by
the
aerobic
soil
and
adsorption/
desorption
data
that
will
be
submitted
to
the
Agency.
These
data
can
be
used
to
evaluate
the
availability
of
1­
naphthol
using
established
EPA
modeling
guidelines.
The
Agency's
proposed
requirement
for
groundwater
monitoring
is
unnecessary
and
is
addressed
in
Aventis'
response
to
Agency
comments
in
the
draft
RED.

Ecological
Effects
Data
requirement
Page:
2
EPA
comment:
The
ecological
toxicity
database
is
complete
except
for:
6.
Aquatic
Plant
Growth
Guideline
122­
2
Aventis'
response:
The
data
requirement
should
be
deleted.
Aquatic
plant
growth
studies
were
submitted
to
the
Agency
in
1992.
An
October
04,
2000
OPP
Guideline
Status
Report
(Chemical
Review
Management
System)
lists
the
guideline
122­
2
status
as
"Acceptable/
Satisfied".
The
studies
are:

MRID
No.
Title
Acceptability
Code
42372101
Lintott,
D.
(1992)
Carbaryl
Technical:
Acute
Toxicity
To
The
Freshwater
Blue­
Green
Alga,
Anabaena
flos­
aquae,
Under
Static
Test
Conditions:
Lab
Project
Number:
J9112004E.
Unpublished
Study
Prepared
By
Toxikon
Environmental
Sciences.
53
P.
June
25,
1992
Upgradable
42372102
Lintott,
D.
(1992)
Carbaryl
Technical:
Acute
Toxicity
To
Duckweed,
Lemna
gibba
G3,
Under
Static
Test
Conditions:
Lab
Project
Number:
J9112004G.
Unpublished
Study
Prepared
By
Toxikon
Environmental
Sciences.
53
P.
January
1,
1992
Upgradable
42372802
Lintott,
D.
(1992)
Carbaryl
Technical:
Acute
Toxicity
To
The
Freshwater
Green
Alga,
Selenastrum
capricornutum
Under
Static
Conditions:
Lab
Project
Number:
J9112004C.
Unpublished
Study
Prepared
By
Toxikon
Environmental
Sciences.
53
P.
June
9,
1992
Acceptable
42431601
Lintott,
D.
(1992)
Carbaryl
Technical:
Acute
Toxicity
To
The
Freshwater
Diatom,
Navicula
pelliculosa,
Under
Static
Test
Conditions:
Lab
Project
Number:
J9112004F.
Unpublished
Study
Prepared
By
Toxikon
Environmental
Sciences.
52
P.
August
10,
1992
Acceptable
42431602
Lintott,
D.
(1992)
Carbaryl
Technical:
Acute
Toxicity
To
The
Saltwater
Diatom,
Skeletonema
costatum,
Under
Static
Test
Conditions:
Lab
Project
Number:
J9112004D.
Unpublished
Study
Prepared
By
Toxikon
Environmental
Sciences.
49
P.
August
10,
1992
Supplemental
11
Page:
2
EPA
comment:
The
ecological
toxicity
database
is
complete
except
for:
7.
Submission
of
a
FETOX
amphibian
toxicity
study
is
required.
Aventis'
response:
The
data
requirement
should
be
deleted.
From
the
published
results
it
is
evident
that
carbaryl
is
practically
non­
toxic
to
the
bullfrog.
Effects
in
plain
leopard
frogs
are
reported
at
levels
well
above
environmental
concentrations.
These
results
were
obtained
testing
U.
S.
native
species.
In
the
proposed
FETOX
assay,
a
non­
native
species
Xenopus
laevis
is
used.
This
African
species
is
unique
in
its
behavior.
Neither
the
species
nor
the
test
methods
are
suitable
for
ecotoxicological
purposes.
As
the
risk
to
amphibians
can
be
evaluated
from
the
studies
cited,
and
as
the
effects
are
only
at
levels
well
above
the
EEC,
this
study
should
not
be
required.

Label
Information
Page:
3
EPA
comment:
For
terrestrial
and
residential
uses:
1.
"Do
not
apply
directly
to
water,
or
to
areas
where
surface
water
is
present
or
to
intertidal
areas
below
the
mean
high
water
mark.
Do
not
contaminate
water
when
disposing
of
equipment
washwater
or
rinsate."
Aventis'
response:
Similar
language
is
already
present
on
Aventis'
SEVIN®
labels.

Page:
3
EPA
comment:
For
terrestrial
and
residential
uses:
3.
"This
product
may
contaminate
water
through
drift
of
spray
in
wind.
This
product
has
a
high
potential
for
runoff
for
several
days
after
application
after
application
(sic).
Poorly
draining
soils
and
soils
with
shallow
water
tables
are
more
prone
to
produce
runoff
that
contains
this
product.

Household
labels
–
Avoid
applying
this
product
to
ditches,
swales,
and
drainage
ways.
Runoff
of
this
product
will
be
reduced
by
avoiding
applications
when
rainfall
is
forecasted
to
occur
within
48
hours.

Agricultural
Label
–
A
level,
well
maintained
vegetative
buffer
strip
between
areas
to
which
this
product
is
applied
and
surface
water
features
such
as
ponds,
streams,
and
springs
will
reduce
the
potential
for
contamination
of
water
from
rainfall­
runoff.
Runoff
of
this
product
will
be
reduced
by
12
avoiding
applications
when
rainfall
is
forecasted
to
occur
within
48
hours."
Aventis'
response:
Aventis
would
like
to
further
discuss
appropriate
label
language
with
the
Agency.
However,
it
should
be
noted
that
light
to
moderate
rainfall
(or
irrigation)
after
application
will
also
help
move
carbaryl
residues
deeper
into
the
soil,
thus
making
them
less
susceptible
to
runoff.
The
language
in
the
last
sentence
should
be
changed
to
read,
"…
when
heavy
rainfall
is….".

Page:
3
EPA
comment:
For
terrestrial
and
residential
uses:
4.
This
pesticide
is
toxic
to
fish
and
aquatic
invertebrates.
Aventis'
response:
Aventis'
SEVIN
labels
currently
state
"This
product
is
extremely
toxic
to
aquatic
and
estuarine
invertebrates."

Page:
3
EPA
comment:
For
terrestrial
and
residential
uses:
5.
This
product
is
highly
toxic
to
bees
exposed
to
direct
treatment
or
residues
on
blooming
crops
or
weeds.
Do
not
apply
this
product
or
allow
it
to
drift
to
blooming
crops
or
weeds
if
bees
are
visiting
the
treatment
area.
Aventis'
response:
Aventis'
SEVIN
labels
currently
contain
similar
language.
13
Draft
RED
Document
1.0
Summary
and
Environmental
Risk
Conclusions
Risk
to
Terrestrial
Organisms
Page:
1
Paragraph:
4
Line:
2
EPA
comment:
As
discussed
in
pp.
44
­
45
and
in
Appendix
D.
Aventis'
response:
The
mammalian
risk
quotients
are
discussed
on
pages
48
to
50
and
in
Appendix
C,
not
as
described
in
this
text.

Fate
and
Water
Assessment
Page:
3
Paragraph:
5
Line:
3
EPA
comment:
…in
the
U.
S.
G.
S
NAQWA
program.
NAQWA…
Aventis'
response:
The
abbreviation
for
the
U.
S.
G.
S.
program
is
NAWQA
Page:
5
Paragraph:
1
Line:
7
EPA
comment:
…estimate
of
possible
concentrations
drinking
water.
Aventis'
response:
missing
word
–
…concentrations
"in"
drinking…
Page:
5
Paragraph:
4
Line:
4
EPA
comment:
…hydrolyzes
in
neutral
(half­
life
=
12
days)
and
alkaline
environments
(pH
9
half­
life
=
3.2).
Aventis'
response:
Missing
units
of
after
second
half­
life.
The
units
are
hours,
so
"=
3.2
hours)".

Page:
5
Paragraph:
4
Line:
5
EPA
comment:
…photolysis
in
water
with
a
half­
life
of
21
days
14
Aventis'
response:
this
is
for
photolysis
in
sterile
water,
not
microbially­
active
water,
so
the
phrase
would
be
more
precise
as
"…
photolysis
in
sterile
water…".

Page:
5
Paragraph:
4
Line:
last
EPA
comment:
(Kf
=1.7
to
3.2).
Aventis'
response:
The
upper
value
Kf
for
carbaryl
should
be
listed
as
3.5
as
referenced
by
EPA
elsewhere
(e.
g.
Table
3,
page
20)
in
the
document.

2.0
Introduction
Page:
6
Paragraph:
2
Line:
1­
3
EPA
comment:
Carbaryl
(1­
naphthyl
N­
methylcarbamate)
is
a
broad­
spectrum
carbamate
insecticide
and
acaricide
registered
for
control
of
over
300
species
of
insects
and
mites
on
over
100
crop
and
noncrop
use
sites,
including
homeowner
uses;
pet,
poultry,
and
livestock
uses;…
Aventis'
response:
Carbaryl
is
no
longer
registered
for
use
on
livestock.
Aventis
CropScience
will
not
support
the
reregistration
of
the
use
on
poultry
(direct
application
and
poultry
quarters
treatment).
We
will
shortly
submit
a
request
for
cancellation
of
this
use
in
accordance
with
section
6(
f)(
1)
of
the
Federal
Insecticide,
Fungicide
and
Rodenticide
Act
(FIFRA).

Page:
6
Paragraph:
3
Line:
2­
3
EPA
comment:
Approximately
2.5
million
pounds
of
carbaryl
are
applied
annually
in
the
U.
S.
A
map
showing
the
widespread
use
of
carbaryl
in
agriculture
is
shown
in
figure
1.
Aventis'
response:
Summation
of
the
data
in
Figure
1
gives
a
total
of
approximately
3.3
million
pounds
of
carbaryl.
Both
the
2.5
and
3.3
million­
pound
figures
are
inconsistent
with
the
value
of
4
million
pounds
cited
on
page
35.
The
2.5
million
pounds
is
an
average
of
usage
over
1987
to
1996
developed
in
a
memo
by
Frank
Hernandez,
July
21,
1998.
The
value
of
2.5
million
pounds
in
the
text
should
be
qualified
with
the
additional
information
on
the
fact
that
it
is
an
average
for
usage
over
1987
to
1996
and
is
not
a
value
for
a
single
year.

Page:
7
Paragraph:
1
Line:
3­
4
15
EPA
comment:
Carbaryl
is
also
used
extensively
for
residential
and
other
non­
agricultural
uses,
being
the
second
most
commonly
insecticide
(sic)
used
in
the
home.
Aventis'
response:
Carbaryl
is
not
registered
for
use
inside
homes.
It
is
registered
for
use
outdoors
in
the
lawn
and
garden
around
homes.
In
addition,
an
evaluation
of
the
Vista
(Triad)
data
for
the
last
seasonal
year
from
October
1999
to
September
2000
shows
retail
sales
for
carbaryl
at
18.7
million
dollars.
Carbaryl
is
listed
as
number
7
based
on
retail
sales
behind
other
active
ingredients
such
as
chlorpyrifos,
diazinon,
imidacloprid,
hydramethylnon
and
tralomethrin.
Therefore
this
sentence
would
be
more
appropriately
worded
as:
"Carbaryl
is
also
used
for
residential
and
other
non­
agricultural
uses,
being
the
seventh
most
commonly
used
insecticide
around
the
home."

Page:
7
Figure
1
EPA
comment:
Figure
2
Aventis'
response:
This
is
labeled
as
Figure
2
when
it
is
Figure
1
3.0
Integrated
Risk
Characterization
Introduction
Page:
8
Paragraph:
1
Line:
last
EPA
comment:
Carbaryl
is
mobile
to
very
mobile
in
the
environment
(Kf
=1.7
to
3.2).
Aventis'
response:
The
upper
value
Kf
for
carbaryl
should
be
listed
as
3.5
as
referenced
by
EPA
elsewhere
(e.
g.
Table
3,
page
20)
in
the
document.
The
classification
of
carbaryl
as
mobile
to
very
mobile
is
inconsistent
with
measured
Koc
values
of
177
to
249.
According
to
the
widely
used
classification
scheme
of
McCall
et
al.
carbaryl
would
be
classified
as
having
medium
mobility
in
soil.
This
classification
of
medium
mobility
is
further
supported
by
the
acceptable
column
leaching
study
(MRID
43320701)
in
which
carbaryl
residues
were
only
slightly
mobile
in
a
number
of
soils.

Aquatic
Organisms
Page:
10,
Paragraph:
1,
Line:
13
EPA
comment:
Submission
of
a
FETOX
amphibian
toxicity
study
is
encouraged.
Aventis'
response:
16
The
data
requirement
should
be
deleted.
From
the
published
results
it
is
evident
that
carbaryl
is
practically
non­
toxic
to
the
bullfrog.
Effects
in
plain
leopard
frogs
are
reported
at
levels
well
above
environmental
concentrations.
These
results
were
obtained
testing
U.
S.
native
species.
In
the
proposed
FETOX
assay
a
non­
native
species
Xenopus
laevis
is
used.
This
African
species
is
unique
in
its
behavior.
Neither
the
species
nor
the
test
methods
are
suitable
for
ecotoxicological
purposes.
As
the
risk
to
amphibians
can
be
evaluated
from
the
studies
cited,
and
as
the
effects
are
only
at
levels
well
above
the
EEC,
this
study
should
not
be
required.

Page:
10,
Paragraph:
3,
Line:
6/
7
EPA
comment:
…resulting
in
a
temporary
impairment
of
burying
behavior
and
increasing
exposure
to
predators.
Aventis'
response:
A
reference
for
this
statement
should
be
added.

Page:
11,
Paragraph:
2,
Line:
7
EPA
comment:
In
a
mesocosms
study,
at
carbaryl…
Aventis'
response:
Typographical
error.
Change
to
"In
a
mesocosm
study,
at
carbaryl…"

Terrestrial
Organisms
Page:
12
Paragraph:
2
EPA
comment:
(use
of
rock
dove
LD50
)
Aventis'
response:
The
reference
cited
for
this
value
in
Table
1
of
Appendix
D
is
currently
not
available
to
Aventis.
Table
1
of
Appendix
D
gives
a
range
of
1000
–
3000
mg/
kg
for
the
LD50.
It
should
be
assured
that
1000
is
indeed
the
correct
value.

Page:
12
Paragraph:
3
Line:
3
­
6
EPA
comment:
On
a
chronic
basis,
the
NOAEC
is
300
ppm
for
the
mallard
duck,
based
on
adverse
reproduction
effects,
including
reduced
egg
production,
decreased
fertility,
increase
incidence
of
cracked
eggs,
increased
embryonic
mortality,
and
reduced
hatching
success.
Aventis'
response:
The
sentence
should
be
changed.
The
embryonic
mortality
and
the
hatching
success
were
not
different
from
the
control.
17
Page:
13
Paragraph:
1
Line:
1
EPA
comment:
…(
rat
LD50
=
307
mg/
kg)
Aventis'
response:
Typographical
error,
the
LD50
is
301
mg/
kg.

Page:
13
Paragraph:
1
Line:
2
–
4
EPA
comment:
…based
on
decreased
fetal
body
weights
and
increased
incomplete
ossification
of
multiple
bones
in
the
laboratory
rat
(LOAEC
=
600
ppm,
NOAEC
=
80
ppm),
has
the
potential
for
mammalian
chronic
effects.
Aventis'
response:
A
new
chronic
reproduction
study
in
rats
has
been
submitted
by
Aventis.
This
study
is
more
relevant
for
an
ecological
risk
assessment
than
the
developmental
study
cited.
The
new
study
resulted
in
a
NOAEC
of
75
ppm.

Page:
13
Paragraph:
3
Line:
1
EPA
comment:
Information
available
in
the
open
literature
suggests
potential
reproduction
effects
of
carbaryl
on
mammals.
Aventis'
response:
The
sentence
should
be
changed
or
deleted.
The
literature
cited
in
the
paragraph
show
ambivalent
results.
While
some
references
seem
to
support
that
sentence,
other
references
do
not
substantiate
such
a
claim.
The
potential
for
reproductive
effects
in
mammals
is
evaluated
in
the
recently
submitted
2­
generation
study
in
rats.
No
reproductive
effects
were
seen
in
that
guideline
study.
The
NOAEC
of
75
ppm
was
based
on
pup
mortality.

Page:
13
Paragraph:
4
Line:
5
EPA
comment:
According
to
surveys
conducted
by
the
American
Beekeeping
Federation
and
the
Washington
State
Department
of
Agriculture,
carbaryl
is
one
of
the
pesticides
most
frequently
mentioned
as
being
associated
with
bee
kills.
Aventis'
response:
A
reference
should
be
provided
for
this
statement.
18
Page:
14
Paragraph:
4
Line:
1­
4
EPA
comment:
The
uses
of
carbaryl
on
crops
(corn,
cotton,
soybeans,
sorghum,
wheat,
barley,
oats,
and
rye),
forests
and
pasture/
rangeland
were
addressed
by
the
US
Fish
and
Wildlife
Service
(USFWS)
in
the
reinitiation
of
consultation
in
September
1989.
The
Service
found
jeopardy
to
a
total
of
86
species
–
6
amphibians,
47
freshwater
fish,
27
freshwater
mussels,
and
5
aquatic
crustaceans.
Aventis'
response:
The
use
of
carbaryl
on
barley,
oats,
rye,
and
cotton
has
been
cancelled.
It
should
be
noted
that
all
Aventis
CropScience
labels
for
the
technical
materials
and
the
end­
use
products
containing
carbaryl
were
amended
to
delete
these
uses.
The
Agency
has
already
approved
the
labeling
changes.
Findings
from
the
assessment
made
by
the
USFWS
should
be
reevaluated
considering
the
cancellation
of
the
use
on
barley,
oats,
rye,
and
cotton.

