United States
Environmental Protection
Agency
Office of
Pesticides and Toxic Substances
Washington, DC 20460
July 1985
Pesticides
SERA
o
o
•o
o
c
3
a
o
00
o
CO
u
CD
O
si!
33
CD
•••
(D
•o
O
(A
E
o
D
o
CD
3
Compound 1080
Special Review
Document 4
-------
COMPOUND 1080
POSITION DOCUMENT 4
U. S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
OFFICE OF PESTICIDE PROGRAMS
401 M STREET, SW
WASHINGTON, D.C. 20460
July 1985
-------
TABLE OF CONTENTS
Page
Executive Summary i
I. Introduction 1
A. Background 1
B. Position Document 1 2
C. Summary of Actions Proposed in PD 2/3 3
D. Organization of Position Document 4 5
II. Comments and Agency Response on the PD 2/3 6
Risk Assessment
A. General Comments on Risk Assessment 6
1. Scientific Advisory Panel 6
2. Wyoming Department of Agriculture 7
3. Montana Department of Agriculture 8
B. Lack of Emergency Treatment 9
1. Agency Position in PD 2/3 9
2. Comments on PD 2/3 9
3. Agency Response 9
C. Acute Toxicitv to Mammals and Birds 10
1. Agency Position in PD 2/3 10
2. Comments on PD 2/3 11
3. Agency Response 11
n. Significant Adverse Effects on Populations of 12
Nontarget Organisms
1. Agency Position in PD 2/3 12
a. General
b. Direct Risks to Nontarget Species
c. Indirect Risks to Nontarget Species
d. Community Level Impacts
e. Relation of Bait Formulations to Risks
2. Comments on PD 2/3 17
a. The Hegdal Study (1979)
b. Predator/Prey Relationship
c. Relationship of Bait Formulations to Risk
d. Other Comments
-------
3. Agency Response 21
a. The Hegdal Study
b. Predator/Prey Relationship
c. Relationship of Bait Formulations
to Risk
d. Other Comments
E. Endangered Species ^
95
1. California Condor z
a. Agency Position in PD 2/3
b. Comments
c. Agency Response
2. San Joacruin Kit Fox
a. Agency Position in PD 2/3
b. Comments
c. Agency Response
3. Black-footed ferret 33
a. Agency Position in PD 2/3
b. Comments
c. Agency Response
4. Aleutian Canada Goose 34
a. Agency Position in PD 2/3
b. Comments
c. Agency Response
III. Comments and Agency Response on PD 2/3 36
Benefits Analysis
A. Agency Position in PD 2/3 36
B. Comments on PD 2/3 37
C. Agency Response 38
D. New Information 40
IV. Comments and Agency Response on Proposed 41
Regulatory Actions in Position Document
A. Rangeland, Cropland, and Non-agricultural 41
Sites: Ground Squirrels
1. Proposed Actions in PD 2/3 41
2. Comments on PD 2/3 42
3. Agency Response 44
-------
B. Rangeland and Cropland: Prairie Dogs 47
1. Proposed Actions in PD 2/3 47
2. Comments on PD 2/3 47
3. Agency Response 49
C. Rangeland, Cropland, and Non-agricultural 50
Sites Except Around Ships and Buildings:
Other Rodents
1. Proposed Actions in PD 2/3 50
2. Comments on PD 2/3 51
3. Agency Response 52
D. Use In and Around Ships and Buildings: 52
Norway Rat, Roof Rat, and House Mice
E. General Comments on Proposed Actions 52
V. Final Regulatory Decision 54
A. General 54
B. Rangeland, Cropland, and Non-agricultural 56
Sites: Ground Sguirrels
1. Reduction and Standardization of Bait 56
Concentration and Application Rates
2. Standardization of Hand Baiting and 57
Post-Baiting Procedures
3. Standardization of Aerial Baiting and 57
Post-Baiting Procedures
4. Special Restrictions for Endangered 58
Species Protection
5. Other Label Statements 58
C. Rangeland and Cropland: Prairie Dogs 58
D. Rangeland, Cropland, and Non-agricultural 59
Sites Except Around Ships and Buildings:
Other Rodents
E. Use In and Around Ships and Buildings: 60
Norway Rat, Roof Rat, and House Mice
F. Summary of Regulatory Reguirements 60
1. General Requirements 60
a. Hand Baiting and Post-Baiting Procedures 60
b. Nontarget Species Label Precaution 61
c. Restricted Use Classification Label 61
Statement
d. Bait Dyes 62
-------
2. Rangeland, Cropland and 62
Non-agricultural Sites: Ground Squirrels
a. Bait Concentration Levels 62
b. Aerial Baiting and Post-Baiting 62
Procedures
c. Endangered Species Label Precautions 63
Rangeland and Cropland: Prairie Dogs
64
a. Bait Concentration Levels 64
b. Use Supervision 64
c. Precontrol Survey for the 64
Black-footed Ferret
d. Endangered Species Label Precautions 65
4. Rangeland, Cropland, and Non-agricultural 65
Sites Except Around Ships and Buildings:
Other Rodents
a. Bait Concentration Levels 65
b. Endangered Species Label Precautions 66
5. Pocket Gophers 66
5. Use In and Around Ships and Buildings: 66
Norway Rat, Roof Rat, and House Mice
VI. Bibliography 67
Appendix A
-------
Executive Summary
The Rebuttable Presumption Against Registration (RPAR)
notice for Compound 1080 (1080) was published in the Federal
Register on December 1, 1976 (41 PR 52792). The presumption
was against the continued registration of all pesticide
products containing 1080 used for rodent control. Compound
1080 was also used for predator control, principally coyotes,
until 1972. At that time the predacidal uses of 1080 were
cancelled. Following a hearing in 1983, EPA determined that
significant new evidence warranted modifications of the 1972
cancellation with respect to the use of 1080 in livestock
protection collars and single lethal dose baits. A
registration for the livestock protection collar was issued
to the Department of the Interior on July 18, 1985.
The RPAR criteria that were determined to have been met
for the rodenticidal uses of 1080 were 1) lack of emergency
treatment, 2) acute toxicity to mammals and birds, and 3)
significant reduction in populations of nontarget organisms
and fatalities to members of endangered species.
Position Document 2/3 (PD 2/3) detailing the Agency's
proposed decision for concluding the RPAR was issued in June,
1983. In PD 2/3 the major actions proposed by the Agency
were the modification in the terms and conditions of use of
1080 to control ground squirrels, prairie dogs and meadow
mice and the denial of the use of 1080 to control several
other species of field rodents. A Federal Register notice
soliciting public comments on PD 2/3 was published on
November 4, 1983 (48 FR 50935). The Agency received comments
on the PD 2/3 from state and county governments, individual
ranchers and farmers, the U.S. Department of Agriculture,
and the Scientific Advisory Panel. All of the comments have
been evaluated and the information has been used in the
development of the Agency's final decision presented in this
document.
The principal use of 1080 is for the control of field
rodents. It is available for this use in the following four
states: California, Colorado, Nevada and Oregon. California
uses over 80% of the 1080 used as a field rodenticide, with
the bulk of the use for the control of ground squirrels.
There are no Federal registrations for any field rodent uses
of 1080, except for a "Special Local Needs" registration
granted under § 24(c) of FIFRA to one county in Oregon.
There is one Federally registered technical product that
contains a use for 1080 to control certain rodents in and
around ships and buildings. California, Colorado and Nevada
use 1080 for field rodent control because they have valid
-------
-11-
"intrastate" registrations. Other states who at one time
used 1080 for field rodent control relied on the U.S. Department
of the Interior's 1080 registration. This registration was
withdrawn following an Executive Order prohibiting the use or
1080 on Federal lands.
The Agency has concluded that a variety of regulatory
measures can be imposed on current intrastate producers and
registrants that will allow the continued use of 1080. A
major condition for continued use will be the requirement
that all intrastate producers, the § 24(c) registrant and the
registrant of technical 1080, submit all data necessary to fully
support Federal registration under §3 of FIFRA. Applications
for Federal registration or amended registration must be
submitted to EPA within 30 days after the publication of the
Federal Register notice announcing the availability of PD 4
and the Agency's intent to cancel registrations that do not
comply with the terms of the notice, or within 30 days of
receipt of written notice to submit an application, which
ever occurs later. The data requirements for registration
are contained in Appendix A of this document.
In addition to requiring a full complement of data, the
Agency is requiring the following regulatory measures to
provide protection for endangered and nontarget species while
data are generated:
0 Utilization of the existing cooperative program
between the California Department of Food and
Agriculture, Department of Fish and Game, and County
Agricultural Commissioners for protection of endangered
species in California.
0 Direct supervision of 1080 field use by state
governmental agency personnel in the range of those
endangered species identified in this document.
0 Limiting use to Certified Applicators or persons
under their direct supervision.
0 Changing product labels to include endangered species
and nontarget precautions, and revised baiting and post-
baiting procedures.
0 Lowering 1080 bait concentrations in the range of the
California condor and for prairie dog control (which
impacts the endangered black-footed ferret).
0 Not registering any new product for field rodent control
until the data necessary to support Federal registration
under § 3 of FIFRA has been submitted and determined to be
adequate.
-------
-111-
A summary of Agency actions and requirements for ground
squirrels, prairie dogs and other rodents is presented at the
end of this Executive Summary.
All current uses of 1080 as a rodenticide may be continued
under certain conditions, with the possible exception of the
use of 1080 in the range of the California condor. Use of
1080 in the range of the condor will be determined prior to
January 1, 1986 following discussion with the Office of
Endangered Species (OES). The Agency recently received a
biological opinion from OES stating that use of the 1080
livestock protection collar in the range of the California
condor would jeopardize the continued existence of this species,
The OES opinion emphasized that there are only a few
condors in existence and there is uncertainty about the
effects of the collar on the condor. Although there are
substantial differences in potential 1080 exposure between
the livestock protection collar and treated grain bait, the
drastic decline in the condor population (OES reported one
source estimated 9 to 11 condors in the wild) may require
prohibition of 1080 use in the range of the condor until
safety data are evaluated by the Agency. If use is permitted
in the range of the condor, it will be at a maximum bait
concentration of .02% beginning in 1986. This level will be
in force until data demonstrating that the .02% level is
ineffective are submitted as well as data establishing the
lowest effective bait concentration. Bait concentrations
higher than .02% will have to be assessed to determine if
they pose risks to the condor.
Because of the potential risk to the endangered black-
footed ferret, the Agency is requiring that the use of 1080
to control prairie dogs be at a maximum concentration of .02%
beginning in 1986. This level will be in force until data
demonstrating that the .02% level is ineffective are submitted
as well as data establishing the lowest effective bait con-
centration. Bait concentrations higher than .02% will have
to be assessed to determine if they pose risks to the black-
footed ferret.
Intrastate producers and the § 24(c) registrant will be
able to sell, distribute and use 1080 at current bait concen-
trations and application rates until field efficacy data are
generated, except as noted in the range of the California
condor and for control of prairie dogs. Depending on the
results of efficacy testing, the Agency may require bait
concentrations for additional products to be lowered while
other tests necessary to support Federal registration are
completed.
-------
-IV-
As other data for Federal registration are submitted and
reviewed by the Agency, additional modifications in the ^®rms
and conditions of use may be required. After reviewing tne
data, the Agency may also determine that some or all uses or
a particular product should not be permitted and will cancel
registrations or deny application for Federal registration,
as appropriate, for those uses in question.
Other states, government agencies, or companies applying
for a 1080 registration for field rodent control will be
required to generate a full complement of data necessary to
support Federal registration prior to the use of 1080. The
Agency believes it is necessary that expansion of the use of
1080 to other states be predicated on an adequate data base
to support such expansion.
SUMMARY OF ACTIONS
Ground Squirrels- California, Colorado, Nevada, Oregon -
all use sites.
"Allow use in 1985 under current label conditions.
""Call in" intrastate products and require data necessary
to support Federal registration under § 3 of FIFRA.
Oregon's § 24(c) would also have to meet § 3 data
requirements. (See Appendix A).
°Allow products to be used at current bait concentrations
and application rates while data are being generated, except
in the range of the California condor. If permitted, use
after 1985 will be at a maximum bait concentration of
.02% in the condor's range unless field efficacy data are
submitted demonstrating that the .02% level is not effective
as well as data establishing the lowest effective bait
concentration. The Agency must determine if higher bait
concentrations will impact the condor. A decision on the
use of 1080 grain baits in the range of the condor will be
made prior to January 1, 1986.