Page:
14
Paragraph:
5
Line:
7
EPA
comment:
The
RPAs
and
RPMs
in
the
1989
B.
O.
may
need
to
be
reassessed…
Aventis'
response:
The
acronyms
used
should
be
explained.

Endocrine
Disruption
Concerns
Page:
15
Paragraph:
3
EPA
comment:
(Report
on
potential
endocrine
effects)
Aventis'
response:
The
paragraph
should
be
deleted.
As
EPA
pointed
out,
the
findings
reported
in
the
literature
were
made
at
concentrations
well
above
the
highest
peak
concentration
modeled.
Therefore
these
findings
are
irrelevant
for
a
risk
assessment
and
at
the
current
stage
of
discussion
about
endocrine
disruption.
If
the
concern
about
the
endocrine
potential
of
carbaryl
persists,
the
issue
should
be
revisited
once
the
Agency's
endocrine
disrupter
screening
and
testing
program,
as
well
as
a
policy
on
how
to
incorporate
positive
findings
into
an
ecological
risk
assessment
have
been
fully
developed.

Page:
15
Paragraph:
4
EPA
comment:
Furthermore,
a
number
of
field
and
laboratory
studies
report
reproduction
effects
with
mammals,
suggesting
that
the
possibility
of
endocrine
disruption
effects
on
wild
mammals
should
be
further
examined.
19
Aventis'
response:
The
statement
should
be
deleted
or
modified.
As
pointed
out
above,
reports
on
reproductive
effects
of
carbaryl
in
the
open
literature
are
at
least
ambivalent.
The
recently
submitted
2­
generation
study
in
rats
demonstrated
the
absence
of
reproductive
effects.
If
the
general
statement
about
the
potential
for
endocrine
disruption
of
carbaryl
is
maintained,
references
(or
a
cross­
reference
within
the
document)
for
the
above
claim
should
be
provided.

Uncertainties
Page:
15
Paragraph:
Last
Line:
4
EPA
comment:
In
the
absence
of
a
valid
two­
generation
rat
reproduction
study,
mammalian
chronic
RQs
were
based
on
a
rat
prenatal
development
study
NOAEC
(MRID#
44732901).
Aventis'
response:
A
new
two­
generation
study
in
rats
was
recently
submitted.
20
4.0
Environmental
Fate
Assessment
Exposure
Characterization
Page:
16
Paragraph:
3
Line:
8
EPA
comment:
Environment
(Kf
=1.7
to
3.2).
Aventis'
response:
The
upper
value
Kf
for
carbaryl
should
be
listed
as
3.5
as
referenced
by
EPA
elsewhere
in
the
document
(e.
g.
Table
3,
page
20).

Page:
16
Paragraph:
3
Line:
last
sentence
EPA
comment:
Detailed
discussion
and
reviews
(DERs)
of
the
studies
that
are
included
in
this
assessment
are
attached
in
Appendix
A.
Aventis'
response:
It
is
inappropriate
to
include
the
DERs
in
the
RED.
A
summary
of
study
findings
is
already
included
in
the
EFED
Chapter.
DERs
should
be
made
available
to
the
public
through
the
regular
procedure
under
the
Freedom
of
Information
Act
after
they
have
been
reviewed
and
cleared
for
confidential
business
information.

Page:
16
Paragraph:
4
Line:
4
EPA
comment:
lower
levels
(generally
less
than
0.01
:
/L).
Aventis'
response:
value
missing
units
­
(generally
less
than
0.01
:
g/
L).

Page:
16
Paragraph:
5
Line:
4
EPA
comment:
…monitoring
data
is
of
limited
utility
in
developing
EECs
for
ecological
and
human
health
risk
assessment.
Aventis'
response:
The
drinking
water
monitoring
program
conducted
by
the
registrant
provides
a
real
world
assessment
of
the
potential
for
human
exposure
to
carbaryl
in
drinking
water
derived
from
surface
water.
Drinking
water
concentrations
derived
from
PRZM/
EXAMS
greatly
overestimate
the
potential
exposure
to
carbaryl
in
drinking
water,
generally
by
several
orders
of
magnitude.
21
Page:
17
Paragraph:
1
Line:
2­
3
EPA
comment:
The
maximum
rate
was
taken
from
the
carbaryl
labels.
Aventis'
response:
It
would
be
of
benefit
for
the
Agency
to
be
explicit
and
list
the
carbaryl
labels
that
were
used
to
develop
the
maximum
application
rates
for
the
model
scenarios.
The
reference
cited
in
the
EFED
Chapter
regarding
the
use
of
carbaryl
on
crops
indicate
that
current
labels
were
not
used
for
the
Agency's
assessment.
Many
of
these
crops
have
been
deleted
from
Aventis'
labels
for
a
few
years
Application
rates,
number
of
applications
per
season,
and
PHI's
also
have
changed
for
several
crops
on
the
labels.

Page:
17
Paragraph:
2
Line:
2
EPA
comment:
For
the
Index
Reservoir
scenario
using
maximum
label
rates,
acute
EEC
values
ranged
from
about
10
:
g/
L
from
sugar
beets
to
about
500
:
g/
L
from
citrus
(Table
6).
Aventis'
response:
Table
6
on
page
33
shows
a
concentration
of
19
:
g/
L
for
sugar
beets
treated
with
the
maximum
label
rate
of
2
x
1.5
lb
ai,
not
10
as
stated
in
this
sentence.
A
low
EEC
value
of
9
:
g/
L
for
sugar
beets
results
from
the
"maximum
reported"
application
scenario
of
1
x
1.2
lb
ai/
A.

Page:
17
Paragraph:
2
Line:
3
EPA
comment:
Chronic
EECs
ranged
from
about
1
to
28
:
g/
L.
Aventis'
response:
Table
6
on
page
33
shows
that
this
is
correct
when
considering
all
of
the
model
scenarios.
However,
either
the
same
maximum
label
rate
reference
should
be
used
as
in
the
preceding
sentence
(in
which
case
the
minimum
chronic
EEC
would
be
2),
or
the
basis
for
the
preceding
sentence
should
be
changed
from
the
maximum
label
rate
to
include
all
application
scenarios
to
keep
the
comparisons
consistent.

Page:
17
Paragraph:
2
Line:
8
EPA
comment:
The
results
of
the
modeling
provide
an
(sic)
conservative,
though
not
unreasonable,
estimate
on
(sic)
possible
concentrations
[in]
drinking
water.
Aventis'
response:
It
should
be
clear
that
Aventis'
surface
water
monitoring
program
provides
a
more
reasonable
estimate
of
the
potential
drinking
water
exposure
to
carbaryl
than
the
modeling
numbers,
which
overestimate
exposure
by
several
orders
of
magnitude.
22
Page:
17
Paragraph:
2
Line:
last
EPA
comment:
…and
model
input
and
output
files
are
attached
in
appendix
B.
Aventis'
response:
The
PRZM
input
files
for
only
the
Index
Reservoir
drinking
water
modeling
were
provided
as
an
electronic
copy.
The
PRZM
input
files
for
the
standard
pond
scenarios
were
not
provided
in
the
draft
RED
so
Aventis
could
not
assess
the
data.
None
of
the
output
files
were
provided.

Page:
18
Figure
2
EPA
comment:
Figure
1.
Generalized
carbaryl
degradation
pathway
Aventis'
response:
This
should
be
labeled
Figure
2,
not
Figure
1.

Page:
19
Table
3
EPA
comment:
Hydrolysis
half­
life
at
pH
9
stated
to
be
5
hours.
Aventis'
response:
The
study
results,
and
the
summary
of
the
study
presented
on
page
20,
show
the
correct
half­
life
at
pH
9
to
be
3.2
hours.

Page:
19
Table
3
EPA
comment:
Aerobic
Aquatic
half­
life
­
4.9.
Aventis'
response:
The
Aerobic
Aquatic
half­
life
is
4.9
days
Page:
19
Table
3
EPA
comment:
Soil
metabolism
T1/
2,
anaerobic,
assumed
stable
23
Aventis'
response:
If
this
guideline
is
satisfied
by
the
data
submitted
for
guideline
162­
3,
it
is
not
clear
why
the
compound
is
assumed
to
be
stable
rather
than
having
a
half­
life
in
line
with
the
72
days
that
resulted
from
the
anaerobic
aquatic
study.
Although
this
parameter
plays
a
fairly
insignificant
role
in
estimating
the
amount
of
carbaryl
available
for
runoff
in
the
models,
it
could
play
a
significant
role
if
one
were
to
use
this
value
in
estimating
leaching
potential
in
subsurface
horizons.

Page:
20
Table
3
EPA
comment:
Batch
Equilibrium
1/
n
values
ranged
from
0.86­
1.02
Aventis'
response:
These
values
are
for
the
desorption
isotherms
only.
For
the
adsorption
isotherms
that
were
used
to
calculate
the
adsorption
Kf
and
Koc
values
listed
in
the
table,
the
correct
range
of
1/
n
values
are
0.78
to
0.84
as
stated
on
page
22.

Page:
20
Table
3
EPA
comment:
Foliar
Dissipation
30
days
Willis
and
McDowell,
1987
Aventis'
response:
The
foliar
dissipation
half­
life
listed
by
EFED
is
incorrect.
Table
IV
of
the
Willis
and
McDowell
review
lists
10
foliar
half­
lives
for
various
formulations
of
carbaryl
applied
to
different
crops.
Five
of
these
half­
lives
are
for
a
study
designed
to
evaluate
a
new
analytical
procedure
for
measuring
carbaryl
residues
on
plants.
This
study
was
conducted
on
plants
grown
in
a
greenhouse,
with
some
of
them
receiving
an
unknown
amount
of
simulated
rainfall.
These
studies
on
greenhouse­
grown
plants
should
not
be
used
to
evaluate
foliar
persistence
in
the
field.
The
foliar
persistence
of
pesticides
can
be
considerably
different
for
residues
on
and
in
plants
grown
in
greenhouses
versus
the
field.
Eliminating
the
half­
lives
for
the
greenhouse­
grown
plants
results
in
the
following
half­
lives
for
carbaryl
on
field
plants:
Cotton,
1.2,
1.3,
1.5
days;
strawberry,
4.1
days;
tomato
1.4
days.
Therefore,
the
longest
half­
life
of
4.1
days
should
be
listed
in
this
table.

Aventis
intends
to
conduct
a
more
thorough
review
of
the
data
on
the
foliar
dissipation
of
carbaryl
and
prepare
a
more
detailed
response
during
the
60­
day
public
comment
period.
24
Persistence
Microbially­
Mediated
Processes
Page:
21
Paragraph:
3
Line:
3
EPA
comment:
with
an
initial
concentration
of
11.2
mg/
L,
degraded
with
a
half­
life
of
4.0
days
in
sandy
Aventis'
response:
The
units
for
ppm
soil
concentration
should
be
given
as
mg/
kg.

Page:
21
Paragraph:
3
Line:
4­
5
EPA
comment:
The
major
degradate
was
1­
naphthol
which
further
degraded
rapidly
to
non­
detectable
levels
within
14
days.
Aventis'
response:
The
data
from
this
study
demonstrate
that
under
aerobic
soil
conditions
the
formation
and
decline
of
1­
naphthol,
starting
from
parent
carbaryl
is
complete
in
less
than
14
days.
The
study
data
show
an
average
maximum
1­
naphthol
level
of
34.5%
of
applied
carbaryl
by
day
1,
declining
to
2.8%
by
day
2,
0%
by
day
4,
0.2%
by
day
7
and
0%
at
day
14.
These
data
suggest
a
preliminary
half­
life
of
less
than
1
day
for
the
major
degradate
1­
naphthol.

Page:
21
Paragraph:
3
Line:
8­
9
EPA
comment:
In
anaerobic
aquatic
soil
carbaryl
with
an
about
10
mg/
L
degraded
with
a
half­
life
of
72.2
days.
Aventis'
response:
Several
words
appear
to
be
missing
from
this
sentence.
One
suggestion:
"Carbaryl
degraded
with
a
half­
life
of
72.2
days
in
anaerobic
aquatic
sediment
with
an
initial
carbaryl
concentration
of
about
10
mg/
L."

Page:
22
Paragraph
carried
over
from
page
21
Line:
4
on
pg
22
EPA
comment:
Chudhry
and
Wheeler,
1988
Aventis'
response:
This
reference
is
not
included
in
the
reference
list
25
Mobility
Page:
22
Paragraph:
1
Line:
1
EPA
comment:
Carbaryl
is
considered
to
be
mobile
to
very
mobile
in
soils.
Aventis'
response:
See
response
directly
below.

Page:
22
Paragraph:
3
Line:
1­
2
EPA
comment:
Based
on
batch
equilibrium
experiments
(MRID
43259301)
carbaryl
was
determined
to
be
very
mobile
to
mobile
in
soils.
Aventis'
response:
The
classification
of
carbaryl
as
mobile
to
very
mobile
is
inconsistent
with
measured
Koc
values
of
177
to
249.
According
to
the
widely
used
classification
scheme
of
McCall,
et
al.
(1980)
wherein
Koc
values
between
150
and
500
denote
medium
mobility
in
soil,
carbaryl
would
be
classified
as
having
medium
mobility
in
most
soils.
This
classification
of
medium
mobility
is
further
supported
by
the
acceptable
column
leaching
study
(MRID
43320701)
in
which
aged
carbaryl
residues
were
only
slightly
mobile
in
a
number
of
soils.
The
mobility
of
carbaryl
would
be
expected
to
be
higher
in
sandy
soils
or
in
soils
of
low
organic
matter.

Field
Dissipation
Page:
22
Paragraph:
5
Line:
3
EPA
comment:
The
submitted
field
and
aquatic
dissipation
studies
were
determined
to
be
unacceptable,
and
did
not
provide
useful
information
on
movement
and
dissipation
of
carbaryl
or
its
degradation
products.
Aventis'
response:
The
field
dissipation
study
(MRID
41982605)
submitted
in
1991
demonstrated
that
carbaryl
dissipated
very
rapidly
(t1/
2
<
1
week)
with
no
measurable
leaching.
The
study
included
two
sites,
one
in
North
Carolina
and
one
in
California.
At
the
North
Carolina
site,
~
95%
of
the
Time
0
residues
had
dissipated
by
the
first
sampling
period
7
days
after
application
(the
planned
first
sampling
at
3
days
was
not
collected
due
to
rain).
Similarly,
~
85%
of
the
Time
0
residues
had
dissipated
by
7
days
after
application
at
the
California
site.
Concerning
the
movement
of
carbaryl,
samples
were
taken
to
a
depth
of
0.9
meters
in
increments
of
0.15
meters.
No
residues
were
found
below
the
upper
0.15
meters.
26
Page:
23
Paragraph:
3
Line:
2
EPA
comment:
Because
of
inappropriate
sampling
intervals,
poor
sample
storage
stability,
lack
of
degradate
monitoring,
rainfall
and
irrigation
that
were
less
than
evapotranspiration,
and
irrigation
water
with
high
pH,
these
studies
do
not
provide
reliable
information
on
the
rate
of
dissipation
of
parent
carbaryl
or
formation
of
degradation
products.

Aventis'
response:
 
The
estimated
half­
life
determined
from
this
study
was
<
3
days.
Sampling
at
intervals
such
that
several
sampling
events
are
taken
prior
to
the
half­
life
of
the
product
is
impractical
for
rapidly
degrading
chemicals
(e.
g.,
those
with
half­
lives
less
than
a
week).
For
this
rapidly
degrading
chemical
an
estimate
of
the
half­
life
should
be
sufficient
for
risk
assessments
even
if
it
is
not
precise.
 
After
the
report
was
submitted
to
California,
the
freezer
storage
stability
recoveries
at
six
and
nine
months
were
measured
but
not
reported.
Rainfall
plus
irrigation
approximated
an
inch
a
week
and
was
more
than
enough
to
maintain
a
good
soil
moisture
for
agricultural
purposes.
 
Sulfuric
acid
is
routinely
added
to
irrigation
water
in
the
region
of
California
where
the
field
test
was
conducted
to
neutralize
the
water's
high
pH.
Although
not
stated
in
the
report,
the
irrigation
water
in
the
California
trial
was
treated
in
the
typical
commercial
fashion.
The
acid
is
injected
into
the
irrigation
pipe
as
water
is
pumped
through
it.
Unfortunately,
the
pH
of
the
water
arriving
at
the
field
after
treatment
was
not
measured.

Aquatic
Field
Dissipation
Page:
24
Paragraph:
2
Line:
3
EPA
comment:
They
(do)
not
provide
useable
information
on
the
dissipation
of
carbaryl
and
1­
naphthol
in
aquatic
field
conditions.
Aventis'
response:
 
The
soil
metabolism
study
referred
to
in
the
report
found
that
the
total
water
soluble
metabolites
did
not
exceed
5%
of
the
total
radioactive
residue,
the
primary
hydrolysis
product,
1­
naphthol,
was
not
found,
and
that
the
only
analyte
of
concern
was
the
parent
insecticide,
carbaryl.
A
soil
metabolism
study
reviewed
concurrently
by
the
Agency
was
issued
later
(MRID
42785101,
classified
"acceptable")
with
similar
results.
Although
the
major
soil
metabolite,
1­
naphthol,
was
found
at
significant
levels
at
day
0
and
day
1,
the
levels
were
less
than
0.7%
by
day
4
and
non­
detectable
by
day
14.
Two
other
metabolites
were
identified
but
never
exceeded
levels
of
1.7%
of
the
total
residue.
Again
the
only
residue
of
concern
was
the
parent
insecticide,
carbaryl.

 
If
present,
1­
naphthol
would
have
been
detected
by
the
residue
method
used
to
measure
the
residues
of
carbaryl
in
the
soil.
27
 
The
estimated
half­
life
determined
from
this
study
was
<
2
days.
Sampling
at
intervals
such
that
several
sampling
events
are
taken
prior
to
the
half­
life
of
the
product
is
impractical
for
rapidly
degrading
chemicals
(e.
g.
those
with
half­
lives
less
than
a
week).

Page:
24
Paragraph:
2
Line:
4
EPA
comment:
Frozen
storage
stability
data
were
provided
for
only
6
months,
although
the
water
samples
were
stored
for
up
to
14
months
and
the
soil
samples
were
stored
for
up
to
17.5
months
prior
to
analysis.
The
data
suggest
that
carbaryl
and
1­
naphthol
degraded
significantly
during
storage.
In
the
six
months
of
storage
carbaryl
degraded
an
average
of
34
%
in
Texas
water
and
39%
in
from
Mississippi.
1­
naphthol
degraded
50%
in
water
from
Texas
and
69%
from
Mississippi.
Degradation
did
not
appear
linear,
and
it
is
not
possible
to
extrapolate
out
to
14
months.
It
was
therefore
not
possible
to
evaluate
the
actual
concentrations
of
carbaryl
and
1­
naphthol
in
the
samples
or
estimate
the
dissipation
rates.
Aventis'
response:
The
existing
6­
month
storage
stability
provides
sufficient
information
to
calculate
the
concentrations
of
carbaryl
in
the
samples.
However,
the
metabolite
1­
naphthol
was
shown
to
degrade
significantly
under
the
same
freezer
conditions.
This
instability
simply
confirms
that
1­
naphthol's
presence
in
the
environment
would
be
very
limited
and
should
not
be
of
concern.

Foliar
Dissipation
Page:
24
Paragraph:
Last
EPA
comment:
The
reported
rates
of
carbaryl
dissipation
from
foliar
surfaces
varies
from
1
days
to
30
days.
In
their
review
of
literature
data
on
pesticide
foliar
persistence,
Willis
and
McDowell
(1987)
report
that
carbaryl
dissipation
rates
varied
from
1.2
to
29.5
days…
For
terrestrial
risk
assessment
modeling
EFED
used
35
days…
Aventis'
response:
As
stated
in
comments
to
Table
3,
the
foliar
dissipation
half­
life
used
by
EFED
for
terrestrial
risk
assessment
is
too
long
and
should
be
corrected.
Table
IV
of
the
Willis
and
McDowell
review
lists
10
foliar
half­
lives
for
various
formulations
of
carbaryl
applied
to
different
crops.
Five
of
these
half­
lives
are
for
a
study
designed
to
evaluate
a
new
analytical
procedure
for
measuring
carbaryl
residues
on
plants.
This
study
was
conducted
on
plants
grown
in
a
greenhouse,
with
some
of
them
receiving
an
unknown
amount
of
simulated
rainfall.
These
studies
on
greenhouse­
grown
plants
should
not
be
used
to
evaluate
foliar
persistence
in
the
field.
The
foliar
persistence
of
pesticides
can
be
considerably
different
for
residues
on
and
in
plants
grown
in
greenhouses
versus
the
field.
Eliminating
the
half­
lives
for
the
greenhouse­
grown
plants
results
in
the
following
28
half­
lives
for
carbaryl
on
field
plants:
Cotton,
1.2,
1.3,
1.5
days;
strawberry,
4.1
days;
tomato
1.4
days.
Therefore,
the
longest
half­
life
of
4.1
days
should
be
used
for
terrestrial
risk
assessment
modeling.

Aventis
will
conduct
a
more
thorough
review
of
the
data
on
the
foliar
dissipation
of
carbaryl
and
prepare
a
more
detailed
response
during
the
60­
day
public
comment
period.

Atmospheric
Transport
Page:
25
Paragraph:
1
Line:
2
EPA
comment:
Waite,
et
al.,
1995
Aventis'
response:
This
reference
is
not
included
in
the
reference
list
Page:
25
Paragraph:
1
Line:
3
EPA
comment:
Beyer
et
al.,
(1995)
Aventis'
response:
This
reference
is
not
included
in
the
reference
list
Page:
25
Paragraph:
3
Line:
5
EPA
comment:
Schomburg
et
al.
(1991)
Aventis'
response:
This
reference
is
not
included
in
the
reference
list
1­
Naphthol
Fate
and
Transport
Page:
26
Paragraph:
2
Line:
1­
2
EPA
comment:
In
an
aerobic
soil
metabolism
study
(MRID
42785101),
1­
naphthol
degraded
rapidly
to
non­
detectable
levels
within
14
days.
29
Aventis'
response:
The
data
from
this
study
demonstrate
that
under
aerobic
soil
conditions
the
formation
and
decline
of
1­
naphthol,
starting
from
parent
carbaryl,
is
complete
in
less
than
14
days.
The
study
data
show
an
average
maximum
1­
naphthol
level
of
34.5%
of
applied
carbaryl
by
day
1,
declining
to
2.8%
by
day
2,
0%
by
day
4,
0.2%
by
day
7
and
0%
at
day
14.
These
data
suggest
a
preliminary
half­
life
of
less
than
1
day
for
the
major
degradate
1­
naphthol.
This
half­
life
can
be
used
for
preliminary
environmental
fate
modeling
to
estimate
EECs
for
1­
naphthol.

Page:
26
Paragraph:
3
Line:
1
EPA
comment:
No
guideline
information
was
submitted
on
1­
naphthol
sorption.
Literature
information
suggests
that
it
is
not
strongly
sorbed.
Aventis'
response:
The
statement
suggesting
that
1­
naphthol
is
not
strongly
sorbed
to
soil
should
be
deleted.
In
support
of
the
1­
naphthol
sorption
statement
the
Agency
has
cited
only
one
paper
by
Karthikeyan
et
al.
(1999)
that
was
conducted
using
aluminum
hydroxide
as
the
sorbent.
Soil
is
composed
of
much
more
than
aluminum
hydroxide,
so
this
study
is
more
of
a
mechanistic
description
of
sorption
to
this
one
component
of
soil
and
not
a
study
of
sorption
to
soil
as
a
whole.
This
cited
study
reported
that
1­
naphthol
does
not
show
significant
sorption
to
aluminum
hydroxide
when
allowed
to
sorb
for
20
hours
in
the
dark
in
the
absence
of
oxygen.
However,
there
was
a
significant
increase
in
sorption
with
increasing
equilibration
time,
and
as
the
Agency
stated,
the
increase
is
influenced
by
pH,
as
would
be
expected
for
an
acidic
phenolic
compound.

Additional
information
available
in
the
literature
demonstrates
that
the
sorption
of
1­
naphthol
to
soil
is
stronger
than
that
seen
for
carbaryl
itself.
Hassett
et
al.
(1981)
have
demonstrated
that
the
sorption
of
1­
naphthol
was
the
result
of
sorption
to
organic
carbon
resulting
in
an
average
Koc
of
431
±
40
for
10
of
the
16
soil
samples
they
tested.
In
the
remaining
6
soil
samples
the
Koc
was
even
higher
(1,645
to
15,618).
Hassett
et
al.
(reference
submitted
as
part
of
30­
day
response
document)
hypothesized
that
the
higher
Kocs
in
these
6
soils,
in
which
the
organic
carbon
to
clay
ratio
was
very
low,
the
clay
surfaces
were
more
accessible
and
the
sorption
of
1­
naphthol
was
apparently
controlled
by
the
clay
fraction.
In
Burgos
et
al.
(1999),
cited
by
EPA
elsewhere
in
the
RED,
it
was
shown
that
there
is
significant
sorption
of
1­
naphthol
to
two
sandy
soils,
and
that
oxidative
coupling
reactions
were
responsible
for
the
strongly
bound
portion.
In
an
earlier
paper
by
Burgos
et
al.
(1996)
it
was
shown
that
both
biologically­
mediated
and
soil­
catalyzed
oxidative
coupling
lead
to
significant
binding
of
1­
naphthol
residues
to
soil.
These
data
indicate
that
1­
naphthol
is
less
mobile
and
less
susceptible
to
leaching
than
carbaryl
itself,
and
they
demonstrate
that
at
least
a
portion
of
the
1­
naphthol
residue
is
tightly
sorbed
to
soil
constituents.

To
meet
the
requirement
by
the
Agency
for
information
on
the
adsorption
and
desorption
of
1­
naphthol,
the
registrant
is
conducting
an
adsorption/
desorption
study
to
meet
the
30
163­
1
guideline.
Study
results
should
be
available
for
submission
to
the
Agency
in
the
first
quarter
of
the
calendar
year
2002.

Aquatic
Exposure
Assessment
Surface
Water
Page:
26
Paragraph
4
Line
1
EPA
comment:
Five
crop
scenarios:
apples,
field
corn,
sweet
corn,
oranges
and
sweet
potatoes
scenarios
were
use
in
modeling
for
surface
water
EEC.
Aventis'
response:
The
fifth
crop
modeled
was
sugar
beets
(not
sweet
potatoes).

Page:
27
Table
4
EPA
comment:
Hydrolysis
half­
life
at
pH
9
stated
to
be
5
hours.
Aventis'
response:
The
study
results,
and
the
summary
of
the
study
presented
on
page
20,
show
the
correct
half­
life
at
pH
9
to
be
3.2
hours.

Page:
27
Table
4
EPA
comment:
(Koc
=
211
for
SCIGROW)
Aventis'
response:
This
is
the
mean
Koc.
According
to
EPA
guidance
the
median
Koc
(209)
should
be
used
for
SCI­
GROW,
although
this
difference
would
not
be
expected
to
affect
the
model
results.

Pages:
27­
28
Table
5
EPA
comment:
Tier
II
surface
water
estimated
environmental
concentration
(EEC)
values
derived
from
PRZM/
EXAMS
modeling
for
use
in
ecorisk
assessment
(calculated
using
standard
pond.)
Aventis'
response:
The
PRZM
input
tables
were
not
provided
for
the
standard
pond
scenarios,
so
the
assumption
is
made
that
the
same
application
methods
were
used
for
the
standard
pond
as
for
the
Index
Reservoir
scenarios
that
were
provided
as
an
electronic
copy
of
a
draft
of
Appendix
B.
31
It
would
be
of
benefit
for
the
Agency
to
state
which
of
the
carbaryl
labels
were
used
to
develop
the
"maximum"
label
application
rate
scenarios.
It
would
be
useful
to
add
another
column
to
this
table
to
specify
which
method
of
application
was
used
to
generate
the
EECs
rather
than
the
generic
"air/
ground"
in
column
1.
There
are
a
number
of
errors
in
the
input
parameters
(noted
below)
that
would
lead
to
changes
in
the
calculated
EECs
and
therefore
the
risk
quotients
for
these
uses.

If
the
modeling
for
the
"average"
scenarios
were
conducted
using
aerial
applications
for
citrus
and
apples
(as
was
the
case
for
the
Index
Reservoir
scenarios),
then
the
model
results
over­
estimate
the
contributions
from
spray
drift.
Few
applications
to
these
crops
are
made
aerially.
Therefore,
the
model
results
over­
estimate
the
contributions
from
spray
drift
since
the
"average"
applications
to
these
crops
are
made
using
ground
airblast
equipment
with
a
spray
drift
of
6.3%
in
the
model
versus
aerial
applications
with
a
spray
drift
of
16%.

The
"average"
scenario
for
sweet
corn
in
Ohio
should
be
3
applications
at
1.1
lb.
ai/
A/
application
(as
noted
in
the
memo,
"Average
application
rate
from
Quantitative
Usage
Analysis
for
Carbaryl,
prepared
July
21,
1998
by
Frank
Hernandez,
OPP/
BEAD")
and
not
the
2
applications
at
3.4
lb.
ai/
A/
application
as
listed
in
the
table.
It
should
be
noted
that
the
"average"
scenario
presented
in
this
table,
2
applications
per
year
at
3.4
lb.
ai
per
application,
exceed
the
maximum
rate
allowed
on
the
label.

The
maximum
label
rate
application
scenario
for
apples
that
is
allowed
by
the
Sevin
brand
XLR
PLUS
label
(E.
P.
A.
Reg.
No
264­
333),
the
Sevin
brand
80WSP
and
CHIPCO
Sevin
brand
80WSP
labels
(E.
P.
A.
Reg.
No
264­
526)
and
the
CHIPCO
Sevin
brand
SL
label
(E.
P.
A.
Reg.
No
264­
335)
is
5
applications
at
3
lb.
ai/
A/
application
made
every
14
days.
The
scenario
used
in
the
model
applies
less
than
the
maximum
amount
of
product
allowed
by
the
labels.
In
addition,
if
the
same
application
timing
was
used
in
the
modeling
for
the
standard
pond
scenario
as
was
used
in
the
index
reservoir
scenario
(applications
made
by
air
every
4
days)
this
would
be
a
violation
of
the
Aventis
labels
which
restrict
applications
to
a
minimum
of
every
14
days.

The
"average"
scenario
for
sugar
beets
in
Minnesota
should
be
1
application
at
1.3
lb.
ai/
A/
application
(as
noted
in
the
memo,
"Average
application
rate
from
Quantitative
Usage
Analysis
for
Carbaryl,
prepared
July
21,
1998
by
Frank
Hernandez,
OPP/
BEAD")
and
not
1
application
at
1.5
lb.
ai/
A/
application
as
listed
in
the
table.

The
"Citrus"
scenario
would
be
more
appropriately
labeled
Oranges.
For
the
average
scenario,
the
3.4
lb.
ai/
A/
application
rate
listed
in
Table
5
is
for
oranges
(as
noted
in
the
memo,
"Average
application
rate
from
Quantitative
Usage
Analysis
for
Carbaryl,
prepared
July
21,
1998
by
Frank
Hernandez,
OPP/
BEAD"),
which
is
the
highest
"average"
application
rate
for
any
type
of
citrus.
Therefore,
this
"average"
scenario
for
oranges
are
at
the
high
end
for
all
citrus
and
overestimates
the
PRZM/
EXAMS
derived
EECs
for
use
in
the
other
citrus
crops.
"Average"
application
rates
for
other
citrus
as
listed
in
the
memo
are:
32
Lemons
–
1.3
applications
at
2.7
lb
ai/
A/
appl
Grapefruit
–
1.6
applications
at
1.4
lb
ai/
A/
appl
Citrus,
other
–
1.8
applications
at
1.8
lb
ai/
A/
appl
The
maximum
label
application
rate
for
citrus
is
7.5
lb
ai
per
application,
not
5
lb
ai,
with
a
maximum
of
20
lb
ai
total
allowed
per
year.
In
California
only,
a
single
application
is
allowed
at
the
rate
of
5
to
16
lb
ai
per
season
for
control
of
California
red
scale
and
yellow
scale.

Estimated
Environmental
Concentrations
for
Terrestrial
Ecological
Risk
Assessment
Page:
29
Paragraph:
2
Line:
2­
4
EPA
comment:
In
the
absence
of
reliable
foliar
dissipation
data
a
dissipation
half­
life
of
35
days
is
used.
Published
literature
shows
that
carbaryl
dissipation
rates
vary,
and
are
among
the
highest
observed
for
any
pesticide
(Willis
and
McDowell,
1987).
Aventis'
response:
As
stated
in
more
detail
above,
some
of
the
foliar
dissipation
half­
lives
listed
in
this
reference
are
high
because
they
were
generated
in
the
greenhouse,
not
in
the
field,
and
therefore
they
should
not
be
used.
Eliminating
the
half­
lives
for
the
greenhouse­
grown
plants
results
in
the
following
half­
lives
for
carbaryl
on
field
plants:
Cotton,
1.2,
1.3,
1.5
days;
strawberry,
4.1
days;
tomato
1.4
days.
Therefore,
the
longest
half­
life
of
4.1
days
should
be
used
for
terrestrial
risk
assessment
modeling.

Page:
29
Paragraph:
2
Line:
6
EPA
comment:
A
more
thorough
description
of
the
model
calculations
and
ELL­
FATE
outputs
are
attached
in
Appendix
B.
Aventis'
response:
No
such
description
or
attachments
were
provided,
so
Aventis
did
not
have
the
opportunity
to
evaluate
the
model.

Page:
29
Paragraph:
2
Line:
last
EPA
comment:
…Tables
4,7,
8
and
9,
Appendix
D.
Aventis'
response:
These
tables
are
in
Appendix
C.
33
5.0
Drinking
Water
Assessment
Water
Resources
Assessment
Page:
29
Paragraph:
3
Line:
3
EPA
comment:
Carbaryl
tends
not
to
partition
to
soil,
aquifer
solids,
or
sediment.
Aventis'
response:
This
sentence
is
misleading
and
should
be
reworded.
Carbaryl
does
partition
onto
these
sorbents,
but
the
sorption
coefficients
are
not
high.
Suggest
rewording
this
such
as:
"Carbaryl
tends
not
to
bind
tightly
to
soil,
aquifer
solids,
or
sediment."