°Applicants for Federal registration and current registrants
must begin studies as soon as possible to determine the
lowest effective bait concentration.
°Require label changes in baiting and post-baiting procedures,
add endangered species and nontarget precautions, and restricted
use labeling.
"Withdraw proposed geographic and seasonal use limitations in
California. Rely on California state programs approved by
EPA.
-------
-V-
Prairie Dogs- Colorado - all use sites.
"Allow use during 1985 under current label conditions.
011 Call in" the Colorado intrastate product and require data
necessary to support Federal registration under § 3 of
FIFRA. (See Appendix A).
"Use after 1985 will be at a maximum bait concentration of
.02% unless field efficacy data are submitted demonstrating
that the .02% level is not effective, as well as data
establishing the lowest effective bait concentration.
The Agency must also determine if higher bait concentrations
will impact the black-footed ferret.
"Applicants for Federal registration must begin studies as
soon as possible to determine the lowest effective bait
concentration.
"Allow use after 1985 only if a validated precontrol survey
for the black-footed ferret is approved by the Agency.
"Restrict use to governmental agencies or persons under
the direct supervision of members of governmental agencies.
"Require label changes in baiting and post-baiting procedures,
add endangered species and nontarget precautions, and restricted
use labeling.
Other Rodents- All use sites.
""Call in" intrastate products and require data necessary
to support Federal registration under § 3 of FIFRA. (See
Appendix A).
"Allow products to be used at current bait concentrations
while data are being generated except in the range of the
California condor. Limitations on 1080 use in the range
of the condor are the same as for ground squirrels.
"Require label changes in baiting and post-baiting procedures,
add endangered species and nontarget precautions, and restricted
use labeling.
"Require that studies to determine the lowest efficacious
bait concentration begin as soon as possible.
"Withdraw proposed seasonal use limitation in California.
-------
-VI-
Pocket Gophers- All use sites.
""Call in" intrastate products and require data necessary
to support Federal registration under § 3 of FIFRA. (See
Appendix A).
°Allow products to be used at current bait concentrations
and application rates while data are being generated.
Norway Rat, Roof Rat and House Mice- In and around ships
and buildings.
"Allow Uses already on Federal label. Full data base for
registration under §3 of FIFRA will be required. (See
Appendix A).
-------
I. INTRODUCTION
A. Background
This. Position Document 4 (PD 4) concludes the Rebuttable
Presumption Against Registration (RPAR) process for sodium mono-
fluoroacetate, commonly called "Compound 1080." Compound
1080 (1080) is currently used as a rodenticide on rangelands,
croplands, and nonagricultural sites. Prior to 1972, 1080
was also used for predator control, principally for coyotes.
In 1972 the predacidal uses of 1080 were cancelled. This
RPAR document addresses only the rodenticidal uses of 1080.
Following a hearing in 1983, EPA determined that significant
new evidence warranted modifications of the 1972 cancellation
with respect to the use of 1080 in livestock protection collars
and single lethal dose baits. Applications for registration
of 1080 livestock protection collars are pending before the
Agency.
Before 1080 was cancelled for predator control, the
Federal registrations for the field rodent uses of 1080 were
maintained by the U.S. Fish and Wildlife Service (FWS), U.S.
Department of the Interior. Following EPA's cancellation of
1080 as a predacide, Executive Order 11643 was issued, which
prohibited the use of 1080 on Federal lands. As a result of
this order, the FWS withdrew all their 1080 Federal registrations.
In most states, ranchers used the FWS product for control of
field rodents and the FWS1 withdrawal of its registration
resulted in 1080 not being available for this use in those
states.
In addition to the FWS registration, three states, California,
Colorado and Nevada, had intrastate products containing 1080
registered for field rodent control. Intrastate products
are products that may be used only within the state in which
they were produced. They are not, however, the same as
"special local needs" registrations issued under the authority
of § 24(c) of FIFRA. Prior to 1972, EPA did not have authority
to regulate products whose manufacture and use were confined
solely to one state, i.e. intrastate products. The 1972
amendments to FIFRA expanded EPA's authority to include
these products. In 1975 the Agency reguired that registrants
of all intrastate products submit an application indicating
their intent to seek Federal registration when asked to do
so by EPA. Government agencies in California, Colorado and
Nevada had existing intrastate products and filed an application
with EPA. EPA allowed all intrastate registrants who filed
an acceptable application to continue to use their products
until "called in " for Federal registration. To date, none
of the 1080 intrastate products has been "called in" for
Federal registration.
-------
-2-
Current intrastate registrations include 46 in California,
three in Nevada, and one in Colorado. There is one § 24(c)
registration, which was issued for one county in Oregon in 1981,
solely for the control of ground squirrels. There is only one
product registered under §3 of FIFRA. This is a technical product
used for formulating 1080 end use products. The technical product
also contains directions for end use for the control of Norway
rats, roof rats, and house mice in and around ships and
buildings. The product may be used to control these rodents
only by bonded certified commercial applicators.
Estimated annual usage of compound 1080 bait was 610,600
pounds for 1977 (approximately 600 Ibs. of active ingredient).
Of the total 1080 usage, 75% is for ground squirrel control
and 7% for prairie dog control. The remaining uses are for
the control of meadow mice, deer mice, wood rat, cotton rat,
chipmunks and several other rodent species. California used
83% of the total 1080 used, Colorado 12% and Nevada and
Oregon 5%. California indicated in their PD 2/3 comments that
annual usage was approximately 100,000 Ibs. of bait, which
would indicate a considerable reduction in usage from 1977.
Compound 1080 is formulated in baits of grain, rolled oats,
or chopped greens for crop and range rodents, and in water bait
stations serviced by pest control operators to control commensal
rodents. Tull Chemical, in Oxford, Alabama, is the sole U.S.
producer of 1080. Compound 1080 is not imported.
B. Position Document 1 (PD 1)
The RPAR notice for compound 1080 was published in the
Federal Register of December 1, 1976 (41 FR 52792). The RPAR
criteria that were determined to have been met or exceeded
for the uses of compound 1080 are: 1) lack of emergency
treatment, 2) acute toxicity to mammals and birds, and 3)
significant reduction in populations of nontarget organisms
and fatalities to members of endangered species. The reasons
for the determination were explained in detail in the RPAR
notice and accompanying Position Document 1 (PD 1).
In PD 1, the lack of emergency human treatment criterion
was based on the following reasons:
- once a sufficiently large dose of compound 1080 is
absorbed into the blood stream, the victim will
inevitably die.
- symptoms of compound 1080 poisoning may not occur
until after a total fatal dose has been absorbed
into the blood stream.
- current compound 1080 labels do not specifically
prohibit use around domestic dwellings where ex-
posure to children is likely; and
-------
-3-
- monoacetin, potentially the most effective medi-
cation for compound 1080 poisoning, is not avail-
able in a pharmaceutical grade.
The acute toxicity criterion was based on a determination
that nontarget mammals and birds are likely to consume lethal
quantities of compound 1080 bait. EPA based its determination
that the significant reduction in nontarget organism populations
criterion was met primarily on acute oral toxicity data. These
data suggested that a risk to various wildlife species, through
direct ingestion of 1080 treated bait, could reasonably be
expected. A lethal guantity could also be consumed indirectly
by eating a rodent poisoned with 1080. Compound 1080 LD5Q
values are not available for many of the nontarget species which
EPA suspected could be affected. Risk to such nontarget
species was based on available toxicity values of target and
nontarget species.
The criterion for endangered species was supported by a
laboratory study that demonstrated a hazard to the endangered
San Joaquin kit fox from secondary poisoning as well as the
available toxicity data. In the secondary hazard study, a
kangaroo rat was killed with a quantity of compound 1080
that could be consumed by the rat by ingesting grain bait
treated with compound 1080. A desert kit fox, a subspecies
of kit fox closely related to the endangered San Joaquin kit
fox, was fatally poisoned by eating the kangaroo rat.
(Schitoskey, 1975)
C. Summary of Actions Proposed in PD 2/3
The Agency issued another Position Document (PD 2/3)
in June, 1983. Notice of the PD 2/3 was published in the
Federal Register on November 4, 1983 (48 FR 50935). In
PD 2/3, the Agency .responded to comments received in response
to PD 1, analyzed the risks and benefits of each 1080 use
and each regulatory option, and proposed regulatory actions.
The proposed actions were designed to reduce risks to the
point where they would be exceeded by the benefits of use.
In some instances, this necessitated proposals to deny
applications for registration; denial is appropriate rather
than cancellation because the products involved were intrastate
products and not federally registered. A summary of proposed
actions by site and pest follows.
Pangeland and pasture use.
ground squirrels- modification of the terms and conditions of
registration
prairie dogs - modification of the terms and conditions of
registration
-------
-4-
pocket gophers
chipmunks
cotton rats
deer mice
kangaroo rats
meadow mice
Cropland use.
no action
denial of
denial of
denial of
denial of
denial of
applications
applications
applications
applications
applications
ground squirrels- modification of the terms and conditions of
registration
prairie dogs - modification of the terms and conditions of
registration
- modification of the terms and conditions of
registration
- no action
- denial of applications
- denial of applications
- denial of applications
meadow mice
pocket gophers
Norway rats
cotton rats
wood rats
Use in and around buildings and ships.
Norway rats
roof rats
house mice
- no action
- no action
- no action
Use on other nonagricultural sites."
ground sauirrels-
(on ditch banks,
levees, canals,
and earthen dams)
chipmunks
cotton rats
Norway rats -
kangaroo rats
wood rats
deer mice
meadow mice -
pocket gophers
modification of the terms and conditions of
registration
denial of
denial of
denial of
denial of
denial of
denial of
denial of
no action
applications
applications
applications
applications
applications
applications
applications
The modifications of the terms and conditions of registration
included reductions in active ingredient concentration,
uniform baiting and post baiting procedures, and seasonal,
geographical and other special restrictions for the protection
of endangered species.
In PD 2/3, the Agency determined that the lack of emergency
human treatment criterion, the acute toxicity criterion, the
significant reduction in nontarget organism populations and
endangered species criterion had not been rebutted.
-------
-5-
In addition, for the criterion for endangered species,
the Agency used the evaluations of the Office of Endangered
Species (OES). OES was asked to identify any endangered
species that might be exposed to 1080 from rodent control
programs. OES noted 12 such species:
San Joaquin kit fox (Vulpes macrotis mutica)
Black-footed ferret (Mustela nigripesl
Morro Bay kangaroo rat (Dipodomys heermanni morroensis)
Salt marsh harvest mouse (Reithrodonotomys raviventris)
California condor (Gymnogyps californianus)
Aleutian Canada goose (Branta canadensis leucopareia)
Bald eagle (Haliaeetus leucocephalus)
Arctic peregrine falcon (Falco peregrinus tundrius)
American peregrine falcon (Falco pregrinus anatum)
Santa Barbara song sparrow (Melospiza melodia graminea)
San Clemente sage sparrow (Amphispiza belli clementeae)
San Clemente loggerhead shrike (Lanius ludovcianus mearnsi)
Of the 12, OES concluded that exposure to 1080 under current
label conditions was not likely to jeopardize the continued
existence of 7 species. They also concluded that new label
restrictions would minimize exposure and adequately protect
the remaining 5 endangered species. These species were: the
California condor, San Joaquin kit fox, black-footed ferret,
Morro Bay kangaroo rat and the salt marsh harvest mouse
(FWS correspondence, July 21, 1978). The Agency concurred
with the OES assessment, but expanded it to include additional
safeguards for the Aleutian Canada goose.
The Agency's concern about endangered species, especially
the San Joaquin kit fox and the black-footed ferret, was
supported by the fact that nontarget species had been killed
by feeding on animals poisoned by 1080.
D. Organization of Position Document 4
Chapter II of this document presents the public comments
on the Agency's PD 2/3 risk assessment and details the Agency's
response to these comments. Chapter III discusses benefits
in the same manner as risks are discussed in Chapter II.
Chapter IV summarizes the Agency's proposed regulatory actions
in PD 2/3, presents the public comments on these proposed
actions, and details the Agency's response to the comments.
Chapter V presents the Agency's regulatory decision.
-------
-6-
II. Comments and Agency Response on the PD 2/3 Risk Assessment
The purpose of this chapter is to review and evaluate
comments submitted to the Agency in response to PD 2/3 and to
revise, as appropriate, the determination of risks used to
reach the risk/benefit decisions on the rodenticidal uses of 1080.
Comments regarding risk associated with the use of 1080
are organized into sections:
A) General comments on risk assessment
B) Comments on lack of emergency treatment criterion
C) Comments on acute toxicity to mammals and birds criterion
D) Comments on significant adverse effects on populations
of nontarget organisms criterion
E) Comments on endangered species criterion
A. General Comments on Risk Assessment
The Scientific Advisory Panel, Wyoming Department of
Agriculture and the Montana Department of Agriculture made
general comments on the risk assessment presented in PD 2/3.
1. Scientific Advisory Panel (SAP)
The SAP commented that there was very little definitive
data to support the contention that 1080 presented a serious
hazard to non-target species, including endangered or threatened
species. The Panel found that the rationale for the action
proposed in PD 2/3, especially the reduction in bait
concentration to 0.02 percent and reduction of the rate of
application to 4 Ibs./acre was not adequately explained or
justified. The Panel also noted that the actions proposed in
PD 2/3 were not, in general, well supported.
Although the Agency agrees with the Panel that there are
data gaps with 1080, as noted several times in PD 2/3, the
Agency also believes there are enough data available to
provide a basis for concern about potential risks to nontargets,
especially endangered species. In particular the Agency
relies on the Hegdal (1979) study, and on other field and
laboratory studies demonstrating the toxicity of 1080 to
various rodent and nontarget species.
Furthermore, as clearly required under FIFRA, the
responsibility for establishing the safety and, if appropriate,
efficacy of a pesticide product rests with the registrant and
not with EPA. A complete data base for 1080 has not been
generated in large part because of the registration status
-------
-7-
of the 1080 field rodenticide use products. As noted in the
Introduction, all of the current 1080 products bearing field
rodent uses are intrastate products and have never been
subject to the data requirements for Federal registration
under § 3 of FIFRA. Existing intrastate products will be
"called in" for Federal registration and new applicants for
1080 registration will have to submit data to support
registration prior to the use of 1080. These actions will
ensure that the potential risk concerns of the Agency are
addressed and will provide a more complete data base to
determine if 1080 can be used without unreasonable adverse
effects on the environment.
2. Wyoming Deparcment of Agriculture
In its comments the Wyoming Department of Agriculture
asked a number of questions:
1) What type of investigations of unreasonable adverse
effects were made to support PD 2/3?
2) Was there a significant effect that triggered the
issuance of PD 1 in 1976, and since 1976, what data
are available to evaluate the effect of reduced 1080
bait concentrations?
3) What sociological study was produced by EPA to support
its regulatory action in PD 2/3?
4) What studies have been produced by EPA on the
conversion of prime farm land to other uses not
compatible with prime farm land objectives when 1080
rodenticides are not available?
In response to the first two questions the Agency has
established RPAR criteria that it believes are indicators of
potential risk. The criteria exceeded were identified and
discussed in both PD 1 and PD 2/3. It is important to recognize
that EPA is not limited to taking regulatory action only
on the demonstrated occurrence of adverse effects.
-------
-8-
EPA did not generate any sociological studies or
evaluations regarding the loss of prime farm land because the
Agency did not believe its proposed 1080 regulatory actions
would cause the impact suggested by Wyoming. Other pesticides
are available for field rodent control and Wyoming and other
applicants have been free to pursue registration of 1080 for
rodent control under § 3 of FIFRA.
Wyoming also commented that "EPA has repeatedly and
arbitrarily denied Wyoming's registration applications for
special local need (Intrastate) use of 1080 for field rodent
control purposes under 162.17 of EPA regulations (Ref 2)
(letters)
The two letters (telegrams from EPA to the Wyoming Department
of Agriculture) submitted by Wyoming in support of this
contention do not present a clear picture of what transpired.
One telegram stated that EPA was not accepting Wyoming's
application for two 1080 products under the intrastate program.
This same telegram also referenced recent court action vacating
an injunction against enforcement of EPA's suspension and
cancellation action against the predacide uses of 1080.
Apparently the Wyoming intrastate products contained predacide
uses as well as field rodent uses and were not acceptable
for inclusion in the intrastate program.
Wyoming and other states interested in obtaining or
expanding 1080 use may submit applications for Experimental
Use Permits to generate data necessary for registration or
otherwise provide such data. The Agency will give such requests
full and fair consideration.
3. Montana Department of Agriculture
Montana commented that PD 2/3 did not reflect their use
experience with 1080 in controlling Columbian ground squirrels
and that data collected from this use should be considered in
the Agency's 1080 decision.
The Agency did not address Montana's use of 1080 to
control Columbian ground squirrels because this use has been
permitted under the emergency exemption procedures of §18 of
FIFRA. Emergency exemption uses are not registered uses and
are issued only in unusual circumstances. In addition, the
Columbian ground squirrel is not a pest species in the four
states using 1080 to control squirrels and Montana's data
could not be extrapolated to other ground squirrel species.
Montana will be required to seek Federal registration for
this use and the data generated in their use experience will
be evaluated as part of the registration process.
-------
-9-
B. Lack of Emergency Treatment
1. Agency Position in PD 2/3
In PD 2/3, the Agency determined that the presumption
against registration for 1080, based on a lack of '--'ergency
treatment, was not rebutted because possible antid^ -es were
either unavailable commercially or were of unproven effectiveness
against a lethal dose. The Agency did note that early first aid
treatment of small ingested amounts can be effective.
The Agency reviewed the available accident history with
1080 since its use began in 1946. The record showed several
poisoning incidents involving humans (adults and children),
and domestic animals, most of which occurred in the early
years of 1080 use. The Agency also noted that the accidents
were mostly associated with use in and around buildings and
that current label instructions and restrictions on use made
it unlikely that humans and pets would now be exposed to 1080.
2. Comments on PD 2/3
One commenter, Montana, observed that there may be an
adequate emergency treatment for 1080 poisoning in humans in
the future. Specifically, they remarked that a pharmaceutical
grade of monoacetin, identified in PD 2/3 as a possible treat-
ment, may become available. This would be desirable and the
Agency urges the development of effective antidotes or emer-
gency treatment.
The same commenter also observed that there is no record
of any human fatalities or poisonings with 1080 from field
rodent use in over 35 years of use. Other State regulatory
officials have also noted the lack of human poisoning or
fatalities from the field rodent uses of 1080.
3. Agency Response
The comments submitted regarding the Agency's concern
about the lack of emergency treatment have not persuaded the
Agency to alter its position. Possible antidotes either are
not available or have not been proven effective against a
lethal dose of 1080. However, as noted in PD 2/3 most of the'
reported human poisoning incidents with 1080 have occurred
with the commensal rodent use rather than the field rodent
use and several measures have reduced risks considerably.
These measures include:
0 upgrading of the label instructions for use.
-------
-10-
0 classification as a restricted use pesticide and
its subsequent use only by certified pest control
operators or governmental agency personnel.
0 prohibition of use in occupied dwellings.
0 placement of bait in tamper proof bait boxes.
0 the generally restrictive state regulations
governing the use of 1080 as a commensal
rodenticide.
The Agency believes for the field rodenticide uses that
the risks regarding the criterion of lack of emergency treatment
are less than the risks for commensal rodent use based on:
0 the accident record, which indicates few if any
confirmed human poisonings,
0 the strict use requirements and safeguards imposed
by the states, particularly California, where over
80% of 1080 rodenticides are used, and
o
classification as a restricted use pesticide and
use only by certified applicators or persons under
their direct supervision.
Notwithstanding the lack of a proven, effective, available
emergency treatment regimen, the Agency is proposing no
additional regulatory actions based on this criterion because
of the risk reduction measures now in place.
C. Acute Toxicity to Mammals and Birds
1. Agency Position in PD 2/3
In PD 2/3 the Agency reported the acute oral toxicity
of 1080 to numerous species of mammals, amphibians,
and birds. The median lethal single oral dosages expressed
as mg of 1080 per kg animal weight (LDsg's) ranged from
0.056 mg/kg for nutria to 60 mg/kg for the opossum. The
Agency noted that even though a 1000-fold range in susceptibility
was indicated to exist between species, all species listed
were very sensitive to the toxic action of 1080 and exceeded
the RPAR criterion for acute toxicity.
Among vertebrates tested, available data indicated that
mammals are generally the most sensitive, notably nutria,
cotton rat, kit fox, and other members of the dog and cat
families.
-------
-11-
2. Comments on PD 2/3
Wyoming observed that acute toxicity data are not meaningful
unless discussed in conjunction with exposure data. The
California Department of Food and Agriculture (CDFA) noted
that "It is a scientifically accepted fact that toxicity and
risk do not go hand in hand and that there are many other
factors involved." CDFA believes that feeding behavior, food
preferences, and aversive conditioning to toxicants were not
given due consideration by EPA. Montana commented that the
risks to birds, mammals and endangered species are often
evaluated from LD50 data of biologically similar or closely
related species. They observed that this may be inaccurate
since available LD50 data shows considerable variation in
susceptability to 1080. Montana also Questioned the reliability
of present LD50 data.
Another commenter (Defenders of Wildlife) stated that
1080 is more toxic to canine and feline species than to
rodents, and that no rodenticide which poses greater risks to
nontarget species than to target pests should be used.
3. Agency Response
The Agency concurs in principal with the comments concerning
the relationship between toxicity and hazard, but notes that
acute toxicity data is an important indicator of potential
risk. The Agency notes that available data indicate a range
of sensitivity to 1080 among related species and also recognizes
that the reliability of some of the LD50 data may be questionable,
However, the Agency's conclusion from these facts, is that while
there is uncertainty in predicting risks to untested species
the available data suggest that 1080 presents risks that may
be high to certain nontarget species. Because of these
uncertainties, the Agency is requiring data that will provide
a better understanding of these risks.
Regarding the comment from Defenders of Wildlife, the
Agency agrees that it would be desirable for all rodenticides
to be more toxic to rodents than to any nontarget species.
However, as noted in the proceeding paragraphs, toxicity is
not the sole component of risk. As the Agency reviews data
for registration and the nature of nontarget risk becomes
clearer, additional risk reduction measures may be required.
-------
-12-
D. Significant Adverse Effects on Populations of
Nontarget Organisms
1. Agency's Position in PD 2/3
a. General
In PD 2/3, the risk assessment was based on the use of
1080 to control ground squirrels. For other pest species
there were insufficient data to form the basis of a complete
risk assessment and no further data have been submitted.
Therefore, the Agency assumes, as it did in PD 2/3, that the
risks posed by the use of 1080 on pests other than ground
squirrels are analagous to those risks from the ground squirrel
use.
b. Direct Risks to Nontarget Species
As noted in PD 2/3, the presence of 1080 treated bait in
the field may pose a risk to nontarget vertebrate species
that may feed on bait and such feeding may result in significant
population reductions in the nontarget species. The Agency
relied on the results of the Hegdal field study as an
indicator of potential direct risks to nontarget species.
The Hegdal study evaluated the hazards to wildlife associated
with aerial 1080 baiting for California ground squirrels.
The study was conducted in Tulare County, California, in the
eastern foothills of the San Joaquin Valley. The study monitored
a large-scale (60,000 acres) operational baiting program conducted
by the Tulare County Agricultural Commission. Ground squirrel
populations were reduced about 85% following baiting.
The results of the study showed that 1) avian species
were not severely impacted and 2) direct effects on rabbits
and nontarget rodents were more substantial than for birds.
c. Indirect Risks to Nontarget Species
Several laboratory studies and the Hegdal field study
indicate that secondary poisoning of animals (poisoning of
animals which eat a victim of direct exposure) can readily
occur with 1080.
One study was conducted by Tucker (1965-72) with domestic
ferrets (Mustela putorius). Ferrets were fed the intact
bodies of albino rats which had been stomach-tubed with 1080
two hours prior to feeding. Ferrets eating rats containing
dosages of 2 mg/kg or less exhibited strong poisoning symptoms
while dosages of 8 mg/kg and above were fatal to ferrets.