Page:
29
Paragraph:
4
EPA
comment:
Under
certain
conditions
carbaryl
can
be
expected
to
persist
in
the
environment.
Under
low
pH
conditions
the
compound
is
stable
to
hydrolysis.
In
anaerobic
environments
metabolism
is
fairly
slow
(t½
=
72
days).
This
suggests
that
carbaryl
may
leach
to
ground
water
and
persist
in
some
aquifers.
Aventis'
response:
This
last
statement
should
be
removed.
In
contrast
to
this
hypothesis
are
the
data
presented
in
the
NAWQA
and
EPA
databases
that
demonstrate
that
carbaryl
is
not
likely
to
leach
to
ground
water
and
is
not
likely
to
persist
in
aquifers.
The
fact
that
carbaryl
has
been
widely
used
in
agricultural
and
urban
settings
for
more
than
35
years,
and
yet
is
found
at
concentrations
greater
than
0.1
:
g
/L
in
only
0.027%
of
the
agricultural
wells,
urban
wells
and
aquifers
sampled
by
NAWQA
(Kolpin,
2001),
indicates
that
this
statement
has
little
merit.
Furthermore,
the
last
sentence
is
in
direct
contradiction
to
the
statement
made
at
the
beginning
of
the
preceding
paragraph
that
carbaryl
"…
has
limited
potential
to
leach
to
ground
water."

Page:
30
Paragraph:
1
Lines
1­
3
EPA
comment:
Surface
water
monitoring
studies
show
that
carbaryl
is
the
second
most
widely
detected
insecticide
after
diazinon.
Carbaryl,
at
typically
low
concentrations,
is
found
in
greater
than
20
%
of
surface
samples
at
concentrations
up
to
7
ppb.
Aventis'
response:
These
summary
statements
are
based
on
the
NAWQA
database,
with
the
exception
of
the
7
ppb
concentration.
The
highest
reported
value
in
the
NAWQA
database
is
5.5
ppb.
The
value
of
7
ppb
does
not
come
from
the
NAWQA
database
but
from
the
report
by
Werner
et
al.
(2000).
In
fact,
a
maximum
carbaryl
concentration
of
8.4
ppb
was
reported
for
surface
water
samples
in
the
California
DPR
surface
water
database
(see
discussion
section).
The
sources
of
the
information
should
not
be
mixed,
or
the
source
of
the
information
should
be
explicitly
stated.
34
Drinking
Water
Exposure
Assessment
Page:
30
Paragraph:
2
Line:
3­
4
EPA
comment:
Carbaryl
is
the
second
most
commonly
detected
insecticide
in
surface
water,
and
can
be
expected
to
contaminate
drinking
water
derived
from
surface
water
bodies.
Aventis'
response:
The
surface
water­
monitoring
program
conducted
by
Aventis
shows
an
insignificant
impact
of
carbaryl
on
drinking
water.

Page:
30
Paragraph:
2
Line:
7
EPA
comment:
The
maximum
reported
value
was
7.0
:
g
/L.
Aventis'
response:
The
maximum
value
reported
in
the
NAWQA
database
is
5.5
:
g
/L.
The
only
carbaryl
detection
reported
in
the
study
by
Werner
et
al.
(2000)
was
7.0
:
g
/L.
The
maximum
value
reported
in
the
California
DPR
Surface
Water
database
is
8.4
:
g
/L.
Since
all
of
the
statistics
made
in
this
paragraph
refer
to
the
NAWQA
data,
the
reference
to
the
maximum
reported
concentration
should
be
5.5
:
g
/L.

Page:
30
Paragraph:
4
Line:
2
EPA
comment:
Older
studies
using
GC
or
GC/
MS
generally
have
poor
recovery
and
quantitation
limits.
Because
of
this
difficulty
in
analysis
the
actual
concentration
of
carbaryl
in
groundwater
and
surface
waters
may
be
higher
than
reported.
Aventis'
response:
The
basis
for
making
this
generalization
is
not
readily
apparent
and
these
statements
should
be
removed.
Comments
regarding
the
recovery
reported
for
the
GC/
MS
method
used
in
the
NAWQA
survey
are
made
below
in
reference
to
statements
made
on
page
34
paragraph
5,
and
are
elucidated
in
the
discussion
section
at
the
end
of
this
response
document.
The
method
detection
limit
(MDL)
reported
for
the
GC/
MS
method
used
for
the
NAWQA
program
is
0.003
ppb
(Zaugg
et
al.,
1995;
Larson
et
al.
,
1999).
The
limit
of
detection
for
the
HPLC/
MS/
MS
method
used
in
the
carbaryl
surface
water
monitoring
study
being
conducted
by
the
registrant
(LOD,
0.002
ppb;
LOQ
0.030
ppb)
is
similar
to
the
GC/
MS
method
used
for
the
NAWQA
program.
In
addition
to
the
GC/
MS
method
used
in
the
NAWQA
program,
carbaryl
was
also
analyzed
by
HPLC/
photodiode­
array
detection
in
a
limited
number
of
samples
with
a
MDL
of
0.008
(Werner
et
al.,
1996).
Therefore,
the
quantification
limits
reported
for
the
GC/
MS
method
used
to
generate
a
majority
of
the
carbaryl
data
in
the
NAWQA
database
is
very
similar
to
the
quantification
limits
for
available
HPLC
methods.
See
the
discussion
section
at
the
end
35
of
this
response
document
for
a
summary
of
the
available
NAWQA
data
obtained
by
the
GC/
MS
and
HPLC/
PDA
methods.

Page:
30
Paragraph:
4
Line:
4
EPA
comment:
More
recent
studies
using
HPLC/
MS
should
provide
better
data
on
the
true
extent
and
magnitude
of
water
contamination
from
the
use
of
carbaryl.
Aventis'
response:
Aventis
believes
that
our
ongoing
targeted
surface
water­
monitoring
program
using
HPLC/
MS/
MS
accurately
reflects
the
extent
and
magnitude
of
carbaryl
exposure
in
drinking
water
derived
from
surface
water.

Drinking
Water
Modeling
Page:
31
Paragraph:
carried
over
from
page
30
Line:
8
EPA
comment:
A
partial
list
of
input
parameters
for
the
PRZM/
EXAMS
modeling
are
given
in
Table
4.
Aventis'
response:
The
partial
list
of
input
parameters
in
Table
4
includes
multiple
conservative
assumptions
likely
to
lead
to
significant
over­
estimation
of
the
potential
surface
water
concentrations
of
carbaryl.

Page:
31
Paragraph:
2
Line:
1
EPA
comment:
For
the
Index
Reservoir
scenario
using
maximum
label
rates,
acute
EEC
values
ranged
from
about
10
:
g/
L
from
sugar
beets
to
about
500
:
g/
L
from
citrus
(Table
6).
Aventis'
response:
Table
6
on
page
33
shows
a
concentration
of
19
:
g/
L
for
sugar
beets
treated
with
the
maximum
label
rate
of
2
x
1.5
lb
ai,
not
10
as
stated
in
this
sentence.
A
low
EEC
value
of
9
:
g/
L
for
sugar
beets
results
from
the
"maximum
reported"
application
scenario
of
1
x
1.2
lb
ai/
A.

Page:
31
Paragraph:
2
Line:
3
EPA
comment:
Chronic
EECs
ranged
from
about
1
to
28
:
g/
L.
36
Aventis'
response:
Table
6
on
page
33
shows
that
this
is
correct
when
considering
all
of
the
model
scenarios.
However,
either
the
same
maximum
label
rate
reference
should
be
used
as
in
the
preceding
sentence
(in
which
case
the
minimum
chronic
EEC
would
be
2),
or
the
basis
for
the
preceding
sentence
should
be
changed
from
the
maximum
label
rate
to
include
all
application
scenarios
to
keep
the
comparisons
consistent.

Page:
31
Paragraph:
2
Line:
6
EPA
comment:
It
is
highly
unlikely
that
any
but
the
most
extensive
targeted
monitoring
would
capture
the
actual
peak
concentrations.
Aventis'
response:
The
role
of
a
peak
concentration
in
dietary
exposure
assessment
is
undergoing
reexamination
within
EPA.
The
current
policy
of
EPA
appears
to
define
a
certain
percentile
as
an
appropriate
value
for
use
in
screening
assessments,
but
the
exact
percentile
to
be
used
is
being
currently
set
by
EPA
management.
(The
most
recent
documents
from
EPA
cite
the
95
th
or
99
th
percentile.)
For
more
comprehensive
assessments,
a
distribution
of
values
is
preferred.

Page:
31
Paragraph:
2
Line:
7
EPA
comment:
The
results
of
the
modeling
provide
a
conservative,
though
not
unreasonable,
estimate
on
possible
concentrations
drinking
water.(
sic)
Aventis'
response:
The
modeling,
performed
according
to
EPA
procedures,
provides
an
upper
bound
estimate
on
potential
concentrations
in
drinking
water
from
surface
water.
Whether
the
modeling
estimates
are
reasonable
depends
on
the
specific
assumptions.
For
carbaryl,
the
three­
year
monitoring
program
(conducted
according
to
EPA
and
ILSI
guidance
available
at
the
time
the
study
was
started)
shows
that
the
model
calculations
are
unreasonable.
These
conservative
assumptions
include
a
3x
factor
on
both
the
aerobic
soil
and
aerobic
aquatic
half
lives,
assuming
the
maximum
drift
rate
for
aerial
applications
throughout
the
county
(in
Florida
citrus
almost
all
applications
are
by
air
blast
with
ground
equipment),
and
the
application
rate
over
a
watershed.
The
conservative
nature
of
the
application
assumption
alone
probably
results
in
an
overprediction
by
at
least
two
orders
of
magnitude.
The
modeling
calculations
assume
an
application
rate
of
17.4
lbs/
acre
of
watershed
annually.
In
Hardee
County,
the
county
with
the
highest
usage
of
carbaryl,
the
average
use
rate
on
a
countywide
basis
is
only
0.31
lb/
acre
(See
Appendix
II).
In
Manatee
County,
the
county
with
the
highest
usage
containing
a
watershed
used
to
supply
drinking
water,
the
average
rate
on
a
countywide
basis
is
0.027
lb/
acre.
37
Page:
31
Paragraph:
2
Line:
8
EPA
comment:
A
more
detailed
assessment
of
the
source
of
water
used
to
provide
drinking
water
and
the
relationship
between
the
areas
where
carbaryl
is
used
and
surface
water
sources
is
required
to
more
accurately
evaluate
possible
human
exposures.
Aventis'
response:
As
mentioned
by
EPA
in
this
document,
ground
water
is
the
source
of
the
majority
of
Florida
drinking
water.
Many
of
the
counties
with
the
highest
use
of
carbaryl
contain
no
watersheds
used
to
provide
drinking
water.
As
discussed
more
fully
in
Appendix
I,
the
watershed
supplying
the
Manatee
County
Water
Treatment
Plant
appears
to
have
the
most
carbaryl
usage
of
drinking
water
watersheds
in
Florida.

Water
Treatment
Effects
Page:
31
Paragraph:
3
Line:
8
EPA
comment:
Since
relatively
(sic)
few
water
treatment
facilities
in
the
U.
S.
use
ozone
the
limited
data
available
do
not
indicate
that
carbaryl
is
likely
to
be
degraded
in
the
majority
of
treatment
plants.
Aventis'
response:
The
monitoring
program
conducted
by
the
registrant
shows
that
removal
occurs
in
some
treatment
plants.
The
effect
of
treatment
seemed
to
be
greater
in
systems
using
carbon
treatment.

Page:
33
Table
6
EPA
comment:
Drinking
Water
EECs
Aventis'
response:
Many
of
the
comments
for
this
table
are
similar
to
those
for
the
EECs
for
ecological
risk
found
in
Table
5.

The
PRZM
model
input
parameters
for
the
Index
Reservoir
scenarios
were
received
as
an
electronic
copy
of
a
draft
of
Appendix
B.
These
input
files
are
very
useful
for
assessing
the
scenarios
that
have
been
modeled.

It
would
be
useful
to
add
another
column
to
Table
6
to
specify
which
method
of
application
was
used
to
generate
the
EECs
(and
thus
the
application
efficiency
and
spray
drift
values).
It
would
be
of
benefit
for
the
Agency
to
state
which
of
the
carbaryl
labels
were
used
to
develop
the
"maximum"
label
application
rate
scenarios.
There
are
a
number
of
errors
in
the
input
parameters
(noted
below)
that
would
lead
to
changes
in
the
calculated
EECs
and
therefore
the
risk
quotients
for
these
uses.
38
The
model
parameters
listed
in
the
electronic
draft
of
Appendix
B
show
that
the
"average"
scenarios
for
citrus
and
apples
were
conducted
using
aerial
applications.
Few
applications
to
these
crops
are
made
aerially.
Therefore,
the
model
results
over­
estimate
the
contributions
from
spray
drift
since
the
"average"
applications
to
these
crops
are
made
using
ground
airblast
equipment
with
a
spray
drift
of
6.3%
versus
aerial
applications
with
a
spray
drift
of
16%.

The
"maximum
label
rate"
application
scenario
for
apples
that
is
allowed
by
the
Sevin
brand
XLR
PLUS
label
(E.
P.
A.
Reg.
No
264­
333),
the
Sevin
brand
80WSP
and
CHIPCO
Sevin
brand
80WSP
labels
(E.
P.
A.
Reg.
No
264­
526)
and
the
CHIPCO
Sevin
brand
SL
label
(E.
P.
A.
Reg.
No
264­
335)
is
5
applications
at
3
lb
ai/
A/
application
made
every
14
days.
The
scenario
used
in
the
model
applies
less
than
the
maximum
amount
of
product
allowed
by
the
labels.
In
addition,
application
timing
was
used
in
the
modeling
for
the
index
reservoir
scenario
(applications
made
by
air
every
4
days)
that
would
be
a
violation
of
the
Aventis
labels
which
restrict
applications
to
a
minimum
of
every
14
days.

The
"average"
scenario
for
sweet
corn
in
Ohio
should
be
3
applications
at
1.1
lb
ai/
A/
application
(as
noted
in
the
memo,
"Average
application
rate
from
Quantitative
Usage
Analysis
for
Carbaryl,
prepared
July
21,
1998
by
Frank
Hernandez,
OPP/
BEAD")
and
not
the
2
applications
at
3.4
lb
ai/
A/
application
as
listed
in
the
table.
The
PRZM
input
file
shows
the
correct
inputs
of
3
applications
at
1.1
lb
ai/
A/
application.

The
"average"
scenario
for
sugar
beets
in
Minnesota
should
be
1
application
at
1.3
lb
ai/
A/
application
(as
noted
in
the
memo,
"Average
application
rate
from
Quantitative
Usage
Analysis
for
Carbaryl,
prepared
July
21,
1998
by
Frank
Hernandez,
OPP/
BEAD")
and
not
1
application
at
1.5
lb
ai/
A/
application
as
listed
in
the
table
and
the
PRZM
input
file.

The
"Citrus"
scenario
would
be
more
appropriately
labeled
Oranges.
For
the
average
scenario,
the
3.4
lb
ai/
A/
application
rate
listed
in
Table
5
is
for
oranges
(as
noted
in
the
memo,
"Average
application
rate
from
Quantitative
Usage
Analysis
for
Carbaryl,
prepared
July
21,
1998
by
Frank
Hernandez,
OPP/
BEAD"),
which
is
the
highest
"average"
application
rate
for
any
type
of
citrus.
Therefore,
this
"average"
scenario
for
oranges
is
at
the
high
end
for
all
citrus
and
overestimates
the
EECs
for
use
in
the
other
citrus
crops.
"Average"
application
rates
for
other
citrus
as
listed
in
the
memo
are:
Lemons
–
1.3
applications
at
2.7
lb
ai/
A/
appl
Grapefruit
–
1.6
applications
at
1.4
lb
ai/
A/
appl
Citrus,
other
–
1.8
applications
at
1.8
lb
ai/
A/
appl
Ground
Water
Resources
Page:
34
Paragraph:
carried
over
from
page
33
Line:
3
EPA
comment:
U.
S.
EPA.
Pesticides
in
Groundwater
Database
(Jacoby
et
al.,
1992)
39
Aventis'
response:
This
reference
is
not
provided
in
the
reference
list.

Page:
34
Paragraph:
3
Line:
3
EPA
comment:
Detections
were
from
(sic)
mainly
from
three
use
sites:
wheat
(5.8
%
of
well
samples
from
wheat
land
use),
orchards
and
vineyards
(1.7
%
of
well
samples
from
orchard
and
vineyard
land
use),
and
urban
(1.8%
of
urban
groundwater
samples).
Aventis'
response:
Updated
information
(noted
below)
is
not
summarized
in
the
same
manner
as
in
this
statement,
so
direct
comparisons
cannot
be
made
easily.
However,
the
updated
information
indicates
a
similar
pattern
of
low
concentrations
of
carbaryl
detections
in
a
limited
number
of
ground
water
resources.

Page:
34
Paragraph:
3
Line:
6
EPA
comment:
Limitations
in
analytical
methodology
(described
elsewhere)
apply
to
groundwater
sample
analysis
also
suggesting
that
there
(sic)
actual
maximum
concentrations
and
extent
of
contamination
may
be
significantly
higher.
Aventis'
response:
This
statement
is
misleading
and
should
be
deleted.
The
validation
of
the
most
widely
used
GC/
MS
method
for
the
data
contained
in
NAWQA
show
recoveries
of
86
to
94%
at
spiking
levels
of
0.1
to
1.0
:
g/
L
with
an
MDL
of
0.003
:
g/
L.
The
HPLC
method
validation
reported
recoveries
of
58
to
64%
%
at
spiking
levels
of
0.1
to
1.0
:
g/
L
with
an
MDL
of
0.018
:
g/
L.
Furthermore,
using
the
GC/
MS
method,
a
mean
recovery
of
115%
was
found
for
field
matrix
spikes
of
carbaryl
at
spiking
levels
of
0.1
:
g/
L.
With
the
GC/
MS
method
MDL
of
0.003
:
g/
L
and
a
mean
recovery
of
115%
for
the
field
matrix
spikes,
this
method
cannot
reasonably
be
characterized
as
stated
by
EPA.
Additional
details
of
the
method
validations
and
field
matrix
spikes
are
provided
in
the
`Discussion
Section'
at
the
end
of
this
response.