-------
-13-
In a similar experiment by Tucker with Peromyscus sp. and
ferrets, ferrets that ingested mice that had been stomach-tubed
to contain one mg 1080 per kg of ferret weight died. Based on
this study, the Agency concluded that if as little as 1 gram
of a 2 oz/100 Ib. (O.i25%) bait were ingested directly by a
ferret, it would most likely be lethal. The Agency also
concluded that if 1080-poisoned prairie dogs were fed upon by
black-footed ferrets, a similar hazard could exist.
Hedgal, Gatz and Fite (1980) fed Richardson's ground
squirrels that had died from eating known amounts of 1080
bait to coyotes with subsequent fatalities. Coyotes that died
had consumed squirrels that had eaten the equivalent of 0.45,
0.47 and 0.75 mg 1080/kg of coyote. Coyotes that survived
had consumed ground squirrels that had eaten the equivalent
of 0.14, 0.28 and 0.35 mg/kg. Therefore, these results
suggest coyotes (LD5Q .12 mg/kg) must consume about 4 times
their LD5Q in poisoned ground squirrels before they are
killed. Apparently the 1080 in the primary consumer is
detoxified; through metabolic processes or other mechanisms,
possibly a dilution effect from the feed.
The Hegdal study was also used to assess secondary risks
to nontarget species. As reported in PD 2/3 / secondary
hazards to raptors and mammalian predators were evaluated by
attaching transmitters to 24 raptors (red-tailed hawks,
turkey vultures, a golden eagle, great horned owls, a screech
owl, common ravens, a common crow), and 42 mammalian predators
(bobcats, coyotes, gray fox, badgers, striped skunks, raccoons,
and opossum) and monitoring their movements before, during,
and after treatment. While the fate of many of these animals
was not determined, five of the six radio-equipped coyotes
and three of the ten radio-equipped bobcats (one bobcat was
emaciated, possibly the result of a trap injury), were found
dead after treatment. Three dead striped skunks (not radio-
equipped) and a coyote (not radio-equipped) were also found
dead after treatment. One striped skunk contained 1080
residues. No other treatment-related mortalities were indicated
among the remaining radio-equipped birds or mammals. Also,
monitoring of 58 active raptor nests indicated no treatment-
related mortalities.
Several aspects of those observations on coyotes and bobcats
were relevant:
0 While the carcasses showed symptoms typical of 1080
poisoning and the general circumstances make it
-------
-14-
highly likely that death was caused by 1080, any
residues of 1080 were below the level of detection.
0 Most of the animals died in locations where their
discovery would have been unlikely had they not been
equipped with transmitters.
0 While the samples are small, the data suggest that
1080 presents a high risk of secondary poisoning
to coyotes and to a lesser degree the bobcat. The
data also suggest a potential for impacts on other
mammalian scavengers.
0 A year later, coyotes appeared to have returned to
pretreatment levels.
The available evidence led the Agency to the conclusion
that secondary poisoning from 1080 can be expected, although
there is considerable uncertainty in quantifying the risk.
The Agency concluded in PD 2/3 that the use of 1080 to
control ground squirrels posed a secondary risk to local popu-
lations of predators. The magnitude of the risk varies with
details of the control program and the status of the non-
target population and may extend to regional populations,
however, the Agency believed the data was insufficient to
support denial of this use.
d. Community Level Impacts
In PD 2/3, the Agency discussed in some detail another
concern regarding the potential effects of non-target predator
mortality resulting from ground squirrel control programs.
This concern was the predator-prey relationship, and the effects
that might result from severely altering the population of a
predator species. The Agency noted that there are many examples
of how the use of pesticides to control insects has created new
pest situations by killing the predators and parasites that
formerly kept potential pests in check. The Agency believed
that the possible similar relationships between ground squirrels
and their predators deserved consideration. The Agency recognized
that while there was little basis for believing that predators
alone could prevent the development of economically damaging
densities of ground squirrels, it was reasonable to suspect
that predators could serve a useful function in population
management.
There is substantial evidence that mammalian predator
populations, while they seldom may be capable of reducing a
high prey population, are capable of suppressing low populations
-------
-15-
of microtines and lagomorphs and retarding their recovery
(Todd, 1980). Support for this position is found in studies
of voles in California (Pearson, 1966, 1971), jackrabbits in
Utah (Wagner and Stoddart, 1972) and rabbits in Australia (Myers,
1980). To the extent that similar relationships may exist
between carnivores and ground squirrels, it follows that the
use of a toxicant that kills substantial numbers of carnivores
at the very time that the populations of the target pests
have been reduced to low levels will have the effect of
enabling the target population to recover at a more rapid
rate. The circumstances under which predation by carnivores
and raptors can effectively suppress low populations of
ground squirrels needs to be determined, but the Agency
suggested that the disruption of natural regulatory mechanisms
is an adverse effect beyond the immediate secondary mortality.
Although the Hegdal study suggested the likelihood of
high mortality to predators, particularly the coyote, the
Agency noted that a year after the Hegdal study, coyote
population indexes had returned to prestudy levels.
-------
-16-
e. Relation of Bait Formulations to Risks
In PD 2/3 the Agency discussed rates of application,
concentrations of 1080 in the bait, and methods of application
and how these variables may affect risks to nontarget organisms
and efficacy in reducing target species. The Agency noted
that, unfortunately, there is a paucity of information that
would directly permit a very accurate assessment of these
relationships. However, the Agency concluded that available
data suggested that risks might be reduced by lowering the
bait concentration without reducing efficacy.
The Agency believed that given the LD$Q, the size of the
animals and the concentration of 1080 in the bait, the amount
of bait required to kill an animal could be estimated. For
target species, a practical lethal "dose" would be an LDgs for
the largest individuals in the population. For nontarget
species, risk may be based on an LD$Q for an average animal
or some more stringent criterion for populations of great
concern, such as endangered species.
A major question from the standpoint of both efficacy and
secondary risk concerns the amount of bait normally consumed
by the target animals. In an area intentionally over-baited
by hand, Sibley (1966) estimated that the average squirrel
found dead had consumed about 100 grains of bait. He did not
indicate his methods or assumptions and the accuracy of this
estimate cannot be assessed. Based on Agency interpretation of
data developed by Marsh (1967), the Agency concluded that the
average squirrel may consume 200 grains of bait if this is
applied when there is one squirrel per active burrow. The
Agency noted that if 200 grains is an expected dose, secondary
risks will depend on residual toxicity of the bait, the lethal
dose for a scavenger, and how many squirrels a scavenger is
likely to eat. Relying on published LD$Q values of uncertain
accuracy and an approximation of average body weight for nontarget
species in California, the Agency calculated the number of
squirrels consuming 200 kernels of various 1080 bait concentrations
required to produce a carnivore LD50, assuming 75 percent
detoxification. These calculations indicated that at a 1080 bait
concentration of 0.1 percent a coyote LD50 would be less than
1 squirrel while the LD 50 for a kit fox would be approximately
0.3 squirrels.
The Agency concluded, therefore that both the coyote and
kit fox are clearly at great risk from applications of 0.05%
to 0.1% formulations if they feed on dead squirrels. The Agency
-------
-17-
also believed these estimates were reasonably consistent with
the observations of Hegdal in his 1979 study and the internal
consistency provided the Agency confidence that its risk
assessment was reasonable. Using the mink as a surrogate for
the black-footed ferret, it appeared to the Agency that there
was virtually no margin of safety under similar patterns of
baiting. A thorough detailed discussion of the Agency's
calculations and assumptions was presented in PD 2/3.
The Agency concluded that broadcast application of 0.08%
and 0.1% bait poses equal or greater secondary risks to carnivores
than hand application of 0.05% bait. It was also suggested that
secondary risks may be reduced by lowering the concentration
of 1080 in the bait.
2. Comments on PD 2/3
The comments in this section have been grouped by major
topic. These topics include comments on the Hegdal study,
predator prey relationships, and relationship of bait
formulation to risk.
a. The Hegdal Study (1979)
As noted several times in PD 2/3, the 1979 Hegdal study
was used to assess the potential risk to non-target species
from 1080 ground squirrel control programs. The California
Department of Food and Agriculture (CDFA) offered several
comments on the Hegdal study. They believed that the study
has many serious flaws, several of which make some of the
data and subsequent conclusions invalid or highly questionable.
CDFA cautions that the decision on ground squirrel control
has important economic and environmental consequences and
basing conclusions on a single study involving highly
questionable research practices is suspect. The major points
that CDFA raised concerning the Hegdal study were:
0 A failure by the Agency to consider that the attachment
of radio transmitters and collars on some species may
alter the behavior, particularly feeding behavior, of
the monitored species. This may result in a tagged
animal feeding more extensively on carrion or sick animals
than it would normally. CDFA cited several studies
demonstrating that the attachment of radio transmitters
alters behavior.
-------
-18-
0 CDFA believes the above point is especially pertinent to
the Hegdal study because researchers trapped predators
up to and including the night before the baiting commenced.
A number of predators had very little time to recover from
soreness or injury incurred during trapping or time to
adjust to the radio-collars. CDFA believed that the
trauma resulting from being trapped, tranquilized, radio-
tagged and released at the same time dead squirrels became
available, increased the likelihood that these animals
were prone to feeding on carrion instead of seeking live
prey.
0 CDFA notes that there was no control group trapped outside
the 1080 baiting area to determine mortality resulting from
trap injuries or the carrying of transmitters.
0 CDFA believed that the practice of luring predators to
traps with fetid scent rather than with live prey biased
the predator sample in that predators with a predisposition
to decomposing flesh were attracted to the traps and
subsequently used in the study. CDFA believes this factor
would explain the high mortality of coyotes and bobcats
(assuming that all deaths were attributable to 1080, a
conclusion that CDFA does not believe is supported by the
study).
0 CDFA believes that the analytical results failing to
identify 1080 in the tissue of many animals supports the
contention that the radios may have contributed to the
deaths of some animals.
0 CDFA questioned the usefulness of the study finding that
necropsies of dead animals generally showed symptoms similar
to those indicated for 1080 intoxication. CDFA notes that
the pathology of 1080 killed animals is often inconclusive.
Without analytical confirmation that 1080 is present in a
carcass, the necropsy alone carries little weight. CDFA
further notes that even with positive analytical data, there
is no assurance at very low levels that some other factor
may not be the cause of death.
-------
-19-
0 CDFA also questioned the Agency's statement that the Hegdal
data suggested that 50 to 100% of the local coyote
population and from zero to 50% of the local bobcat
population may be killed from 1080 ground squirrel control
programs in similar circumstances. CDFA noted that of
seven coyotes tagged, one died before treatment, one
survived, and there is no conclusive evidence that the
other five died from 1080. They also note that even if
the five post treatment deaths were attributed to 1080,
the percentage kill would be five out of seven or 71%
of the population sampled. A similar comment on the
bobcat fatalities was noted with CDFA commenting that
20% not 50% would be the upper limit since two of ten
bobcats were killed. (CDFA apparently did not include
the third dead bobcat since it had an injured leg and
was emaciated).
b. Predator Prey Relationship
CDFA did not share the Agency's concern regarding the
possible adverse effects of predator population suppression.
CDFA believes the Agency has failed to consider a major difference
between mammalian prey and predators and insect prey and
predators. CDFA points out that many insect predators are
species-specific in selecting prey, while coyotes and bobcats
are opportunistic in selecting prey- Montana also noted
that predators are opportunistic and that if one prey species
becomes less abundant, predators will switch to more easily
obtainable prey. CDFA noted that in many mammalian predator-prey
relationships the number of prey has a greater influence on
the number of predators than the reverse. They further note
that there is evidence that predation probably stimulates
reproduction in the prey species and keeps the prey population
at a higher sustained level than if no predation existed at
all (Howard 1974) .
CDFA contends that there is no evidence that any incidental
loss of predators such as coyotes or bobcats during rodent
control with 1080 has resulted in a predator-prey imbalance or
that the disruption of natural regulatory mechanisms has resulted
in adverse effects beyond secondary mortality in the immediate
area.
-------
-20-
The Montana Department of Agriculture (Montana) also
believes it is unrealistic and at odds with field observations
that predators can maintain a prey species at a low level.
They also note that population reductions of such species may
be masked or magnified by other population trends unrelated
to 1080 use and that populations of nontarget predators,
especially coyotes, recover rapidly.
Defenders of Wildlife stated that 1080 may cause local
depletions in populations of nontarget species, specifically
coyotes by secondary poisoning and that such depletions, if
frequent, may disrupt entire biotic communities. The same
commenter stated that local bobcat populations would be
threatened by use of 1080 particularly in light of trapping
of bobcats.
c. Relationship of Bait Formulation to Risk
CDFA commented that the Agency overlooked several factors
that influence the formulation of effective rodenticide baits.
Some of these factors include:
0 The LD100 in a free feeding situation.
0 How much an animal consumes in a free feeding situation.
0 How quickly symptoms occur to cause the target species
to stop feeding.
0 The kind of bait used.
0 The type of carrier used.
0 Consideration of toxicant loss during storage, slough-
off losses in the storage sack, loss because of light
rains.
CDFA believes that knowing only the LD95 and weight of the
animal in formulating bait concentration invites the development
of resistance problems. CDFA does note that an optimum bait
concentration for 1080 can be determined by a series of
laboratory feedings and subsequent field tests.
CDFA also takes issue with the Agency's assumptions
regarding the amount of bait consumed by a ground squirrel,
an important factor in estimating secondary poisoning risks.
The 200 kernels per squirrel average used by the Agency is
regarded as too high by CDFA and they point to Sibley's study
(1966), which showed that dead squirrels consumed about 100
kernels of bait and further noted that Sibley intentionally
overbaited.
CDFA also contended that there were no data to support the
Agency's assumption that broadcast application of 0.08 percent
and 0.1 percent bait posed a equal or greater secondary risk
to carnivores than hand application of 0.05 percent bait.
-------
-21-
CDFA believes that the evidence does not show that aerial
broadcasting presents a greater secondary risk to any carnivore,
d. Other Comments
Several commenters (CDFA, Montana, Wyoming Department of
Agriculture) state generally that the data on the effects of
1080 on populations of non-target organisms are incomplete or
subject to differing interpretations. They concluded from
this that compound 1080 has not been shown to cause significant
adverse effects on populations of nontarget organisms.
Several comments were made about the toxic action and
other intrinsic characteristics of 1080. For instance,
commenters disagreed with the Agency that 1080 is uniformly
distributed in the bodies of 1080-poisoned animals (CDFA,
Wyoming); that it is moderately accumulated from sublethal
doses (CDFA); that it has no significant flavor (CDFA); that
it induces regurgitation in some species of poisoned animals
(CDFA, Wyoming); that 1080 is a fast-acting poison (CDFA,
Wyoming).
Three commenters, (CDFA, Montana, Wyoming) argued that
laboratory toxicity testing has little relevance to field
exposures to 1080 and there is no credible evidence that 1080
affects nontarget populations.
3. Agency Response
a. The Hegdal Study
As noted, the Hegdal field study was a major component of
the Agency's risk asessment in PD 2/3. Notwithstanding
CDFA's many objections to the conduct of the study and the
Agency's conclusions drawn from it, EPA concludes that the
study clearly demonstrates secondary poisoning risks from
1080 field rodent control programs. CDFA acknowledges that
some coyotes and occasionally even a bobcat may be killed
from such programs.
The Agency also recognizes that trap injuries, placement
of radio transmitters and use of carrion to lure test animals
may have affected the results to some degree and that a control
group would have been useful for comparison. However, the
Hegdal study was a major cooperative effort involving EPA,
U.S. Fish and Wildlife Service and California regulatory
personnel. Efforts were made to safeguard the integrity of
the study and the Agency notes that carrion lures are commonly
used in studies to attract coyotes. Further, there are no
data to suggest that a substantial portion of the coyote
population do not eat carrion whenever it is available.
-------
-22-
The Agency recognizes that there is always room for
improvement in study design and execution, but as CDFA noted
in their comments, no study is perfect and faults can be
found with any study. The Agency does not believe that the
criticism of the study advanced by CDFA detracts from the
basic conclusions regarding secondary risk. However, the
Agency also recognizes the inappropriateness of relying
solely or too heavily on one study.
The Hegdal study confirmed the sensitivity of coyotes, and
to a lesser extent bobcats to 1080 from actual field trials.
Prior to the Hegdal study there was a sense that actual field
mortalities, although they occurred, were not significant.
The level of mortality, 5 of 6 tagged coyotes killed, does
indicate the potential for significant secondary poisoning
effects of this species even when considering the "biases"
that CDFA has suggested. The fact that coyotes are secondarily
killed by 1080 is evidence in the Agency's opinion to pursue
a conservative course in protecting other species, particularly
endangered species. As noted previously, coyote sensitivity
to 1080 is especially important as an indicator of potential
risk to the endangered San Joaguin kit fox and the black-footed
ferret.
b. Predator-Prey Relationship
The discussion of predator-prey relationships in PD 2/3
was presented to underscore the complexities that exist in
the biotic communi-ty. The Agency is not suggesting that the
coyote alone can keep ground squirrel populations in check.
The Agency also agrees that the coyote/ground sguirrel
relationship is not comparable in any specific sense to those
relationships that exist in the insect community. Nevertheless
the Agency does not believe it prudent to totally ignore
predator-prey relationships.
-------
-23-
Since these relationships are not well understood, the
Agency concludes nonselective killing of nontargets is best
avoided. As noted in the discussion of the Hegdal study, to
achieve this end the Agency is requiring that data be developed
to determine if reduced bait concentrations can reduce potential
risks.
c. Relationship of Bait Formulation to Risk
CDFA's comments on the factors that influence the formulation
of effective rodenticide baits are well taken. EPA acknowledges
CDFA and the County Agriculture Commissions expertise in this
matter. However, data have not been presented to show that
reduced bait concentration levels and application rates will
not be efficacious. The Agency concludes it is necessary to
establish minimum effective bait concentration and application
rates and determine if a reduction in potential hazard to
nontarget wildlife will follow.
d. Other Comments
The Agency agrees that many useful data on 1080 and
population effects are missing, incomplete or subject to
differing interpretations. However, existing data reasonably
show the potential for population effects in the coyote and
suggest the potential for population effects in other sensitive
nontarget species. These risk concerns may be reduced
significantly by the use of 1080 at reduced bait concentration
and application rates.
The Agency assumes that the purpose of those comments
disagreeing with the Agency's characterization of 1080's intrinsic
properties was to support the contention that 1080 does not
inherently pose any risk of secondary poisoning. The Agency's
discussion of the properties of 1080 may be open to differences
in interpretation . However, the Hegdal study and available
LD50 data indicate that certain measures are necessary to
reduce nontarget risk.
The Agency agrees that more extensive field data would be
preferrable to extrapolations from laboratory data in estimating
risks to nontarget species. In addition, the Agency does not
agree that laboratory data has little revelance in risk
assessment. In the absence or limited availability of field
data, the Agency has relied on laboratory data, existing field
data, and reasonable exposure assumptions. The Agency concludes
that this risk criterion is not rebutted merely by intrinsic
problems of extrapolation between laboratory and field testing.
-------
-24-
E. Endangered Species
The Agency's concern over the potential adverse effects
on endangered species is an important factor in shaping the
regulatory measures discussed in this document. The focus of
this concern is to ensure that the use of 1080 to control
rodents will not jeopardize the continued existence of any
endangered species. Of the endangered species identified by
the Fish and wildlife Service and EPA as species that might
be exposed to 1080 from rodent control programs, two are of
particular concern. They are the California condor and the
black-footed ferret. The reason for the high level of concern
for these two species is that their survivability is more
tenuous than the other identified endangered species.
A summary of the Agency's risk assessment in PD 2/3,
comments, and Agency response on the the California condor,
the San Joaquin kit fox, the black-footed ferret, and the
Aleutian Canada goose follows. Two other endangered species,
the Morro Bay kangaroo rat and the salt marsh harvest mouse,
were also identified in PD 2/3 as requiring modifications
in the terms and conditions of use, but no comments were received,
Many of the risk concerns discussed in Section D. of this
chapter, population effects on nontarget organisms, were
also important factors in the Agency's determination of risks
to endangered species.
-------
-25-
1. California Condor
a. Agency Position in PD 2/3
In PD 2/3 the Agency noted the complexity of assessing
the risk to condors from exposure to 1080. The key component
of 1080 exposure was the extent that condors fed on poisoned
ground squirrels and which parts of squirrel carcasses condors
consumed. The Agency believed that there was ample evidence
that condors will feed on poisoned ground squirrels, however,
the extent was not known nor was there any clear indication
as to which parts of the squirrel carcass condors will not
consume. For purposes of the Agency's risk assessment it
assumed that a condor may consume the contents of squirrel
cheek pouches and stomach as well as unmetabolized, absorbed
1080.
Utilizing the work of Sibley (1966) on the amount of 1080
treated grain contained in a ground squirrels cheek pouches
as a starting point, and making certain "margin of safety"
assumptions that took into account existing LD 50 values for
the turkey vulture and the interspecific variations in sensitivity
to 1080, the Agency estimated that a condor may have to feed
on only a few 1080 poisoned squirrels to be at risk. The
Agency also expressed concern about consumption of individual
squirrels that might contain substantially more 1080 than an
"average" squirrel.
The Agency also noted that the condor's current plight
was due primarily to its failure to reproduce. The Agency
cited certain preliminary data on the effects of 1080 on the
testes of birds, which showed that there was no evidence that
1080 affected avian testicular development. Notwithstanding
this preliminary data, however, the Agency noted its continued
concerns regarding the possible reproductive effects of 1080
on the condor.
b. Comments
The CDFA devoted much of its comment on PD 2/3 to the Agency's
analysis of risk to the condor. In addition to specific comments
on Agency interpretation of data, CDFA made some general comments
about the condor and its plight that are discussed below. This
is followed by an analysis of more specific comments on data
interpretation, risk analysis, etc. CDFA contended, among other
things that:
i. The condor was on its way to extinction long before
California was settled by Europeans.
-------
-26-
ii. There is no evidence that there was an upsurge in
mortality among condors following the first few
years of use of 1080.
iii. There is no evidence that 1080 has been responsible
for a single condor death.
iv. No hard evidence or subjective evidence was found to
support any mortality to condors from 1080 used for
field rodent control.
v. There is no convincing subjective or hard evidence to
support the theory that 1080 may be implicated in
the reproductive failure of condors.
CDFA offered several comments concerning the use of the
turkey vulture as a surrogate species for the condor. They noted
that:
i. The turkey vulture is the closest representative to
the condor and is a reasonable surrogate.
ii. There is no substantiated evidence that 1080 has
been responsible for the death of turkey vultures
(Hegdal et al. 1979) .
iii. Turkey vultures, which feed on 1080 poisoned ground
squirrels, are as plentiful today in California as
they have ever been after 35 years of 1080 use.
CDFA offered several comments on the Agency's discussion of
condor feeding habits and in particular commented that
information referenced by EPA in PD 2/3 establishes that
ground squirrels represent only a small part of the condor's
diet. Thus, CDFA argued that any ground squirrels consumed by
a condor would be diluted with other foods in its diet, which
makes the risk to the condor much less than the Agency contended.
CDFA commented that the Agency failed to account for the
tendency of turkey vultures and condors to regurgitate food, that
this characteristic is common in many avian scavengers and is
probably the mechanism by which they protect themselves from
a variety of toxic substances.
CDFA also objected to the importance the Agency placed on
the fact that ground squirrels will carry grain in their cheek
pouches and this could lead to a high concentration of 1080
treated bait that may be ingested by a condor. Based on its
considerable field experience and expertise, CDFA made several
points regarding pouching. They are:
-------
-27-
i. Pouching is rarely done by young squirrels, which
make up the bulk (80%) of the ground squirrel
population at the time of the year when most baiting
is done.
ii. Even when pouching is occurring, it may not mean that
a high percentage of squirrels have pouched 1080 bait.
iii. When some squirrels began to feel sick they will empty
their cheek pouches.
iv. Some ground squirrel species only rarely have anything
in their pouches.
v. It is not uncommon for a scavenger to remove the skin
from the skull, thus scattering any bait which may
be in the pouches.
CDFA further argued that there is very little risk to
condors based on the failure to find any condor dying from
1080 poisoning. Specifically, CDFA took issue with the
Agency's contention that 1080 is a slow acting poison and
therefore if a condor were poisoned it would likely succomb
in an area remote from the control operation and probably not
be found. CDFA also cites that time spent by condors at a
feeding site is an important factor in determining whether or
not a dead condor might be found. They also note that most
of the condor nesting, roosting and feeding areas are well
known and monitored by those working with condors making the
likelihood that a dead condor would go unnoticed small.
c. Agency Response
While the Agency generally does not disagree with CDFA's
overview comments about the condor, the report by Hagen
(1972) that a condor killed by flying into a transmission
line contained 1080 in its digestive tract is still of concern.