Page:
34
Paragraph:
3
Line:
last
EPA
comment:
…and
updated
information
is
available
at:
http://
water.
wr.
usgs.
gov/
pnsp/
ja/
est32/.
Aventis'
response:
This
web
page
was
last
updated
in
1998.
A
more
recent
update
by
Kolpin
was
posted
June
11,
2001
at:
http://
water.
wr.
usgs.
gov/
pnsp/
pestgw/
and
is
the
source
of
the
updated
information
included
in
the
`Discussion
Section'
at
the
end
of
this
response.
40
Surface
Water
Resources
Monitoring
Data
Page:
34
Paragraph:
4
Line:
5­
6
EPA
comment:
Because
of
limitation
in
the
analytical
methods
used
there
is
some
question
as
to
the
accuracy
of
carbaryl
analysis.
Aventis'
response:
This
generalized
statement
needs
to
be
qualified
or
deleted.
Whereas
the
authors
of
reports
written
as
part
of
the
NAWQA
program
have
been
clear
about
the
potential
limitations
of
the
quantitative
nature
of
the
carbaryl
data
in
the
database,
they
have
also
been
clear
about
the
validity
of
the
qualitative
nature
of
the
data.
The
use
of
the
multiresidue
method
in
the
NAWQA
program
does
have
some
limitations
as
a
result
of
the
large
numbers
of
diverse
pesticides
and
degradation
products
that
they
are
monitoring.
However,
the
QC/
QA
data
generated
as
part
of
the
NAWQA
program
(described
in
the
discussion
section
on
surface
water
at
the
end
of
this
response)
demonstrate
the
validity
of
the
detections
of
carbaryl
in
the
studies.
The
monitoring
study
conducted
by
the
registrant,
and
reported
in
this
section,
does
not
have
the
same
potential
limitations
in
the
analytical
method
since
the
method
is
looking
specifically
for
carbaryl.
Therefore,
the
analytical
method
used
by
the
registrant
does
not
raise
questions
about
the
accuracy
of
the
carbaryl
analysis.

Page:
34
Paragraph:
4
Line:
5­
6
EPA
comment:
Poor
analytical
methods
probably
have
resulted
in
lower
detection
rates
and
lower
concentrations
than
actually
present.
Aventis'
response:
This
generalized
statement
should
be
deleted
for
reasons
provided
above
and
in
the
discussion
section.

NAQWA
(sic)
41
Change
to
NAWQA
Page:
34
­
35
Paragraph:
5
Lines:
5­
8
EPA
comment:
Carbaryl
analytical
results
are
fairly
poor,
with
a
typical
mean
percent
recovery
of
24%
(
 
=
15)
in
laboratory
quality
control
samples,
and
a
method
detection
limit
(MDL)
of
0.003
ug/
L.
This
suggests
that
the
values
reported
do
not
represent
the
maximum
concentrations
that
exist,
and
that
surface
water
contamination
may
be
more
widespread
than
the
data
show.
Aventis'
response:
These
statements
are
misleading
and
should
be
updated
with
further
quality
control
data
supplied
by
NAWQA.

A
discussion
of
the
analytical
method
used
in
the
NAWQA
program
is
presented
in
the
USGS
Open­
File
Report
95­
181
(see
Zaugg
et
al.
(1995)
in
references).
The
mean
percent
recovery
of
24%
noted
above
can
be
found
in
Table
9
of
this
report
and
is
by
no
means
"typical".
A
mean
recovery
value
of
24%
was
reported
for
reagent­
grade
water
fortified
at
a
level
of
0.03
µ
g/
L
with
a
method
detection
limit
said
to
be
0.003
µ
g/
L.
Additional
recoveries
for
fortified
water
samples
(reagent­
grade,
ground
and
surface
waters)
ranged
from
10
to
202%
(see
discussion
section).
The
carbaryl
data
in
the
NAWQA
database
are
amended
with
an
"E"
qualifier
to
indicate
the
variability
found
with
this
method,
not
because
the
carbaryl
concentrations
are
underestimated.

Additional
evaluations
of
field
blank,
field
matrix
spike
and
lab
control
spike
samples
as
part
of
the
NAWQA
program
can
be
found
in
a
provisional
report
by
Martin
(1999).
This
report
demonstrates
the
lack
of
detection
of
carbaryl
in
100%
of
the
field
blanks,
and
median
recoveries
of
94.4%
in
306
field
matrix
spikes
and
93.0%
in
1000
lab
control
spikes,
each
at
spiking
levels
of
0.1
µ
g/
L.
These
data
suggest
an
adequate
level
of
detection
of
carbaryl
in
the
method
used
in
the
NAWQA
survey
of
surface
and
ground
water.
See
the
additional
discussion
at
the
end
of
this
document
for
further
information
regarding
recoveries
in
spiked
surface
and
ground
water.

Page:
35
Paragraph:
2
Line:
7
EPA
comment:
…at
about
0.1
percent
of
the
amount
used
in
the
basins
(Larson
et
al.,
1999)
http://
water.
wr.
usgs.
gov/
pnsp/
rep/
wrir984222/
load.
html.
The
estimated
carbaryl
use
on
in
agricultural
applications
is
about
4
million
pounds
suggesting
that
400,000
pounds
are
delivered
to
the
nations
streams
draining
agricultural
areas.
42
Aventis'
response:
This
estimated
use
of
carbaryl
for
agricultural
applications
over­
estimates
the
use
of
carbaryl
by
about
1
million
pounds.
BEAD
and
USGS
data
cited
on
pages
6
and
7
are
consistent
with
lower
total
pounds
of
carbaryl
applied.
In
addition,
0.1
percent
of
4
million
pounds
would
be
4,000
pounds,
not
400,000
pounds.
If
the
1987
–
1996
average
of
2.5
million
pounds
carbaryl
is
used
in
the
calculation,
the
total
load
suggested
to
be
delivered
to
streams
draining
agricultural
areas
would
be
2,500
pounds.

Registrant
Monitoring
Study
Page:
35
Paragraph:
4
Line:
11
EPA
comment:
Carbaryl
was
analyzed
by
HPLC/
MS
with
a
limit
of
detection…
Aventis'
response:
The
analytical
method
used
by
the
registrant
in
the
surface
water
monitoring
study
uses
tandem
mass
spectrometry
(MS/
MS)
as
the
detection
method.
This
type
of
detection
involves
quantification
of
"daughter"
ions
from
a
selected
mass
fragment
and
is
more
selective
than
an
MS
method.
Therefore,
to
accurately
reflect
these
differences,
the
method
should
be
labeled
as
HPLC/
MS/
MS.

Page:
36
Paragraph:
3
Line:
9
EPA
comment:
In
several
cases
finished
water
had
higher
concentration
than
raw
water,
and
finished
water
had
detectable
carbaryl
when
the
raw
did
not.
The
highest
concentration
measured
was
in
finished
water
(0.18
ppb).
Raw
water
sampled
at
the
same
time
had
much
lower
concentration
(0.010).
43
Aventis'
response:
This
statement
is
misleading
and
certainly
does
not
consider
the
analytical
uncertainty
for
concentrations
below
the
level
of
quantification
and
near
the
level
of
detection.
There
were
only
two
cases
when
finished
water
was
greater
than
raw
water
when
the
concentrations
in
finished
water
were
greater
than
0.01
ppb
(only
one­
third
of
the
quantification
limit).
One
case
was
when
the
raw
water
was
0.009
ppb
and
the
finished
water
was
0.011
ppb.
These
two
analyses
are
essentially
equivalent,
especially
considering
that
they
are
only
about
a
third
of
the
quantification
limit.
The
other
case
was
at
the
Deerfield
community
water
system.
This
drinking
water
facility
uses
a
small
river
without
a
reservoir
as
a
source
for
a
small
Community
Water
System.
Farms
are
located
immediately
upstream
of
the
facility.
The
intake
is
also
not
continuous
(shut
down
over
weekends).
Therefore,
getting
a
matching
sample
is
quite
difficult,
especially
for
a
short
duration
spike
as
a
result
of
spray
drift,
summer
thunderstorm,
or
perhaps
a
spill
that
almost
immediately
enters
the
river
a
runoff
event.
The
rarity
of
this
event
is
demonstrated
by
the
absence
of
residues
of
this
magnitude
the
next
year
(2000).
Samples
collected
through
this
time
of
the
year
in
2001
also
do
not
indicate
a
similar
event.
Although
the
data
from
this
site
cannot
be
used
to
determine
the
peak
concentration,
the
data
provide
a
distribution
of
residues
through
the
three
year
period
which
will
define
up
to
the
99
th
percentile
concentration
of
the
distribution.

The
Deerfield,
Michigan
community
water
system
is
one
of
the
systems
in
which
the
greatest
variability
of
residues
would
be
expected.
Most
of
the
other
community
water
systems
are
located
on
larger
rivers,
lakes,
or
reservoirs.

Because
the
design
of
study
called
for
analysis
of
finished
water
only
when
there
were
residues
in
the
raw
water,
there
was
only
one
finished
sample
analyzed
when
the
raw
water
contained
no
residues.
This
sample
was
collected
at
the
Deerfield
community
water
system
at
the
sampling
interval
after
the
finding
of
0.16
ppb
in
the
Deerfield
system.
The
residue
level
in
this
sample
was
0.004
ppb.
The
difference
between
0.004
ppb
and
non­
detect
is
insignificant,
and
if
real
can
probably
be
attributed
to
water
at
much
higher
concentrations
remaining
in
the
system
from
the
previous
week.

Page:
36
Paragraph:
4
Line:
1
EPA
comment:
Non­
targeted
monitoring,
such
as
the
NAWQA
program,
has
shown
much
higher
concentrations
occur
indicating
that
this
study,
while
useful,
can
not
be
used
to
describe
the
overall
distributions
that
occur
throughout
the
entire
use
area.
Aventis'
response:
The
targets
of
the
drinking
water
monitoring
conducted
by
the
registrant
and
the
NAWQA
program
are
different.
The
NAWQA
program
characterized
surface
water
concentrations
within
a
study
area
while
the
Aventis
drinking
water
monitoring
measured
residues
in
inlets
and
outlets
of
drinking
water
facilities.
Also
the
drinking
water
monitoring
program
considered
only
use
areas
with
drinking
water
supplies.
However,
44
for
FQPA
dietary
assessments,
the
appropriate
target
is
drinking
water
rather
than
surface
water.

The
main
reason
why
the
drinking
water
monitoring
study
did
not
show
residues
as
high
as
in
the
NAWQA
program
is
the
location
of
the
sampling
points.
Drinking
water
supplies
tend
to
be
located
on
larger
surface
water
bodies
than
NAWQA
sampling
points
(or
in
other
words,
the
intakes
for
community
water
systems
tend
to
be
downstream
of
NAWQA
sampling
points).
This
additional
time
allows
for
additional
degradation
and
dilution
to
occur.
Finding
the
highest
concentration
at
the
Deerfield,
Michigan
system
is
not
surprising
since
this
intake
is
on
one
of
the
smallest
surface
water
bodies
included
in
the
monitoring
study
(see
response
to
Page:
36,
Paragraph:
3,
Line:
9
above
for
a
more
detailed
explanation).

Page:
36
Paragraph:
4
Line:
4
EPA
comment:
This
study
does
not
provide
sufficient
information
to
allow
estimation
of
actual
peak
and
mean
concentrations
that
actually
occur
in
all
use
areas.
Aventis'
response:
Because
most
of
the
samples
did
not
contain
carbaryl
residues,
accurate
estimates
of
the
actual
peak
and
mean
concentrations
can
not
be
obtained.
However,
the
distributions
obtained
from
all
sites
can
be
used
to
define
up
to
the
99
th
percentile
concentration.
The
average
cannot
be
accurately
determined;
however,
the
time­
weighted
average
is
only
slightly
above
the
limit
of
detection
(and
certainly
less
than
0.01
ppb)
at
all
20
sites.

The
peak
concentration
in
this
study
was
measured
at
a
community
water
system
on
a
small
river.
The
registrant
agrees
that
the
sampling
schedule
was
not
adequate
to
determine
the
true
peak
in
such
systems.
Most
of
the
other
community
water
systems
are
located
on
larger
rivers,
lakes,
or
reservoirs.
Therefore,
the
peak
values
are
not
likely
to
be
an
order
of
magnitude
greater
than
the
amounts
present
in
the
collected
samples.

The
distributions
obtained
in
this
study
are
suitable
for
use
in
dietary
exposure
assessments.
When
EPA
policy
establishes
what
percentile
concentration
is
an
appropriate
regulatory
endpoint,
then
these
percentiles
can
be
determined
for
each
of
the
community
water
systems
monitoring.
These
percentiles
can
then
be
compared
with
DWLOC
values
in
screening
assessments.

Page:
37
Paragraph:
2
Line:
1
EPA
comment:
Only
limited
information
was
submitted
on
sampling
site
selection…
Aventis'
response:
The
summary
in
Appendix
I
of
this
response
provides
a
description
of
the
sites
considered
for
the
monitoring
study
and
the
rationale
for
the
selection
of
the
twenty
sites.
This
information
demonstrates
that
the
community
water
systems
selected
for
this
study
45
are
representative
of
the
systems
that
are
most
likely
to
contain
the
highest
concentrations
of
carbaryl
residues.

Page:
37
Paragraph:
3
Line:
3
EPA
comment:
This
should
include
an
explanation
of
why
this
study
did
not
observe
concentrations
as
high
as
those
found
in
other,
non­
targeted
studies,
and
how
the
results
of
this
study
can
be
related
to
concentrations
that
occur
throughout
the
country.
Aventis'
response:
The
main
reason
why
the
drinking
water
monitoring
study
did
not
show
residues
as
high
as
in
the
NAWQA
program
is
the
location
of
the
sampling
points.
Drinking
water
supplies
tend
to
be
located
on
larger
surface
water
bodies
than
NAWQA
sampling
points
(or
in
other
words,
the
intakes
for
community
water
systems
tend
to
be
downstream
of
NAWQA
sampling
points).
This
additional
time
allows
for
additional
degradation
and
dilution
to
occur.
Finding
the
highest
concentration
at
the
Deerfield,
Michigan
system
is
not
surprising
since
this
intake
is
on
one
of
the
smallest
surface
water
bodies
included
in
the
monitoring
study
(see
response
to
Page:
36,
Paragraph:
3,
Line:
9
above).

Since
the
drinking
water
study
targeted
drinking
water
systems
in
high­
use
watersheds,
the
data
from
this
study
are
representative
of
the
drinking
water
systems
most
likely
to
contain
carbaryl.

Sacramento­
San
Joaquin
River
Delta
Page:
37
Paragraph:
4
Line:
4
­
5
EPA
comment:
Carbaryl
was
found
to
be
the
sole
causative
agent
at
one
of
20
sites…
The
toxicity
seemed
to
persist
for
several
days…
Aventis'
response:
The
statement
should
be
revised.
The
reference
cited
(Werner
et
al.,
2000)
lists
carbaryl
as
"the
primary
toxicant"
(not
as
the
"sole
causative
agent"),
even
though
an
unknown
was
also
found
at
the
same
time.
No
information
about
the
"unknown"
is
provided.
Both
conclusions
of
"sole
causative"
and
of
"primary
toxicant"
cannot
be
substantiated
without
further
evidence
about
the
nature
and
concentration
of
the
unknown.
Actually,
for
another
site
the
authors
concluded
about
the
unknown
found
there
"in
3
of
21
samples,
toxicity
observed
could
not
be
entirely
explained
by
the
identified
primary
toxicants."
Additionally,
it
is
at
least
questionable
if
the
analytical
method
employed
would
detect
all
potential
toxicants
beside
the
insecticides
it
was
set
up
for.

The
toxicity
seeming
to
persist
is
not
explained
or
substantiated
in
the
reference.
The
citation
of
such
dubious
results
should
be
removed
from
the
RED.
46
6.0
Hazard
and
Risk
Assessment
for
Aquatic
Organisms
Hazard
assessment
for
Aquatic
organisms
Estuarine/
Marine
Fish
Page:
39
Paragraph:
2
Line:
6
EPA
comment:
…carbaryl
water
concentration
of
1.2
:
g/
ml…
Aventis'
response:
To
be
consistent
with
the
rest
of
the
document
the
units
should
be
presented
in
ppm
("
carbaryl
water
concentration
of
1.2
ppm")

Aquatic
Plants
Page:
40
Paragraph:
2
Line:
6
EPA
comment:
Guideline
122­
2
is
not
fulfilled.
Aventis'
response:
The
chapter
should
be
revised.
As
detailed
above
(comments
to
Page
2
of
the
Memorandum),
studies
were
submitted
in
1992.
The
status
for
this
requirement
in
an
October
04,
2000
OPP
Guideline
Status
Report
(Chemical
Review
Management
System)
lists
the
guideline
122­
2
status
as
"Acceptable/
Satisfied".

Risk
Assessment
for
Aquatic
Organisms
Page:
40
Paragraph:
4
Line:
3
EPA
comment:
…corresponding
levels
of
concern
(LOCs)
is
presented
in
Appendix
D.
Aventis'
response:
The
risk
quotients
are
currently
listed
in
Appendix
C.