Sublethal effects in birds often include ataxia, a loss of
coordination of voluntary muscle movements. As pointed out
in PD 2/3 there was no way to conclude that the 1080 did not
impair the condor's faculties or any indication how the
condor came in contact with the 1080. CDFA offered no specific
comment on the Hagen report.
Notwithstanding general agreement about the lack of evidence
implicating 1080 in the death of any condor, the Agency's concern
is focused on the fact that the condor is in critical danger of
becoming extinct and condors are exposed to 1080 by feeding on
poisoned ground squirrels. The Agency's position and responsibility
is to ensure that the use of 1080 for rodent control does not
-------
-28-
result in additional pressure on the condor's ability to survive.
Therefore, all available data have been evaluated from a
conservative margin of safety perspective.
The Agency agrees with CDFA's assessment regarding the
turkey vulture, but believes its use as a surrogate species
for the endangered condor has certain limitations. The
Agency will not accept LD50 values for the turkey or black
vulture as an approximate LD50 for the condor.
The Agency noted in PD 2/3 that while turkey vultures and
condors are both carrion feeders and belong to the same family,
they are of different genera and as such their susceptability to
1080 may or may not be similar. There is no way to predict
the susceptability of a particular species to 1080. Therefore,
the Agency believes that in order to provide the greatest
degree of protection for this species, the toxicity of the
most sensitive avian species should be considered in determining
a regulatory position.
Regarding CDFA's comments on condor feeding habits, the
Agency recognizes that the condor does not limit its feeding
to ground squirrels, but there is evidence that condors
commonly feed on them. In addition, in the absence of specific
information, EPA assumes that condors will feed on stomachs
and cheek pouches of ground squirrels, though not necessarily
selectively. The Agency is particularly concerned about
those situations where ground squirrel carcasses are plentiful
and other food sources are scarce. This scenario could lead to
feeding episodes where poisoned ground squirrels constitute all
or most of a condor's diet.
The Agency agrees with CDFA that the ability to regurgitate
may be an important factor in self protection for the condor,
but, as CDFA also noted, this factor alone may or may not be
enough of a safety factor to save an exposed individual; coyotes
may also regurgitate poisons, yet they are susceptable to 1080.
The points raised by CDFA on pouching habits of ground
squirrels do not significantly diminish the Agency's concern,
described in PD 2/3, concerning the potential for a condor to
feed on squirrels that might contain large amounts of treated
bait in their pouches.
The Agency does not disagree with the observations
and comments of CDFA regarding the time spent by condors at
feeding sites, or the high level of monitoring and tracking
devoted to individual condors. The Agency is persuaded
that the high level of observation and interest in the condor
-------
-29-
by various groups may make the likelihood of a dead condor
going undetected less than the Agency originally speculated.
California, however, did not support its contention with any
estimates of what fraction of condors that have died in
recent years were, in fact, found.
In summary, CDFA has raised many valid points concerning
condor feeding habits, characteristics, exposure, and lack of
evidence attributing any condor death to 1080. However, the
Agency believes that the comments presented by CDFA do not
eliminate the risk potential to the condor identified by the
Agency in PD 2/3. The Agency must ensure that the use of 1080 as
a field rodenticide does not impose an increased burden on the
condor's ability to survive. The Agency views the loss of a single
condor to 1080 as a threat to the survival of the species. This
position necessitates a conservative evaluation of data such
as LDSO's, that considers not only biologically closely related
species but the most sensitive avian species. It also necessitates
consideration of feeding and exposure situations that may not be
likely to occur, but are still possible under certain conditions
and must be considered.
The Agency has proposed continued use of 1080 in the range
of the condor, but at reduced bait concentration levels. However,
a recent biological opinion from the Office of Endangered Species
(OES) that use of the 1080 livestock protection collar would
jeopardize the continued existence of the condor necessitates
further discussion with OES on the use of 1080 baits in the
condor's range. The Agency considers the condor's range to be that
as defined by OES in their biological opinion on the livestock
collar. It includes the California counties of Fresno, Kern,
King, Los Angeles, Monterey, San Benito, San Luis Obispo,
Santa Barbara, Tulare, and Ventura.
It is possible that 1080 baits will be prohibited in
the range of the California condor until safety data are
submitted and evaluated.
The Agency believes the regulatory decision contained in
Chapter V of this document reflects consideration of the
field and regulatory expertise of CDFA, the California Department
of Fish and Game, and the County Agricultural Commissioners'
Offices.
2. San Joaguin Kit Fox
a. Agency Position in PD 2/3
In PD 2/3 the Agency noted that while there were no
confirmed reports of kit foxes being killed from 1080 used in
ground sguirrel control programs, the potential for adverse
effects was very high considering the approximate 1080 LD 50
-------
-30-
value for the kit fox of 0.22 mg/kg. The Agency noted that
this degree of sensitivity approached that of the coyote,
which is approximately 0.1 mg/kg, and Hegdal (1979) clearly
demonstrated that 1080 ground squirrel control programs
resulted in substantial coyote mortality.
b. Comments
CDFA offered several comments on the Agency's assessment
of risk to the San Joaguin kit fox.
In PD 2/3 the Agency cited reports by researchers that
indicated that:
i. San Joaquin kit foxes will eat ground squirrel carrion
(Laughrin 1970; Morrell 1971, 1972).
ii. 1080 rodent control programs may kill some kit foxes,
but do not threaten the population (Swick 1973,
Morrell 1975).
iii. Field evaluations have indicated that ground squirrels
frequently consume several lethal doses of 1080 before
death (Swick 1973).
On point i, CDFA commented that kit fox prey almost
entirely on live kangaroo rats, pocket mice, San Joaqu.in
antelope ground squirrel, and probably as carrion, two species
of rabbit and minimally the California ground squirrel.
On point ii, CDFA noted that the quote attributed to
Swick, concerninq statements from the San Luis Obispo County
Agriculture Department regarding kit fox losses in rodent
control programs, could not be verified.
On point iii, CDFA questioned EPA's characterization of
the amounts of 1080 found in ground squirrels as "several
lethal doses."
CDFA noted that none of the authors discussed above came
up with evidence that a single kit fox was ever killed by
1080 and that no other kit fox fatalities have been attributed
to 1080
CDFA objected to the Agency citing the 1975 Schitosky
study involving one desert kit fox as evidence for the RPAR
criteria for endangered species being met. CDFA noted that
the test, which involved gavaging a kangaroo rat with a
massive dose of 1080 and then feeding it to the kit fox, in
no way ressembles what occurs in the field.
-------
-31-
CDFA noted that according to Morrell (1975) the estimated
kit fox population was a minimum of 5,066 and a maximum of
14,800 adults with a mean figure of 10,000. In another
section of their response they state that these figures
represent the current kit fox population. CDFA cites Dana
(1976) that there were believed to be only 1,000-3,000 kit
foxes when they were placed on the endangered list in 1967.
They also point out that the kit fox recovery team recommended
recently that the kit fox status be downgraded from endangered
to threatened.
CDFA speculated that alternative rodenticides may be
less efficacious and present more of a hazard to nontarget
animals, particularly birds, including condors. They cite
Schitosky (1975), who contends that zinc phosphide is the
safest to kit foxes among three rodenticides tested (including
strychnine and 1080) but its use results in a greater percentage
of squirrels dying above ground rather than in their burrows.
This presumably would increase the risk to condors and other
birds.
CDFA also speculated that resident foxes in ground
squirrel control areas may have developed an aversion to 1080
killed ground squirrels and/or developed a tolerance through
ingestion and emesis of sublethal doses.
Defenders of Wildlife commented that the use of 1080 is
inappropriate in the range of the kit fox, even when one-mile
buffer zones around active dens are excluded from treatment.
c. Agency Response
Regarding CDFA's comment on the minimal role ground
squirrels play in the kit fox diet, the Agency agrees that
the California ground sauirrel may indeed make up a small or
even insignificant portion of the diet. However, as with
the condor, the Agency is concerned about unusual circum-
stances, when "normal" or preferred food is not available or
is scarce. In these instances the California ground squirrel
carcasses could become a large part of the kit fox's diet on
an interim basis.
The Agency has no reason to doubt the accuracy of the
Swick comment and notes that the observation may have been
made by an employee no longer with the county. In any event,
it underscores the Agency's concern that the potential for
poisoning of kit foxes does exist.
Regarding CDFA's questions about ground squirrel
consumption of "several lethal doses" of 1080, the Agency
notes that the size of the lethal dose in relation to the
amount consumed is important for some purposes. CDFA does
not dispute that data show that dead ground squirrels have
-------
-32-
been recovered and analyzed and shown to contain significant
quantities of unmetabolized 1080. Moreover, it is not in dispute
that kit foxes could ingest such 1080 by eating the ground
sguirrels. Although CDFA is correct that the Agency does not
report the LD100 for the California ground squirrel it does
report the LD50, a value which is appropriate for making
calculations of potential hazard. The focus of Agency concern
in this regard is that the poisoning action of 1080 is such
that it may allow, in some cases, time for squirrels to
consume more bait than needed to cause death. This fact
heightens the concern about the amount of 1080 a kit fox
might be exposed to from feeding on a 1080 killed ground
squirrel. Further, as expressed in PD 2/3, the Agency is
concerned with the loss of a single kit fox, although the
level of concern is less than for the condor.
The Agency acknowledges that there is no hard evidence
that any kit fox has been killed with 1080, however there are
convincing facts such as the feeding habits of the kit fox,
LD 50 values of surrogate species and the California ground
squirrel, that indicate the potential for kit fox loss from
1080 baiting programs. Additionally, Spencer (1945) reported
kit fox mortality associated with ground squirrel baitinq with
1080, although no confirmatory chemical analysis was conducted.
The Agency agrees that the Schitosky study was not
conducted in such a manner to allow any definitive conclusions
about the fate of a kit fox feeding on 1080 killed rodents.
The Agency has the following response concerning CDFA's
comments on the estimated kit fox population and its current
endangered status. The San Joaquin Kit Fox Recovery Plan
published by the U.S. Fish and Wildlife Service, January 31,
1983, does note that the 1969 population was estimated at
between 1,000 and 3,000 animals. The Recovery Plan also
notes that the Morrell study of 1975 was a more thorough
quantitative study than earlier ones, but that the population
estimate of 14,832 was of doubtful value since no correction
was made for the loss of suitable habitat. The Recovery Plan
presents adjusted figures based on the Morrell study of a
maximum population of 6,961. The Plan also notes that the
Morrell data must be interpreted with caution because only 1
percent of the total range was surveyed. The Plan estimates
the San Joaquin kit fox population prior to 1930 may have
been between 8,667 and 12,134 based on densities of 1 to 1.4
foxes per square mile and a range of 8,667 square miles. It
is further noted that the 1975 adjusted population estimate
of 6,961 represents a possible population decline of 20-43%
in the last half century.
-------
-33-
The Recovery Plan does not present a definitive picture of
the population at the time the kit fox was listed nor the
current population although it appears that it has stabilized.
The Plan does not support the implication of CDFA's comment
t'hat the kit fox population has increased since endangered
species listing. Further, no mention of downgrading the status
of the kit fox was noted by the Agency in its review of the
Recovery Plan. The San Joaguin kit fox is still listed as an
edangered species by OES.
Regarding CDFA's comments about the use of zinc phosphide
resulting in more ground sguirrel deaths above ground, the
Agency notes that the mode of poisoning action with zinc
phosphide does not generally result in secondary poisoning
concerns. When ingested, zinc phosphide generates phosphene
gas, which normally renders the carcasses relatively harmless as
the gas is guickly dissipated. The Agency is also of course
concerned with the question raised by CDFA regarding the use
of less efficacious rodenticides that result in more baiting.
This includes concern over lowering the concentration of a
pesticide to a level that is ineffective.
The Agency acknowledges that CDFA's hypothesis about
aversive conditioning and 1080 tolerance build up in the kit
fox is possible, but there are no specific data on this
phenomena.
The Agency believes the 1 mile buffer zone from any kit fox
den that is now reguired by California, in addition to the other
stringent regulatory controls imposed by California on 1080
use, as well as the modifications in the terms and conditions
of use reguired by this document will provide added protection
for the kit fox until more definitive data become available.