Estuarine/
Marine
Fish
Page:
42
Paragraph:
1
Line:
17/
18
EPA
comment:
Chronic
toxicity
studies
with
an
estuarine/
marine
fish
species
is
required.
47
Aventis'
response:
This
requirement
should
be
waived.
Given
the
relatively
short
half­
life
of
carbaryl
in
the
aquatic
environment
and
the
low
acute
risk,
it
is
unlikely
that
estuarine/
marine
fish
species
would
be
exposed
to
a
chronic
risk.

Page:
42
Paragraph:
2
Line:
1
EPA
comment:
There
is
one
carbaryl
use
in
particular
that
presents
a
major
acute
and
chronic
risk
to
estuarine/
marine
fish.
Aventis'
response:
This
sentence
should
be
rephrased.
While
there
might
be
an
acute
risk
from
the
application
to
oyster
beds,
given
that
there
is
only
one
application
every
six
years
according
to
the
reference
cited
by
EPA,
it
is
improbable
that
estuarine/
marine
fish
would
be
exposed
to
a
chronic
risk.

7.0
Hazard
and
Risk
Assessment
for
Terrestrial
Organisms
Hazard
Assessment
for
Terrestrial
Organisms
Mammalian
Page:
46
Paragraph:
4
Line:
1
EPA
comment:
With
a
rat
LD50
of
307
mg/
kg…
Aventis'
response:
Typographical
error,
the
rat
LD50
is
301
mg/
kg.

Risk
Assessment
for
Terrestrial
Organisms
Avian
Risk
Nongranular
Formulations
Page:
47
Paragraph:
4
Line:
5
EPA
comment:
…levels
of
concern
(LOCs)
is
presented
in
Appendix
D.
Aventis'
response:
The
risk
quotients
are
currently
listed
in
Appendix
C.
48
Page:
48
Paragraph:
1
Line:
3
EPA
comment:
…
for
34
of
43
uses
at
maximum
reported
rates,
and
for
37
of
72
uses
at
"average"
rates.
(Appendix
D,
…
Aventis'
response:
The
risk
quotients
are
currently
listed
in
Appendix
C.

Granular
Formulations
Page:
48
Paragraph:
2
Line:
5
EPA
comment:
…for
any
of
the
granular
carbaryl
uses
(Appendix
D,
Table
6).
Aventis'
response:
The
risk
quotients
are
currently
listed
in
Appendix
C.

Mammalian
Risk
Risk
to
Herbivores/
Insectivores:
Nongranular
Formulations
Risk
Quotients
for
Herbivores/
Insectivores
Based
on
Less
than
Maximum
Label
Use
Rates
Page:
48
Paragraph:
3
Line:
3
&
4
EPA
comment:
…
(Appendix
D,
Table
10a)
and
maximum
reported
(Doane
data)
use
rates
data
available
for
43
uses
(Appendix
D,
Table
10b)
Aventis'
response:
The
risk
quotients
are
currently
listed
in
Appendix
C.

Risk
Quotients
for
Herbivores/
Insectivores
Based
on
Maximum
Label
Use
Rates
Page:
48
Paragraph:
6
Line:
1
EPA
comment:
Carbaryl
is
moderately
toxic
to
small
mammals
on
an
acute
oral
basis
(rat
LD50
=
307
mg/
kg)
Aventis'
response:
Typographical
error,
the
rat
LD50
is
301
mg/
kg.
By
using
the
lower
LD50
all
acute
mammalian
risk
quotients
will
change
slightly.
49
Page:
49
Paragraph:
1
Line:
3
EPA
comment:
…corresponding
levels
of
concern
(LOCs)
is
presented
in
Appendix
D.
Aventis'
response:
The
risk
quotients
are
currently
listed
in
Appendix
C.

Risk
to
Granivores:
Nongranular
Uses
Chronic
risk:
Nongranular
Uses
Page:
50
Paragraph:
2
Line:
8
EPA
comment:
…summarized
in
Appendix
D,
Table
9.
Aventis'
response:
The
risk
quotients
are
currently
listed
in
Appendix
C.

Reproduction
Effects
Page:
50
&
51
Paragraph:
4
/
1
EPA
comment:
(Review
of
alleged
reproduction
effects
of
carbaryl).
Aventis'
response:
The
paragraphs
should
be
changed.
The
literature
cited
in
these
paragraphs
show
ambivalent
results.
While
some
references
seem
to
support
the
claim
of
reproductive
effects,
other
references
do
not.
The
potential
for
reproductive
effects
in
mammals
was
evaluated
in
the
recently
submitted
2­
generation
study
in
rats.
No
reproductive
effects
were
seen
in
this
guideline
study.
The
NOAEC
of
75
ppm
was
based
on
pup
mortality.

Page:
51
Paragraph:
4
EPA
comment:
Feeding
2
or
20
mg/
kg
of
carbaryl
to
pregnant
rhesus
monkeys
(Macacca
mulatta)
Aventis'
response:
This
paragraph
should
be
deleted.
As
there
are
no
native
monkey
species
in
the
U.
S.,
this
reference
is
irrelevant
for
U.
S.
wildlife
species.
Additionally,
the
reference
cited
is
only
a
brief
abstract
article
consisting
of
one
17­
line
paragraph.
Such
information
should
not
be
the
basis
for
use
in
a
RED
risk
assessment.
50
9.0
References
(non­
MRID)

Some
of
the
references
cited
in
EPA's
list
are
not
full
scientific
articles,
but
only
abstracts
from
meetings
(e.
g.
DeNorsica,
1973;
Doughtery
et
al.
,
1971,
Chapin
et
al.
1997).
Such
"publications"
should
not
be
used
as
references
considered
in
risk
assessments.
Without
a
sufficient
description
of
methods
and
a
presentation
of
detailed
results
these
studies
cannot
be
evaluated
to
determine
if
the
findings
are
or
are
not
scientifically
plausible.
Similarly,
at
least
three
of
the
references
(Gladenko
et
al.
1970,
Krylova
et
al.
1975,
Smirnov
et
al.
1971)
cited
as
proof
for
reproductive
toxicity
are
in
Russian
in
Cyrillic
writing
making
an
appropriate
and
timely
evaluation
difficult.
Due
to
the
limited
review
time
during
the
30­
day
comment
period,
the
registrant
could
not
peruse
all
references.
A
more
detailed
response
will
be
provided
during
the
60­
day
comment
period.

Page:
59
EPA
comment:
Carmel,
R.
F.,
Imhoff,
J.
C.,
Hummel,
P.
R.,
Cheplick,
J.
M.
and
Donigan,
A.
S.,
1997.
Aventis'
response:
The
first
name
should
be
Carsel.

Page:
59
EPA
comment:
Nkedi­
Kizza
and
Brown
(1988)
Aventis'
response:
The
date
should
be
1998.

Appendix
A:
Environmental
Fate
Study
Reviews
(DERs)

Page:
62
ff
EPA
comment:
(Environmental
fate
DERs
are
included)
Aventis'
response:
The
DERs
should
not
be
included
in
the
RED.
Publication
of
DERs
together
with
the
RED
is
unusual
and
will
put
Aventis
in
a
competitive
disadvantage.

Appendix
B:
Refined
Water
Memo
EPA
comment:
MEMO
TO
BE
INSERTED
WHEN
APPROVED
51
Aventis'
response:
This
memo
was
provided
as
an
electronic
copy
and
needs
to
be
inserted
into
the
document.
It
included
text
that
repeated
several
sections
of
the
EFED
document
and
it
included
PRZM
input
tables
for
the
drinking
water
concentrations
using
the
Index
Reservoir
scenario.
It
would
have
been
of
benefit
to
have
the
same
PRZM
inputs
for
the
"standard
pond"
scenarios
that
were
used
to
estimate
surface
water
concentrations
used
in
the
aquatic
risk
assessments.

Appendix
C:
Ecological
Risk
Assessment
Toxicity
Endpoints
Used
in
the
Risk
Assessment
Page:
129
(e­
version)

EPA
comment:
Aventis'
response:

Mammalian
acute
oral
LD50
rat
=
307
mg/
kg
The
correct
LD50
is
301
mg/
kg
Mammalian
chronic
(reproduction)
NOAEC
rat
=
80
ppm
The
result
of
the
recently
submitted
2­
generation
rat
study
should
be
used
(75
ppm)

Avian
Acute
and
Chronic
Risk
Page:
130
(e­
version)
Paragraph:
1
Line:
1
EPA
comment:
Since
the
avian
LC50
is
greater
than
5,000
ppm
(Appendix
E),
Aventis'
response:
The
toxicity
data
are
currently
listed
in
Appendix
D.

Page:
132
–
135
(e­
version)
EPA
comment:
(Acute
Risk
Quotients
in
Tables
4
and
5,
as
well
as
throughout
the
document
were
a
reference
is
made
to
these
quotients)
Aventis'
response:
As
the
acute
risk
quotients
are
calculated
on
the
basis
of
an
LC50
of
>
5000
ppm,
the
quotients
should
be
given
as
"<
(value)",
not
just
the
value.
The
values
should
also
be
changed
accordingly
throughout
the
document
where
a
reference
is
made
to
these
quotients.
52
Risk
from
Exposure
to
Non­
granular
Products
Page:
137
–
147
(e­
version)
EPA
comment:
(Text
and
tables
7
­
10)
Aventis'
response:
Text
and
tables
should
be
revised.
A
rat
LD50
of
307
mg/
kg
was
used
to
calculate
the
acute
risk
quotients.
The
correct
value
is
301
mg/
kg.
For
calculation
of
the
chronic
risk
quotient
a
NOAEC
of
80
ppm
was
taken
from
a
developmental
study.
The
NOAEC
of
75
ppm
from
a
more
relevant
2­
generation
rat
study
recently
submitted
should
be
used
instead.

Risk
from
Exposure
to
Granular
Products
Page:
147
&
148
(e­
version)
EPA
comment:
(Text
and
Table
11)
Aventis'
response:
Text
and
tables
should
be
revised.
A
rat
LD50
of
307
mg/
kg
was
used
to
calculate
the
acute
risk
quotients.
The
correct
value
is
301
mg/
kg.

Aquatic
Plants
Page:
152
(e­
version)
EPA
comment:
Based
on
a
single
core
aquatic
plant
toxicity
study
available…
…recommended
that
toxicity
studies
with
Lemna
gibba,
Anabaena
flos­
aquae,
Skeletonema
costatum,
and
a
freshwater
diatom
be
submitted.
Aventis'
response:
The
respective
studies
were
submitted
to
the
Agency
in
1992
(see
comments
above
to
Page
2
of
the
Memorandum
for
a
complete
list
and
status).

Appendix
D:
Toxicity
Assessment
Page:
157
(e­
version)
EPA
comment:
Table
1
(spelling
of
author
in
MRID
No.
00160000)
53
Aventis'
response:
The
author
of
MRID
No.
00160000
should
be
"Hudson
et
al.
".
Also,
it
is
not
obvious
why
the
same
reference
is
one
time
classified
"core"
and
six
times
"supplemental".
The
agency
should
reconsider
if
the
use
of
a
"supplemental"
study
(i.
e.,
rock
dove)
in
calculating
all
acute
RQ
values
is
justified.

Birds,
Chronic
Toxicity
Page:
158
(e­
version)
Paragraph:
3
EPA
comment:
Bird
kills
attributed
to
carbaryl
and
involving
blackbirds,
ducks,
starlings,
grackles
turkey,
and
cardinals
have
been
reported
in
Pennsylvania,
Virginia,
New
Jersey,
North
Carolina
and
Michigan
(#
1002048­
001,
#1000802­
001,
#1007720­
020,
##
1000799­
003,
#1004375­
004).
Aventis'
response:
The
paragraph
should
be
moved
to
the
acute
bird
section.
Also,
only
individuals
familiar
with
this
information
will
recognize
the
numbers
as
the
incident
numbers
from
the
EIIS
database.
An
appropriate
reference
should
be
inserted
here
and
in
similar
citations.

Page:
158
Paragraph:
1
Line:
2
&
3
EPA
comment:
Exposure
to
carbaryl
at
levels
equal
to
or
greater
than
1000
ppm
in
the
mallard
duck
results
in
adverse
reproduction
effects,
such
as
decrease
in
number
of
eggs
produced
include
cracked
eggs,
fertility,
embryonic
mortality,
and
hatching
success.
Aventis'
response:
The
sentence
should
be
changed.
The
embryonic
mortality
and
the
hatching
success
were
not
different
from
the
control.

Mammals,
Acute
and
Chronic
Page:
158
&
159
(e­
version)
EPA
comment:
(rat
LD50
of
307
mg/
kg,
NOAEC
80
ppm)
Aventis'
response:
The
acute
LD50
value
for
rat
should
be
corrected
to
301
mg/
kg,
and
the
chronic
NOEAC
to
75
ppm
from
the
2­
generation
rat
study.
54
Freshwater
Fish,
Acute
Page:
161
(e­
version)
EPA
comment:
Table
6
Aventis'
response:
The
study
classification
of
reference
MRID
40098001
(Mayer
&
Ellersieck,
1986)
should
be
reconsidered
(and
handled
in
a
consistent
fashion).
A
number
of
times
the
reference
is
classified
"core",
while
in
other
instances
the
classification
is
"supplemental".
The
reference
is
an
overview
article
with
little
description
of
test
methods,
analytical
procedures,
GLP,
or
study
details.
The
results
are
generally
listed
in
extensive
tables
(although
summarized
in
the
text
for
some
chemicals).
Such
a
review
article
cannot
be
regarded
as
a
"core"
study
equivalent
to
the
guideline
studies
that
have
to
be
prepared
by
registrants.
Also,
such
studies
with
insufficient
test
method
descriptions
should
not
be
used
in
a
risk
assessment
as
the
primary
source
of
information.
A
submission
based
on
such
data
would
have
certainly
been
rejected
by
the
Agency
Freshwater
Invertebrates,
Acute
Page:
163
(e­
version)
EPA
comment:
Table
9
Aventis'
response:
The
study
classification
of
reference
MRID
40098001
(Mayer
&
Ellersieck,
1986)
should
be
reconsidered
(and
handled
in
a
consistent
fashion).
A
number
of
times
the
reference
is
classified
"core",
while
in
other
instances
the
classification
is
"supplemental".
The
reference
is
a
review
article
with
little
description
of
test
methods,
analytical
procedures,
GLP,
or
study
details.
The
results
are
generally
listed
in
extensive
tables
(although
summarized
in
the
text
for
some
chemicals).
Such
an
overview
article
cannot
be
regarded
as
a
"core"
study
equivalent
to
the
guideline
studies
that
have
to
be
prepared
by
registrants.
Also,
such
studies
with
insufficient
test
method
descriptions
should
not
be
used
in
a
risk
assessment
as
the
primary
source
of
information.

Estuarine
and
Marine
Invertebrates,
Acute
Page:
165
(e­
version)
EPA
comment:
Table
13,
reference
for
glass
shrimp:
Mayer
&
Ellerersieck
Aventis'
response:
The
reference
should
be
corrected
in
Mayer
&
Ellersieck.
55
Page:
167
(e­
version)
Table
15
EPA
comment:
Table
15,
reference
for
MRID
No.
00265665
Aventis'
response:
The
reference
for
MRID
No.
00265665
should
also
contain
the
citation
of
an
author.
56
DISCUSSION
1.
Surface
Water
Concentrations
Summary
of
Registrant
Surface
Water/
Drinking
Water
Monitoring
Program
In
section
V,
page
31
EPA
states
that
the
modeling
simulations
provide
a
conservative,
though
not
unreasonable,
estimate
on
possible
concentrations
in
drinking
water.
The
data
from
the
registrant
drinking
water
monitoring
program
provide
the
best
estimate
of
concentrations
of
carbaryl
in
drinking
water.
This
study
uses
the
sampling
design
for
acute
endpoints
recommended
in
industry/
EPA
meetings
during
1999
(weekly
sampling
during
times
of
peak
concentrations
over
a
three
year
period).
Twenty
sites
representing
the
highest
carbaryl
use
areas
were
selected
based
on
the
information
provided
in
Appendix
I.
These
included
16
sites
in
agricultural
areas
and
4
locations
in
urban
areas.
Samples
were
collected
from
the
inlet
and
outlet
water
at
each
sampling
interval.
Outlet
samples
were
only
analyzed
when
residues
were
present
in
the
inlet
samples.
The
analytical
method
had
a
limit
of
quantification
of
0.030
ppb
and
a
limit
of
detection
of
0.002
ppb.

Table
1
summarizes
the
results
of
the
monitoring
at
each
of
the
20
community
water
systems.
The
maximum
concentration
observed
was
0.16
ppb
(average
of
four
samples,
the
highest
was
0.18
ppb)
in
a
finished
water
sample
from
the
Deerfield
community
water
system
located
on
the
River
Raisin
in
Lenawee
County,
Michigan.
There
were
only
five
other
samples
above
the
limit
of
quantification
of
0.030
ppb.
One
was
a
raw
water
sample
containing
0.31
ppb
from
the
Little
Potato
Slough
Mutual
community
water
system
near
Lodi
in
San
Joaquin
County,
California
(the
source
is
the
Little
Potato
Slough).
The
corresponding
finished
water
sample
was
0.007
ppb.
A
second
one
was
a
raw
water
sample
in
Brockton,
MA
which
contained
0.031
ppb.
No
detectable
residues
were
found
in
the
corresponding
finish
water
sample.
The
last
three
samples
were
from
the
Shades
Mountain
plant
of
the
Birmingham
community
water
system
on
the
Cahaba
River
in
Jefferson
County,
Alabama
.
Two
were
raw
and
finished
samples
of
0.038
and
0.032
ppb
at
the
same
sampling
interval
in
2001.
.
The
other
sample
was
0.035
ppb
in
the
raw
water
in
a
2000
sample
(the
corresponding
finished
sample
did
not
contain
carbaryl
residues.
e.
All
residues
were
transient
so
the
time
weighted
average
concentration
of
carbaryl
in
each
of
the
years
was
0.005
ppb
or
less
at
all
20
community
water
systems.
57
Table
1.
Summary
of
Results
from
the
Carbaryl
Drinking
Water
Monitoring
Study.