3. Black-footed ferret
a. Agency Position in PD 2/3
The Agency concluded that there was no "margin of safety"
for the black-footed ferret from the use of 1080 to control
prairie dogs. This conclusion was based on the LD 50 values
of the prairie dog, the amount of 1080 treated bait that a
prairie dog might consume, assumptions about the amount of
1080 that might not be metabolized following ingestion of
bait by the prairie dog, and LD 50 values of species related
to the black-footed ferret.
Specifically, the Agency suggested that if a one kg
prairie dog consumed as little as 4 grams of a .11% bait (the
concentration of the only bait available for prairie dog
control), 4.4 mg of 1080 would be ingested. The Agency went
on to note that only 0.3 mg of absorbed 1080 may be sufficient
to kill a prairie dog based on existing LD 50 values and
therefore much of the 1080 consumed by a prairie dog would
not be detoxified.
-------
-34-
b. Comments
CDFA commented that the Agency's attempt to show no
margin of safety for the ferret was not very convincing.
They contended that if 75 percent of the 1080 consumed
was detoxified, a rate of detoxification suggested by the
Agency in PD 2/3, only 1.1 mg of 1080 would remain in the
carcass of a prairie dog, using the example cited by the Agency
in part (a) of this section. They further speculated that if
the black-footed ferret is egually susceptible to 1080 as the
domestic ferret (LD50 of 1.41 mg/kg) then a LD 50 would still
not be reached if a one kg black-footed ferret consumed the
entire prairie dog containing 1.1 mg. CDFA also commented
that a one kg ferret would not consume more than 10 percent
of its weight or in this case 100 grams of food a day. Their
conclusion was that there is apparently little if any risk to
the black-footed ferret.
c. Agency Response
The Agency is not persuaded by CDFA's analysis that
little or no risk exists. The Agency cannot unduly rely on
the 75% detoxification factor, which has been calculated from
a single small test. As noted in several discussions in this
document, the wide variability in toxicological response of
related species to 1080 is of concern to the Agency and it
cannot base its calculations solely on one surrogate species.
In this regard, the Agency also notes that another species of
the same genus, the mink (M_. vison) , has LD 50 values reported
between 0.49 - 1 mg/kg, which is less than the reported LD 50
value for the domestic ferret (1.41 mg/kg-more recent data show
an LD5Q of 1.23 mg/kg). Therefore, the Agency believes that
it may be reasonably extrapolated from available data that the
use of 1080 may pose risk to black-footed ferrets.
4. Aleutian Canada goose
a. Agency Position in PD 2/3
In PD 2/3 the Agency indicated that USDI had concluded
that the use of 1080 under current label conditions was not
likely to jeopardize the continued existence of the Aleutian
Canada goose. EPA believed that a greater degree of concern
was warranted because 1080 is used to control ground sguirrels
on the dikes of rice fields, an area where geese may feed
when the fields are fallow. The Agency also noted that
while the geese may feed primarily on herbaceous matter,
there was concern that they may also feed on treated grain
bait when available
-------
-35-
b. Comments
CDFA recommended that USDI's determination be relied
upon and expressed concern that alternative rodenticides may
result in geese fatalities.
c. Agency response
The Agency has considered the points raised by CDFA and
concludes that the risks to the Aleutian Canada goose can be
significantly reduced when 1080 is used in accordance with
the CDFA program recently approved for strychnine and other
toxicants.
-------
-36-
III. Comments and Agency Response on the Position
Document 2/3 Benefits Analysis
A. Agency Position in PD 2/3
In the PD 2/3, the Agency evaluated the economic
impacts of denying registration of 1080 for federal and
intrastate registered rodenticidal uses. The general approach
of that economic analysis was to evaluate the economic impacts
assuming 1080 was cancelled and users shift to alternative
materials or systems. The analysis identified or estimated
the quantity of registered 1080 bait used, listed registered
alternatives, and determined changes in treatment costs
associated with the use of the most likely alternative rodent
control methods.
Compound 1080 is primarily used to control ground sguirrels,
prairie dogs, and field rodents. It is perceived that control
of ground squirrels and prairie dogs is necessary because these
rodents compete with cattle for grass and leave holes in the
ground which domisticated animals can step in or which are
sources of erosion. The field rodents eat grain.
The Agency's conclusions regarding economic effects were
based on information from a number of sources, principally the
Preliminary Benefit Analysis. In addition, rebuttal comments
from registrants, users, USDA and other parties, federal and
state vertebrate pest control specialists and available
published data were considered in the Agency's conclusions.
The Preliminary Benefit Analysis focused on agricultural
uses (pastures and croplands) to control rodent pests in forage,
and nonaqricultural uses primarily for the control of commensal
rodents. Within these two categories, the Agency identified
major and minor uses to the extent that data permitted, estimated
quantities of 1080 used and listed available alternatives
including an evaluation of their efficacy and availability.
The Agency faced considerable data limitations in performing
the economic analysis of denying 1080 uses. In most cases,
data necessary for accurately determining economic effects
were lacking. For example, almost no field test data were found
regarding the comparative efficacy of 1080 and alternatives
under the complex set of climatic, topographic and soil
conditions on a site-by-site, species-by-species basis.
Likewise, few statistically valid field surveys or data were
found on the extent of pest infestation and the damage caused
by various rodent species on individual sites. As a result
of these data limitations, the Agency relied on the judgement
of knowledgeable Federal, state and local vertebrate pest con-
trol specialists in the evaluation of economic consequences
of denying uses of 1080. Thus, this analysis often resulted
in qualitative discussions of impacts rather than quantitative
estimates.
-------
-37-
In summary the Agency concluded that 1080 was the most
efficacious and cost effective single dose rodenticide available
for control of field and commensal rodents. Treatment costs with
strychnine and zinc phosphide, the principal alternatives to
1080, increased costs an estimated 60 to 80% for the control
of ground squirrels and prairie dogs. These increased costs
resulted from both higher bait costs and the need in many
instances to prebait. The Agency concluded that chemical
controls (including strychnine and zinc phosphide) and non-
chemical controls can be effective when properly used, however,
they were usually more labor intensive, more costly and
generally thought to be less effective than 1080.
The Agency concluded that local impacts on agricultural
productivity or production costs could occur in Colorado,
Oregon, and California, where there are active 1080 use
programs, but that the economic impact of denying all uses of
1080 would not seriously affect U.S. production or prices of
major cotnmodites or services.
B. Comments on PD 2/3
No significant, supported, new benefit information was
submitted in response to the PD 2/3. USDA, California, Montana,
Nevada, South Dakota, Wyoming, several farm and ranch
organizations, as well as individual farmers commented on
various aspects of the benefits discussion presented in PD 2/3.
USDA expressed concern that the Agency's action would lead
to the unavailability of efficacious, cost effective, and
safe tools for the control of field rodents. They recommended
that the Agency delay any final decision on 1080 until the
results of several ongoing cooperative studies were completed.
USDA maintained that these studies would provide critical
information on environmental impacts and costs.
CDFA commented that if 1080 were not available, annual
losses of $2 billion would be incurred eventually. California
based farm and ranch organizations also cited the $2 billion
figure, but neither CDFA or any organization submitted
documentation to support this estimate.
Montana expressed concern that the use and benefits
information they had gathered during the use of 1080 on an
emergency basis under §18 of FIFRA for control of Columbian
ground sguirrels was not included in PD 2/3 and should be
included in the Agency's decision.
Nevada commented that 1080 may be preferable in some
situations to available rodenticides and recommended that
certain uses proposed by the Agency for denial be kept.
-------
-38-
The South Dakota Department of Agriculture (South Dakota)
made two comments that related directly to the benefit analysis.
First they commented that the Agency had not sufficiently
addressed the "testimony of field rodent control professionals
that the zinc phosphide yields inconsistent results for
prairie dog control." The second comment concerned the adverse
economic impact suffered by South Dakota agricultural
producers because of the unavailability of 1080 for prairie
dog control in their state. South Dakota estimated that the
cost of zinc phosphide was $784,000 greater that the cost of
treating the same acreage with 1080.
Wyoming claimed that state agricultural production would
be severely impacted if 1080 were not made available.
Several other commenters, mostly ranchers and farmers,
stated that there was a need for field rodent control, and
that 1080 was an efficacious and viable chemical.
C. Agency Response
The future availability of 1080 will now depend upon the
willingness of registrants and other interested parties to
develop the data necessary to support Federal registration.
While data are being generated the Agency has decided to
allow the continued use of 1080 for all rodent species and
sites contained on current labels with certain modifications,
with the possible exception in the range of the California
condor. The Agency notes that USDA's comments about ongoing
studies were made in November, 1983 but, to date, the results
of these studies have not been submitted to the Agency.
In the absence of supporting data, the Agency cannot
utilize the $2 billion agricultural impact estimate reported
by CDFA and other California agricultural organizations.
The Agency does note that it would appear that a chemical of
that much economic importance would be readily supported with
data necessary for Federal registration.
The Agency did not consider the benefits and use information
from the Montana Columbian ground sguirrel programs in developing
PD 2/3 because the Agency only addressed Federally registered
or intrastate products. As noted earlier the Columbian ground
squirrel use was permitted under the emergency exemption
provisions of § 18 of FIFRA. This provision permits use only
in unusual or emergency circumstances. However, the benefits
information submitted by Montana has been reviewed in the
preparation of this position document. The Agency concluded
that the Columbian ground squirrel information cannot be
extrapolated to other field rodent uses and does not alter
the Agency's benefit assessment in PD 2/3. The information
gathered by Montana will be useful in supporting a Federal
registration for Columbian ground squirrel control.
-------
-39-
Nevada did not provide any quantitative benefits information
necessary for the Agency to assess the economic impact of
their suggestion.
In response to South Dakota's comments the Agency
notes that it reviewed all pertinent data, gathered information
from zinc phosphide prairie dog control programs conducted in
South Dakota and elsewhere, and collected information from local
state, and Federal vertebrate pest control specialists in
assessing the efficacy of zinc phosphide. The Agency believes
its efficacy assessment of z,inc phosphide presented in PD 2/3
is correct.
On South Dakota's second comment, the Agency notes that
it did not specifically address the impact on benefits in
states that did not have intrastate registrations, which
would include South Dakota. The Agency did note in PD 2/3
that areas that did not have 1080 available were at a comparative
disadvantage where alternatives were more costly or not as
effective. The Agency has some questions about the dollar
amount presented by South Dakota in that it was assumed that
the entire acreage of prairie dog inhabited lands was treated.
In addition the $784,000 is small compared to the total
estimated value of cattle, calf, and sheep production in the
State ($1.4 billion in 1982). As detailed in Section IV, the
Agency intends to allow the continued use of 1080 for prairie
dog control under certain strict conditions. South Dakota,
however, will have to obtain a §3 Federal registration prior
to the use of 1080 in South Dakota for prairie dogs or other
rodents, as will all states that do not have a current intra-
state registration for 1080.
The Agency could not evaluate the statements of Wyoming
and the several ranchers and farmers who commented on the
agricultural impact of not being able to use 1080, because no
supporting data were submitted.
The Agency also notes with interest the development of
certain benefits information associated with the use of 1080
to control prairie dogs. The high potential risks to the
endangered black-footed ferret from the use of chemical controls
has resulted in questions being raised about the ultimate
benefits of chemical control programs. Certain new information,
which is discussed briefly in section D of this chapter,
indicates that improved range management and cultural practices
may be viable alternatives to toxicants under certain conditions.
Although this information relates directly to prairie dog
control, the principals may be applicable to the control of
ground squirrels and other rodents as well.
-------
-40-
D. New Information
Recent studies with the management of lands containing
large prairie dog populations suggest that cultural control
may be a viable alternative to the use of 1080 or other
toxicants in some circumstances. Prairie dog towns are
generally associated with range areas which are heavily
grazed by cattle. It appears that prairie dogs rely on the
cattle to keep the grass in and around the prairie dog town
short. If the cattle are removed, the prairie dogs are
not always able to keep the grass short. Tall grass provides
cover for animals- which prey on prairie dogs, such as coyotes
As the grass cover increases, the efficiency of the predators
increases to the point that the prairie dog population begins
to decline. Snell (1984) describes a successful program in
which a 110 acre prairie dog town was reduced to less than 10
acres by reducing grazing of the prairie dog town by cattle.