Site
Major
Uses
Maximum
Concentration
(ppt)
TWA
Conc.
(ppt)*
in
Outlet
Water
Inlet
Water
Outlet
Water
1999
2000
2001*
*
1999
2000
2001**
1999
2000
Manatee,
FL
citrus
9
3
ND
11
ND
NA
1
1
West
Sacramento,
CA
orchards,
nuts
3
24ND310NA1
1
Lodi,
CA
orchards,
nuts
12
31
ND
4
7
NA
1
1
Riverside,
CA
grapes,
tree
crops
8ND
ND
ND
NANA
1
1
Lake
Elsinore,
CA
citrus
ND
3
6
NA
NA
Analysis
Pending
1
1
Corona,
CA
citrus
ND
ND
ND
NA
NA
NA
1
1
Beaumont,
TX
various
agricultural
ND
ND
ND
NA
NA
NA
1
1
Point
Comfort,
TX
rice,
tree
crops
18
5ND
ND
NDNA
1
1
Penn
Yan,
NY
grapes,
apples
ND
23
ND
NA
ND
NA
1
1
Westfield,
NY
grapes,
apples
21
5
ND
ND
9
NA
1
1
Jefferson,
OR
vegetables,
strawberries
ND
10
ND
NA
ND
NA
1
1
Coweta,
OK
pecans
4
ND
***
ND
NA
***
1
1
Pasco,
WA
apples,
potatoes
2
3
ND
ND
ND
NA
1
1
Manson,
WA
apples
ND
ND
ND
NA
NA
NA
1
1
Deerfield,
MI
vegetables
10
4
ND
160
ND
NA
5
1
Brockton,
MA
cranberries
31
27
ND
ND
3
NA
1
1
East
Point,
GA
home
and
garden
18
18
4
3
8
ND
1
1
Midlothian,
TX
home
and
garden
14
ND
14
ND
NA
ND
1
1
Cary,
NC
home
and
garden
4ND
ND
ND
NANA
1
1
Birmingham,
AL
home
and
garden
23
35
38
ND
ND
32
1
1
*
Annual
Time
Weighted
Concentration,
outlet
values
substituted
for
inlet
values
when
available;
values
below
the
detection
limit
were
considered
to
be
half
the
detection
limit.
**
Results
represent
one
to
six
months
of
sampling
into
the
third
year
program.
***
No
results
available
for
the
third
year
of
sampling.
58
ND
Not
detected.
NA
No
outlet
samples
analyzed
due
to
carbaryl
residues
not
being
detected
in
inlet
samples.

Summary
of
Surface
Water
Data
from
the
NAWQA
Program
In
Section
1
page
3,
Section
4
page
28
and
in
Section
5
page
34,
EPA
has
summarized
the
available
surface
water
monitoring
data
from
the
NAWQA
program
as
having
detections
in
46%
of
the
36
NAWQA
study
units
between
1991
and
1998
with
a
maximum
concentration
of
5.5
ppb.
The
following
tables
summarize
the
carbaryl
analyses
presently
available
from
this
database.

Table
2
is
a
summary
of
the
carbaryl
detections
in
the
updated
database
analysis
recently
reported
by
Larson
(2001).
This
analysis
was
conducted
only
for
samples
collected
during
a
one­
year
period
of
the
most
intensive
sampling
from
each
of
the
sampling
sites.
Numerous
samples
were
excluded
from
this
analysis
as
described
by
Larson:

"A
few
sites
with
sufficient
sampling
for
pesticides
were
excluded
from
the
analysis,
in
order
to
minimize
bias
caused
by
over­
representation
of
a
particular
land
use
or
agricultural
setting.
…
The
sampling
requirements
for
a
site
to
be
included
in
the
analysis
were
a
minimum
of
8
samples
collected
in
6
or
more
months
during
the
1­
year
period.
In
addition,
samples
must
have
been
collected
during
the
expected
period
of
elevated
pesticide
concentrations.
At
most
of
the
sites
used
in
this
analysis,
20
to
30
samples
were
collected
during
the
selected
1­
year
period.…
Not
all
samples
collected
during
the
year
at
each
site
were
used
in
the
calculation
of
the
summary
statistics,
however.
Samples
collected
as
part
of
a
fixed­
frequency
sampling
schedule
were
included,
along
with
a
much
smaller
number
of
samples
collected
during
selected
high
or
low
flow
conditions.
Samples
collected
over
a
storm
hydrograph,
or
as
part
of
a
study
of
diurnal
variability,
were
excluded
in
order
to
avoid
bias
resulting
from
repeated
sampling
during
extreme
conditions.
"

Table
2.
Carbaryl
Detections
Reported
in
Pesticides
in
Streams
Update
(Larson,
2001)

Site
Type
Number
of
Sites
Number
of
Samples
Carbaryl
Detection
Frequency
(%)
Maximum
Estimated
Concentration
(
µ
g/
L)
All
>=
0.01
(
µ
g/
L)
>=
0.05
(
µ
g/
L)
>=
0.10
(
µ
g/
L)
Agricultural
Streams
62
1560
9.2
5.
7
1.8
0.
9
5.2
Urban
Streams
22
611
43
37
19
12
3.2
Integrator
A
31
595
15
11
2.7
1.
2
0.43
A
Large
streams
and
rivers
59
Results
in
Table
3
and
Table
4
show
a
breakdown
of
all
the
carbaryl
analyses
reported
in
the
USGS
NAWQA
database,
which
was
downloaded
from
their
web
site
July
16,
2001.
The
data
are
reported
separately
for
the
GC/
MS
and
HPLC/
PDA
analyses.

Table
3.
Frequency
of
Carbaryl
Detections
by
GC/
MS
in
Different
Concentration
Ranges
Reported
in
the
NAWQA
Database
as
of
July
16,
2001
Land
Use
Type
Number
of
Samples
<=
MDL
C
>0.003
to
0.01
>0.01
to
0.1
ppb
>0.1
to
1
ppb
>1
ppb
No.
%
No.
%
No.
%
No.
%
No.
%

All
Samples
10379
8388
80.82
617
5.94
1065
10.26
283
2.73
26
0.25
Agricultural
4349
3888
89.40
188
4.32
225
5.17
46
1.06
2
0.
05
Urban
1763
921
52.24
161
9.13
463
26.26
195
11.06
23
1.30
Mixed
A
3648
3022
82.84
247
6.77
345
9.46
33
0.90
1
0.
03
Other
B
619
557
89.98
21
3.39
32
5.17
9
1.
45
0
0
A
Large
streams
and
rivers.
Includes
all
of
the
"Integrator"
sites
listed
in
Larson,
et
al.
.,
1999
and
many
more.
B
Includes
forest,
rangeland,
mining,
etc.
C
The
method
detection
limit
(MDL)
for
carbaryl
analyzed
by
the
GC/
MS
method
is
0.003
µ
g/
L,
but
updated
MDLs
presented
in
the
database
may
be
higher
for
some
analyses
and
are
included
in
this
category.

Table
4.
Frequency
of
Carbaryl
Detections
by
LC/
PDA
in
Different
Concentration
Ranges
Reported
in
the
NAWQA
Database
as
of
July
16,
2001
Land
Use
Type
Number
of
Samples
<=
MDL
C
>0.008
to
0.01
>0.01
to
0.1
ppb
>0.1
to
1
ppb
>1
ppb
No.
%
No.
%
No.
%
No.
%
No.
%

All
Types
5516
5348
96.9
5
9
0.
16
93
1.
69
54
0.
98
12
0.
22
Agricultural
2528
2509
99.2
5
1
0.
04
13
0.
51
3
0.12
2
0.
08
Urban
1189
1064
89.4
9
4
0.
34
64
5.
38
47
3.
95
10
0.
84
Mixed
A
1523
1501
98.5
6
4
0.
26
15
0.
98
3
0.2
0
0
Other
B
276
274
99.2
8
0
0
1
0.
3610.
360
0
A
Large
streams
and
rivers.
Includes
all
of
the
"Integrator"
sites
listed
in
Larson,
et
al.
.,
1999
and
many
more.
B
Includes
forest,
rangeland,
mining,
etc.
C
The
method
detection
limit
(MDL)
for
carbaryl
analyzed
by
the
LC/
PDA
method
is
0.008
µ
g/
L,
but
updated
MDLs
presented
in
the
database
may
be
higher
for
some
analyses
and
are
included
in
this
category.
60
Summary
of
Carbaryl
Analytical
Methods
used
in
the
NAWQA
Program
In
a
number
of
instances
throughout
their
review,
EPA
has
made
reference
to
the
"poor
recovery"
for
carbaryl
noted
in
a
NAWQA
summary
document
(Larson,
1999).
In
this
document,
reference
is
made
to
mean
percent
recovery
of
24%
for
carbaryl
with
a
method
detection
limit
(MDL)
of
0.003
ppb.
The
Agency
cites
this
low
mean
recovery
several
times
as
evidence
that
the
concentrations
of
carbaryl
reported
in
the
database
widely
underestimate
the
actual
concentrations
of
carbaryl
in
the
water
samples.
This
claim
is
misleading
and
should
be
removed
from
each
location
in
the
draft
RED
for
reasons
discussed
below.

Two
analytical
methods
were
developed
as
part
of
the
NAWQA
program
and
both
of
them
have
been
used
in
the
analysis
of
carbaryl.
The
first
method,
used
for
a
majority
of
the
NAWQA
data
reported
for
carbaryl,
is
a
GC/
MS
method
with
an
MDL
of
0.003
ppb
(Zaugg,
et
al.,
1995).
The
second
method,
used
for
a
limited
number
of
samples
in
which
carbaryl
was
analyzed,
is
an
LC/
Photodiode­
Array
(PDA)
method
with
an
MDL
of
0.008
ppb
(Werner
et
al.
.,
1996).
In
the
NAWQA
database
the
quantitative
data
for
carbaryl
determined
by
the
GC/
MS
method
are
flagged
with
an
"E",
as
are
data
for
several
other
analytes,
indicating
that
the
analysts
have
noted
"the
potential
for
variable
performance"
in
the
analysis
of
carbaryl.
None
of
the
carbaryl
data
in
the
NAWQA
database
has
been
corrected
for
procedural
recoveries
that
were
noted
in
the
documents
described
above.
Both
of
these
methods
are
discussed
below
in
relation
to
the
recoveries
found
for
the
methods
and
the
potential
impact
this
could
have
on
the
analytical
data
for
carbaryl.

Gas
Chromatography/
Mass
Spectroscopy
Method
The
analytical
method
most
used
in
the
NAWQA
program
for
the
analysis
of
carbaryl
in
water
samples
is
the
GC/
MS
method
described
by
Zaugg,
et
al.,
1995.
In
this
multi­
residue
method,
the
analytes
are
first
removed
from
the
water
sample
by
sorption
on
a
C­
18
solid
phase
and
are
subsequently
eluted
from
the
solid
phase,
separated
by
GC
and
quantified
by
mass
spectroscopy
with
selected
ion
monitoring.
The
identity
of
each
analyte
is
confirmed
by
the
appropriate
combination
of
retention
time
and
the
ratios
of
three
mass
ions
that
are
characteristic
for
the
analyte.

The
recoveries
for
carbaryl
spiked
at
different
levels
into
three
different
types
of
water
and
analyzed
by
the
GC/
MS
method
are
shown
in
Table
5
Mean
percent
recoveries
of
151
and
202%
were
found
for
carbaryl
fortified
at
0.1
and
1.0
µ
g/
L
in
reagent
grade
water.
A
preliminary
MDL
of
0.046
µ
g/
L
was
calculated
for
the
0.1
µ
g/
L
spiking
level.
Mean
percent
recoveries
of
10
and
75%
were
found
for
carbaryl
fortified
at
0.1
and
1.0
µ
g/
L
in
a
surface
water
sample
collected
from
the
South
Platte
River.
However,
carbaryl
was
detected
at
0.18
µ
g/
L
in
this
water,
or
nearly
twice
the
low
spike
level,
raising
questions
about
the
validity
of
this
result.
Mean
percent
recoveries
of
94
and
86%
were
found
for
carbaryl
fortified
at
0.1
and
1.0
µ
g/
L
in
a
ground
water
sample
collected
from
a
well
in
Denver.
A
mean
recovery
value
of
24%
was
reported
for
reagent­
grade
water
fortified
at
a
level
of
0.03
µ
g/
L
with
a
method
detection
limit
calculated
at
0.003
µ
g/
L.
61
Table
5.
Recovery
and
Precision
for
Multiple
Determinations
of
Carbaryl
in
GC/
MS
Method
for
Carbaryl
Spiked
in
Different
Water
Samples
Water
type
Spike
Concentration
(
µ
g/
L)
Mean
Recovery
(%)
MDL
Calculated
Reagent
Grade
0.
1
151
0.046
Reagent
Grade
1.
0
202
Surface
A
0.1
10
Surface
A
1.0
75
Ground
B
0.1
94
Ground
B
1.0
86
Reagent
Grade
0.
03
24
0.
003
A
Surface
water
was
collected
from
the
South
Platte
River
near
Henderson,
Colorado.
This
water
was
found
to
contain
significant
concentrations
of
several
pesticides
including
0.18
µ
g/
L
carbaryl.
This
concentration
was
subtracted
from
the
values
determined
to
give
corrected
results.
B
Ground
water
was
collected
from
the
Denver
Federal
Center
Well
15.

Whereas
the
values
reported
by
Zaugg,
et
al.
(1995)
are
of
interest
in
validating
the
analytical
method,
they
are
not
as
useful
in
evaluating
the
validity
of
the
data
contained
in
the
NAWQA
database.
Therefore,
quoting
the
mean
recovery
value
of
24%
for
reagent
grade
water
spiked
with
carbaryl
at
0.03
µ
g/
L
as
evidence
that
the
concentrations
reported
in
the
database
underestimate
the
actual
concentrations
of
carbaryl
present
in
the
water
samples
is
misleading.
A
more
useful
measure
of
the
validity
of
the
values
in
the
database
lies
with
the
quality
control
checks
that
have
been
incorporated
into
the
analysis
of
samples
in
the
NAWQA
program.

In
a
preliminary
report,
Martin
(1999)
reported
the
quality
control
data
collected
as
part
of
the
NAWQA
surface
and
ground
water
programs
by
the
1991
NAWQA
Study
Unit
teams
or
the
National
Water
Quality
Laboratory
(NWQL)
during
1992
to
1996.
The
data
that
were
compiled
includes
field
blanks,
laboratory
control
spikes
and
field
matrix
spikes,
which
are
defined
below
by
Martin.

"Field
blanks
were
collected
at
the
field
site
with
pesticide­
grade
blank
water
and
are
exposed
to
the
field
and
laboratory
environments
and
equipment
similarly
to
environmental
samples.
Field
blanks
measure
the
frequency
and
magnitude
of
contamination
(one
type
of
positive
bias)
in
environmental
water
samples
from
sources
in
the
field
and/
or
laboratory.
Contamination
is
the
main
cause
of
falsepositive
detections
(detecting
a
pesticide
in
a
sample
when,
in
truth,
it
is
absent)."

"Laboratory
control
spikes
measure
the
bias
and
variability
of
the
analytical
method
at
a
particular
concentration.
One
laboratory
control
spike
is
measured
in
each
analytical
set
of
environmental
samples.
The
laboratory
control
spike
has
62
the
target
pesticides
spiked
into
pesticide­
grade
blank
water
at
the
laboratory
and
extracted,
processed,
and
analyzed
like
environmental
samples.
Laboratory
control
spikes
analyzed
by
GCMS
were
spiked
at
0.1
:
g/
L…"

"Field
matrix
spikes
measure
the
bias
and
variability
of
the
analytical
method
PLUS
any
potential
effects
caused
by
(1)
degradation
of
pesticides
during
shipment
to
the
laboratory,
(2)
inferences
in
the
determination
of
pesticides
from
unusual
characteristics
of
the
environmental
water
sample
("
matrix
effects"),
and
(3)
other
chemical
processes
that
cause
bias
or
variability
in
the
measurements
of
pesticides
in
environmental
water
samples.
Field
matrix
spikes
analyzed
by
GCMS
were
spiked
at
0.1
:
g/
L,…"

All
of
the
carbaryl
analyses
in
the
field
blanks,
field
matrix
spikes
and
lab
control
spikes
were
conducted
following
the
same
method
described
by
Zaugg
et
al.,
1995
that
was
used
to
generate
a
majority
of
the
carbaryl
data
contained
in
the
NAWQA
database.
The
data
below
were
excerpted
from
Tables
1
to
4
of
the
Martin
report.
Carbaryl
is
found
in
these
tables
under
parameter
82680.

Out
of
145
samples
taken
as
ground
water
field
blanks,
carbaryl
was
not
detected
in
any
of
the
samples
indicating
a
lack
of
false
positives.
Out
of
171
samples
taken
as
surface
water
field
blanks,
carbaryl
was
reported
in
two
samples
(1.2%
false
positives)
at
reported
concentrations
of
0.009
and
0.012
µ
g/
L.

A
summary
of
the
results
for
the
field
matrix
spikes
and
the
lab
control
spikes
is
presented
in
Table
6
Mean
recovery
for
the
306
field
matrix
spikes
was
115%
of
the
spiking
level
of
0.1
µ
g/
L
with
a
median
recovery
of
94.4%
and
a
90
th
percentile
recovery
of
200%.
This
indicates
the
potential
for
the
method
to
over­
estimate
the
concentration
of
carbaryl
present
in
the
water
samples
and
is
consistent
with
the
initial
data
reported
for
the
reagent
water
samples
by
Zaugg
et
al.
(1995).
Mean
recovery
for
the
1000
lab
control
spikes
was
99.6%
of
the
spiking
level
of
0.1
µ
g/
L
with
a
median
recovery
of
93%
and
a
90
th
percentile
recovery
of
185%.
These
data
suggest
an
adequate
level
of
detection
of
carbaryl
in
QC
samples
that
were
analyzed
as
part
of
the
same
process
used
in
the
NAWQA
survey
of
pesticides
in
surface
and
ground
water.
63
Table
6.
Percent
Recoveries
of
Carbaryl
Detected
by
the
NAWQA
GC/
MS
Method
in
Laboratory
Control
Spikes
and
Field
Matrix
Spikes
at
a
Spiking
level
of
0.1
µ
g/
L
Sample
Type
Number
of
Samples
10
th
Percentile
Recovery
(%)
Median
Recovery
(%)
Mean
Recovery
(%)
90
th
Percentile
Recovery
(%)
Maximum
Recovery
(%)

Field
Matrix
Spike
306
40
94.4
115.0
199.9
456
Laboratory
Control
Spike
1000
20
93.0
99.6
185.1
329
The
following
disclaimer
was
taken
verbatim
from
the
provisional
report
by
Martin
(1999)
and
pertains
to
the
data
provided
above
on
the
recovery
of
carbaryl
in
the
field
matrix
spike
samples.
"The
field
matrix­
spike
data
have
not
been
reviewed
thoroughly,
are
provisional,
and
are
subject
to
change.
Further
review
of
the
field­
spike
data
is
expected
to
identify
spikes
that
have
extremely
high
or
low
recoveries
because
the
spikes
either
were
improperly
collected
or
incorrectly
documented
in
the
NAWQA
QC
data
base.
The
expected
result
of
further
review
is
a
data
set
of
field
matrix
spikes
with
fewer
extreme
values
than
the
provisional
data
set
described
in
this
paper;
consequently,
the
provisional
data
set
provides
a
conservative
estimate
of
the
quality
of
the
NAWQA
pesticide
data.
Interpretations
of
field
matrix
spike
data
in
this
paper
are
not
expected
to
change
greatly
as
a
result
of
further
review
of
the
data,
however,
the
statistics
and
confidence
limits
reported
in
the
text
and
tables
will
change
on
further
review
(especially
for
pesticides
with
low
numbers
of
field
spikes
[less
than
50])."
64
High­
Performance
Liquid
Chromatography/
Photodiode­
Array
Method
Another
analytical
method
used
in
the
NAWQA
program
for
the
analysis
of
carbaryl
in
water
samples
is
the
LC/
PDA
method
described
by
Werner,
et
al.,
1996.
This
method
was
used
for
the
analysis
of
carbaryl
in
a
limited
number
of
samples
as
noted
above.
In
this
multi­
residue
method,
the
analytes
are
first
removed
from
the
water
sample
by
sorption
on
a
Carbopak­
B
solid
phase
extraction
cartridge
and
are
subsequently
eluted
from
the
solid
phase,
separated
by
HPLC
and
quantified
by
light
absorption
using
a
photodiode­
array
detector.
The
identity
of
each
analyte
is
confirmed
by
the
appropriate
combination
of
retention
time
and
light
absorption
characteristics.
The
recoveries
for
carbaryl
spiked
at
different
levels
into
three
water
samples
and
analyzed
by
this
method
is
shown
in
Table
7
The
recoveries
ranged
from
58%
to
84%
for
the
different
water
and
spiking
levels.
Laboratory
control
spikes
in
organic­
free
water
resulted
in
a
mean
recovery
of
61%
over
a
two­
year
sampling
period.
These
results
indicate
reasonable
levels
of
carbaryl
recovery
from
each
of
the
different
types
of
water
evaluated
for
the
method.

Table
7.
Recovery
and
Precision
for
Multiple
Determinations
of
Carbaryl
in
LC/
PDA
Method
for
Carbaryl
Spiked
in
Different
Water
Samples
Water
type
Spike
Concentration
(
µ
g/
L)
Mean
Recovery
(%)
MDL
Calculated
Organic­
Free
0.1
82
0.
008
Organic­
Free
1.0
70
Surface
A
0.1
84
0.
016
Surface
A
1.0
84
Ground
B
0.1
58
0.
018
Ground
B
1.0
64
Organic
Free
0.5
61
C
A
Surface
water
was
collected
from
the
South
Platte
River
at
Englewood,
Colorado.
B
Ground
water
was
collected
from
Jefferson
County,
Colorado
(Arvada
Well
14).
C
National
Water
Quality
Laboratory
results
produced
using
5
operators
and
7
instruments
over
2
years
(about
350
data
points).

Summary
of
Surface
Water
Data
from
the
California
DPR
Surface
Water
Database
In
Section
5
pages
34
to
37
EPA
has
summarized
surface
water
monitoring
data
from
various
sources.
One
source
not
included
in
this
discussion
is
the
California
Surface
Water
Monitoring
Database.
The
number
of
analyses
and
the
detections
of
carbaryl
residues
reported
in
the
database
are
summarized
in
Table
8.
Carbaryl
was
detected
at
levels
above
the
LOQ
in
only
5.1%
of
the
2,690
samples
analyzed.
The
mean
concentration
of
carbaryl
in
the
140
samples
above
the
LOQ
was
0.42
ppb.
The
highest
concentration
of
carbaryl
that
was
detected
was
8.4
ppb.
65
An
analysis
of
the
data
in
the
California
Department
of
Pesticide
Regulation's
surface
water
database
as
of
July
15,
2000
was
conducted
for
carbaryl.
The
following
summary
of
the
contents
of
the
database
is
adapted
from
information
provided
by
the
California
DPR.
The
database
contains
monitoring
results
for
pesticides
in
samples
taken
from
California
rivers,
creeks,
urban
streams,
agricultural
drains,
the
Delta,
and
urban
stormwater
runoff.
As
of
July
15,
2000,
the
database
contained
the
results
of
30
studies
conducted
by
federal,
state,
and
local
agencies,
private
industry,
and
an
environmental
group.
A
total
of
4,660
samples
were
taken
in
16
counties
from
January
1991
through
March
2000.
Each
record
in
the
database
is
the
result
of
one
analysis
for
a
pesticide
active
ingredient
or
breakdown
product.
The
database
contains
a
total
of
92,296
analytical
records.
Only
information
on
the
analytical
detection
of
carbaryl
in
these
water
samples
is
summarized
in
Table
8
below.

Table
8.
Carbaryl
Detections
Reported
in
California
DPR
Surface
Water
Monitoring
Database
Land
Use
Type
Number
of
Samples
<=
LOQ
>0.003
to
0.01
>0.01
to
0.1
ppb
>0.1
to
1
ppb
>1
ppb
No.
%
No.
%
No.
%
No.
%
No.
%

All
Samples
2690
2553
94.9
1
13
0.48
55
2.04
55
2.04
14
0.52
Concentrations
of
analytical
results
that
are
reported
below
the
limit
of
quantification
are
reported
as
a
zero
in
the
database
concentration
field.
The
LOQs
for
the
different
methods
used
to
generate
the
data
contained
in
the
database
ranged
from
0.003
to
0.5
µ
g/
L,
with
a
majority
of
the
samples
analyzed
with
an
LOQ
of
0.05
µ
g/
L
or
less
(Table
9)
.

Table
9.
Limits
of
Quantification
for
Carbaryl
Analytical
Methods
Reported
in
California
DPR
Surface
Water
Monitoring
Database
LOQ
(
µ
g/
L)
0.003
0.041
0.044
0.05
0.07
0.1
0.
5
Number
of
Samples
Analyzed
267
238
168
1353
92
53
146
66
2.
Ground
Water
Concentrations
In
Section
5
page
34
EPA
summarized
information
on
the
detection
of
carbaryl
in
groundwater
from
the
EPA
Pesticides
in
Groundwater
Database,
the
EPA
STORET
database
and
the
NAWQA
database.
Each
of
the
databases
shows
a
pattern
of
very
low
levels
of
carbaryl
detection
in
few
groundwater
resources.
These
analyses
confirm
several
statements
made
by
the
Agency
that
carbaryl
has
limited
potential
to
impact
groundwater
resources.
However,
on
page
2
of
the
Memorandum
issued
June
28,
2001,
in
conjunction
with
the
EFED
RED
chapter
for
carbaryl,
EPA
is
requiring
additional
information
on
"Surface
and
groundwater
monitoring
in
urban
and
suburban
use
areas
(non­
guideline)."
Based
on
the
characteristics
of
carbaryl
and
the
available
data
demonstrating
limited
impact
of
carbaryl
on
ground
water
resources,
additional
studies
to
evaluate
the
potential
for
carbaryl
to
contaminate
groundwater
are
unnecessary
and
unwarranted.

Summary
of
Ground
Water
Data
from
the
NAWQA
Program
In
Section
5,
pages
33
­
34,
EPA
has
summarized
ground
water
monitoring
data
available
for
carbaryl.
The
database
that
contains
the
most
extensive
evaluation
of
the
impact
of
the
most
recent
uses
of
carbaryl
on
ground
water
is
the
NAWQA
database.
One
deficiency
of
the
NAWQA
program
is
that
samples
are
targeted
to
agricultural
and
urban
areas
but
not
to
areas
treated
with
the
specific
chemical
being
analyzed.
However,
given
the
use
patterns
of
carbaryl,
the
use
of
carbaryl
has
certainly
occurred
near
a
number
of
these
wells.
Another
deficiency
is
that
when
residues
are
found,
that
while
they
may
be
representative
of
residues
in
ground
water,
they
may
not
be
representative
of
residues
in
ground
water
used
for
drinking
water
due
to
the
location
of
the
sampled
wells
relative
to
potable
drinking
water
wells.

EPA
cited
a
1998
review
of
the
NAWQA
database
by
Kolpin
and
stated:

"Carbaryl
was
detected
at
greater
than
the
detection
limit
(0.003
:
g/
L)
in
1.1
%
of
groundwater
samples
from
1034
sites
across
the
U.
S.
by
U.
S.
G.
S.
NAQWA
(sic)
program.
The
maximum
observed
concentration
was
0.021
:
g/
L."

This
1998
analysis
has
been
extended
by
additional
study
data
collected
by
the
NAWQA
program.
The
additional
data
continue
to
show
a
limited
number
of
low
level
detections
of
carbaryl
in
ground
water
samples.
Table
10
below
summarizes
a
more
recent
provisional
review
by
Kolpin
(2001)
of
the
updated
NAWQA
database.
Not
all
of
the
water
samples
were
used
to
calculate
the
summary
statistics
as
noted
by
Kolpin:

"To
preclude
bias
in
these
summary
statistics
from
wells
that
were
sampled
more
than
once,
the
data
set
was
condensed
such
that
each
well
had
a
single
pesticide
analysis.
This
generally
was
the
first
sample
collected.
However,
subsequent
samples
were
selected
if
these
samples
contained
more
pesticide
data
(i.
e.,
a
larger
number
of
pesticides
were
analyzed).
Wells
that
were
designed
to
be
a
part
of
both
a
land­
use
study
and
a
major
aquifer
survey
were
used
in
each
summary.
67
Because
of
uncertainties
in
the
source
of
water
and
contributing
land­
use
area,
springs
and
drains
were
excluded
from
these
summaries."

Table
10.
Carbaryl
Detections
Reported
in
Pesticides
in
Ground
Water
Update
(Kolpin,
2001)

Site
Type
Number
of
Samples
Carbaryl
Detection
Frequency
(%)
Maximum
Estimated
Concentration
(
µ
g/
L)
All
>=
0.01
(
µ
g/
L)
>=
0.05
(
µ
g/
L)
>=
0.10
(
µ
g/
L)
Agricultural
LandUse
Wells
1244
0.40
0.16
0.0
0.
0
0.019
Urban
Land­
Use
Wells
634
2.1
1.
3
0.0
0.
0
0.031
Major
Aquifers
1849
0.59
0.54
0.05
0.05
0.539
68
REFERENCES
Burgos,
W.
D.,
J.
T.
Novak
and
D.
F.
Berry.
1996.
Reversible
Sorption
and
Irreversible
Binding
of
Naphthalene
and
 
­Naphthol
to
Soil:
Elucidation
of
Processes.
Environ.
Sci.
Technol.,
30:
1205­
1211.

Burgos,
W.
D.,
D.
F.
Berry,
A.
Bhandair,
and
J.
T.
Novak.
1999.
Impact
of
Soil­
Chemical
Interactions
on
the
Bioavailability
of
Naphthalene
and
1­
Naphthol.
Water
Research,
33:
3789­
3795.

Hassett,
J.
J.,
W.
L.
Banwart,
S.
G.
Wood,
and
J.
C.
Means.
1981.
Sorption
of
 
­naphthol:
Implications
concerning
the
limits
of
hydrophobic
sorption.
Soil
Sci.
Soc.
Am.
J
45(
1):
38­
42.

Kolpin,
D.
W.
June
11,
2001.
Pesticides
in
Ground
Water,
Summary
statistics;
Results
of
the
National
Water
Quality
Assessment
Program
(NAWQA),
1992­
1998.
Available
for
download
from
http://
water.
wr.
usgs.
gov/
pnsp/
pestgw/.

Larson,
S.
J.
June
11,
2001.
Pesticides
in
Streams,
Summary
statistics;
Results
of
the
National
Water
Quality
Assessment
Program
(NAWQA),
1992­
1998.
Available
for
download
from
http://
water.
wr.
usgs.
gov/
pnsp/
pestsw/.

Martin,
J.
D.
October
27,
1999
.Quality
of
Pesticide
Data
for
Environmental
Water
Samples
Collected
for
the
National
Water­
Quality
Assessment
Program,
1992­
96
and
Examples
of
the
Use
of
Quality­
Control
Information
in
Water­
Quality
Assessments.
Available
for
review
at:
http://
water.
wr.
usgs.
gov/
pnsp/
rep/
qcsummary/

McCall,
P.,
D.
Laskowski,
R.
Swann,
and
H.
Dishburger.
1980.
Measurement
of
Sorption
Coefficients
of
Organic
Chemicals
and
Their
Use
in
Environmental
Fate
Analysis.
In:
Test
Protocols
for
Environmental
Fate
&
Movement
of
Toxicants.
Proceedings
of
a
Symposium,
Association
of
Official
Analytical
Chemists,
94th
Annual
Meeting,
October
21,
22,
1980.
pp
89­
109.

Werner,
I.,
L.
A.
Denovic,
V.
Conner,
V.
De
Vlaming,
H.
Bailey
and
D.
E.
Hinton.
2000.
Insecticide­
Caused
Toxicity
to
Ceriodaphnia
dubia
(Cladocera)
in
the
Sacramento­
San
Joaquin
River
Delta,
California.
Environmental
Toxicology
and
Chemistry,
19:
215­
227.

Werner,
S.
L.,
M.
R.
Burkhardt
and
S.
N.
DeRusseau.
1996.
METHODS
OF
ANALYSIS
BY
THE
U.
S.
GEOLOGICAL
SURVEY
NATIONAL
WATER
QUALITY
LABORATORY—
DETERMINATION
OF
PESTICIDES
IN
WATER
BY
CARBOPAK­
B
SOLIDPHASE
EXTRACTION
AND
HIGH­
PERFORMANCE
LIQUID
CHROMATOGRAPHY.
U.
S.
Geological
Survey
Open­
File
Report
96­
216,
42
pp.
Available
for
download
from
http://
wwwnwql.
cr.
usgs.
gov/
Public/
pubs/
OFR96­
216/
OFR96­
216.
html.
69
Zaugg,
S.
D.,
M.
W.
Sandstrom,
S.
G.
Smith
and
K.
M.
Fehlberg.
1995.
METHODS
OF
ANALYSIS
BY
THE
U.
S.
GEOLOGICAL
SURVEY
NATIONAL
WATER
QUALITY
LABORATORY—
DETERMINATION
OF
PESTICIDES
IN
WATER
BY
C­
18
SOLID­
PHASE
EXTRACTION
AND
CAPILLARY­
COLUMN
GAS
CHROMATOGRAPHY/
MASS
SPECTROMETRY
WITH
SELECTED­
ION
MONITORING.
U.
S.
Geological
Survey
OpenFile
Report
95­
181,
49
pp.
Available
for
download
from
http://
wwwnwql.
cr.
usgs.
gov/
Public/
pubs/
OFR95­
181/
OFR95­
181.
html.
70
Confidential
Business
Attachment
APPENDIX
1
Surface
Water
Monitoring
for
Residues
of
Carbaryl
in
High
Use
Areas
of
the
United
States
(Stone
Environmental,
Inc.
Report
#99­
1005­
F)
(hard
copy
provided).
71
Confidential
Business
Attachment
APPENDIX
2
Calculation
of
County
Average
Carbaryl
Use
Rates
(hard
copy
provided)