This method may reduce the need for broad scale poisoning of
prairie dogs, especially where it is practical to control the
level of cattle grazing or where overgrazing has contributed
to the prairie dog infestation.
-------
-41-
IV. Comments and Agency Response on Proposed Regulatory
Actions in Position Document 2/3
This chapter summarizes the Agency's proposed actions in
PD 2/3 for each use, presents the comments on these proposals,
and the Agency's response to the comments.
A. Rangeland, Cropland, and Non-Agricultural Sites: Ground
Squirrels
1. Proposed Actions in PD 2/3
In PD 2/3, the Agency proposed several modifications
in the terms and conditions of use. Most of these modifications
were to be incorporated on product labels.
a. Standardize bait concentrations at a maximum of 0.02
percent active ingredient and application rate of one level
teaspoon per burrow for hand application, a maximum of 0.02%
active ingredient and application rate of four pounds per acre
for broadcast application of grain bait, and a maximum of 0.02%
active ingredient and a application rate of five to ten pounds
per acre for broadcast application of green bait.
b. Standardize hand baiting and post-baiting procedures
as follows:
i. Baiting should not be done unless tests indicate
satisfactory bait acceptance will occur in areas
to be treated.
ii. Keep pets and domestic animals away from treated
areas.
iii. Clean up all accidentally spilled bait immediately.
iv. Do not place bait in piles.
v. Pick up and burn or bury deeply all visible carcass
of animals killed by 1080.
vi. Do not use within 1/4 mile of a dwelling without
first notifying the occupants.
vii. All bait must be dyed yellow in accordance with the
California Vertebrate Pest Control Handbook.
c. Standardize aerial baiting and post-baiting procedure
as follows:
i. Use in accordance with the Guidelines for Applying
Rodent Baits by Aircraft for Control of Ground
Sguirrels in the California Vertebrate Pest Control
Handbook.
ii. Pick up and burn or bury deeply all visible carcasses
of animals killed by 1080.
iii. All bait must be dyed yellow in accordance with the
California Vertebrate Pest Control Handbook.
-------
-42-
d. To protect the California condor, the following
requirement was proposed:
Do not use in the Sespe-Sierra range of the California
condor or in that portion of the coastal range south
of Monterey County. Do not use in the coastal range
north of San Luis Obispo County during the months of
August through December.
e. To protect the San Joaquin kit fox, the following
requirements were proposed:
Do not use in that portion of the range of the San
Joaquin kit fox that has been closed to night hunting
by the California Fish and Game Commission. In the
remainder of the range, consult the California Fish
and Game Commission prior to use.
f. To protect the Morro Bay kangaroo rat, the following
requirement was proposed:
Do not use in the range of the Morra Bay kangaroo
rat, as defined by the United States Fish and Wild-
life Service.
g. To protect the salt marsh harvest mouse, the
following requirement was proposed:
Do not use in the range of the salt marsh harvest
mouse, as defined by the United States Fish and
Wildlife Service.
h. To protect the Aleutian Canada goose the following
requirements were proposed:
Do not use from September 1 to April 30 for the
control of ground squirrels on croplands within the
areas closed by the California Department of Fish
and Game to the hunting of Canada Geese. Additionally,
any baiting for ground squirrels in the wetlands
areas of Contra Costa, San Joaquin, Solana and Yolo
Counties from September 1 to April 30 is permitted
only if bait stations are use in accordance with the
California Vertebrate Control Handbook.
2. Comments on PD 2/3
Many commenters responded to these proposals. On the
proposal to reduce and standardize 1080 bait concentrations
at .02%, commenters (Montana, CDFA, Colorado Department of
Agriculture, New Mexico Department of Agriculture and USDA)
expressed concern that the proposed bait concentrations
-------
-43-
had not been shown to be or would not be efficacious. USDA
commented that any reduction of active ingredient not based
on efficacy data may significantly reduce efficacy, thus
rendering the bait ineffective for achieving control, while
increasing costs because of the subsequent need for multiple
applications. This could potentially increase the hazard to
nontarget species. USDA also cited work by Salmon and
Lickliter (1982) that showed that standard bait concentrations
could not be applied across species. USDA pointed out
that if the target animal does not ingest enough bait to
cause death, not only is control ineffective but future
control may also become more difficult because of bait shyness.
USDA recommended that any reduction and standardization
of bait concentration and application rate be based on scientific
studies and that the proposal in PD 2/3 be delayed until data
are developed to determine the field efficacy of various
dosage rates. USDA noted that such studies could be done
rapidly at reasonable cost.
The California Farm Bureau Federation characterized the
proposed bait reduction requirement as unacceptable and noted
that California officials had already reduced the bait concentration
to those levels that retained the necessary effectiveness. They
also cited the State's extensive experience with 1080 as a reason
for retaining the current range of bait concentrations. The
California Cattleman's Association echoed the Farm Bureau's
sentiment regarding the State's experience with 1080. They also
expressed concern that lowering bait concentrations to levels
that would make baits ineffective would force excessive repeated
use and would result in less effective control and more toxic
bait in the environment.
On proposals to standardize hand and aerial baiting and
post-baiting procedures, commenters (USDA, Montana, and
California) pointed out that some of these measures (keeping
pets away, and burning or burying visible carcasses) were not
always possible. Specifically, USDA noted that it was not
practical under range conditions to pick up and burn carcasses
because of the risk of plague or parasite infestations to
those humans responsible for disposal. They also noted that
soil conditions may prohibit burying and fire hazards prohibit
the burning of carcasses. Montana noted that it was impractical
and unrealistic to pick up and burn or bury carcasses on large
or inaccessable baited areas. Montana recommended that carcasses
be buried singularly, in order to lessen the risk from a
concentration of carcasses, that could be dug up by scavengers.
Commenters also noted that dye colors for baits should be
allowed to vary depending on local conditions and the California
Vertebrate Pest Control Handbook has legal force' only in
California.
-------
-44-
On the proposal to impose geographic restrictions to
protect the California condor, commenters (CDFA and USDA) said
that the geographic range restrictions were not justified in
light of California's efforts to protect this endangered
species. USDA agreed that since the risk to condors was small
and theoretical in nature, the Agency should permit the use of
1080 for ground squirrel control in the range of the condor
consistent with current California policy. California's policy
is that a label must specify when a pesticide is to be used
in defined endangered species areas. They further noted that
the Agency's final position on strychnine used for ground
squirrel control in the range of the condor was consistent
with California's policy. CDFA offered extensive comments on
the Agency's risk analysis to the condor which were discussed
in Section II of this document.
On the proposal concerning geographic restrictions for the
protection of the San Joaauin kit fox, USDA stated that EPA
had chosen to ignore the data of Swick (1973) and Morrell
(1975) and had instead relied on assumptions based on LD 50's
and studies on another species, the coyote. They further
noted that the Agency admits that they were unaware of reports
verifying that kit foxes had been killed by 1080. USDA
recommended that EPA concur with California policy and regulation
for protection of the San Joaquin kit fox. CDFA also presented
numerous comments on the Agency's risk analysis to the kit
fox. These comments were addressed in Section II of this
document.
On the proposal to establish geographic restrictions to
protect the Morro Bay kangaroo rat and the salt marsh harvest
mouse respectively, USDA objected to the Agency proposal to
prohibit use in the range of these species and notes that PD 2/3
does not provide scientific justification for these actions nor a
discussion of the potential impact such action would have.
On the proposal to place seasonal restrictions on 1080
baiting to protect the Aleutian Canada goose, CDFA commented
on the lack of demonstrated risk to the goose and the fact
that EPA ignored the advice of USDI. They also noted that
they would concur with restrictions for the goose if they
were in accord with the recommendations of California
governmental agencies. These comments were discussed in
Section II of this document.
3. Agency Response
Concerning proposal (a), reduction and standardization of
bait concentration and application rate, the Agency agrees
that any reduction to levels that are ineffective could have
adverse consequences. The Agency also agrees that the soundest
way to determine the optimum bait concentration and application
rate is with valid test data. The Agency also agrees with
USDA's comment that such tests could be performed quickly
and at reasonable cost. However, no data have been generated
-------
-45-
or even initiated in the several years since the beginning of
the Special Review or in the last year since the issuance of
PD 2/3. The Agency has modified its position to permit use
at current bait concentrations and application rates until
data for registration are generated except in the range of
the condor. However, as discussed more fully in Chapter V, a
recent biological opinion submitted to the Agency from USDI's
Office of Endangered Species (OES) on the use of the 1080
livestock protection collar necessitates further evaluation
of the use of 1080 baits in the range of the condor. OES
stated that the use of the collar would jeopardize the
continued existence of the condor. Use in 1985 baiting
programs will be allowed under current label requirements.
If permitted, beginning in 1986, the use of 1080 in the
range of the condor will be at a maximum bait concentration
of .02% until field efficacy data are generated. If the
.02% level is not effective, the level will be changed when
data establishing the lowest effective bait concentration
are submitted as well as data showing that this level will
not impact on the condor.
Concerning the standardization of hand baiting and
post-baiting procedures, the Agency agrees that the requirement
to keep pets and domestic animals away is overly restrictive,
and would, as a practical matter, prohibit use since there
are very few places where free-roaming pets and livestock do
not occasionally occur. The requirement has been revised.
The Agency also agrees that the proposed requirement to
bury or burn all visible carcasses cannot be reasonably
implemented in some situations and has revised this requirement
The Agency disagrees that its proposal to require dyeing
of rodenticide baits as specified in the California Vertebrate
Pest Control Handbook is too restrictive and warrants
modification. One of California's purposes in dyeing 1080
grain baits is to protect seed eating birds. Field trials
with Auramine O, a yellow dye, showed limited impacts on
avain species, which may have been due to repellency caused
by the dye. Therefore, the Agency is requiring that all
1080 grain baits used for field rodent control be dyed yellow.
-------
-46-
The Agency has modified its position concerning the
imposition of certain geographic and seasonal use restrictions
to protect the endangered California condor, San Joaquin kit
fox, Morro Bay kangaroo rat, salt marsh harvest mouse and the
Aleutian Canada goose.
The Agency proposed these restrictions to provide adequate
protection for the identified endangered species. However,
it is the Agency's belief that in this instance the development
and administration of programs to protect most endangered
species will be most effectively carried out at the state and
local level. All the endangered species identified in the
preceeding paragraph are potentially exposed to 1080 rodent
baits only in California. Based on comments and documents
submitted by various State Agencies in California responsible
for pesticide regulation and fish and wildlife management,
the Agency is now convinced that the best interests of
endangered species protection will be served by withdrawing
the proposed geographical and seasonal restrictions for all
species. The Agency believes that certain measures instituted
in California are of particular importance in protecting
endangered species. These measures include:
0 The stringent regulatory controls imposed by California
officials on the use and possession of 1080.
0 The Joint Policy Statement regarding rare and endangered
species signed by CDFA, the California Department of Fish
and Game, and the California Agricultural Commissioners
Association.
0 The considerable field experience and training of
state employees involved with the distribution and
use of 1080.
The Agency will require that specific programs developed
for endangered species by California officials be submitted
to the Agency for review and approval prior to 1080 use in the
range of the California condor and kit fox. Existing programs
for the salt marsh harvest mouse, Morro Bay kangaroo rat and
Aleutian Canada goose are now acceptable. Any subsequent
program changes must also be submitted to EPA for review. As
part of EPA's review the Office of Endangered Species, USDI,
will be consulted. Specific label instructions and cautions
regarding endangered species will be required. These label
statements are detailed in Chapter V.
-------
-47-
B. Rangeland and Cropland; Prairie Dogs
1. Proposed Actions in PD 2/3
In PD 2/3, the Agency proposed the following modifi-
cations of the terms and conditions of registration for this
use:
a. Standardize bait concentration at a maximum of 0.02%
active ingredient at a dosage of one teaspoon per burrow
spread over a three square feet area.
b. Standardize baiting and post-baiting procedures in
the same manner as proposed for "Rangeland and Cropland
Rodents: Ground Squirrels."
c. Restrict use to governmental agencies or persons under
the direct supervision of members of governmental agencies,
d. Allow the use of 1080 in a prairie dog town only if
a precontrol survey, conducted in accordance with a
protocol approved by EPA, in consultation with USFWS,
does not indicate the presence or possible presence
of a blackfooted ferret.
2. Comments
USDA, and several state and county agencies responded to
the above proposals. Concerning proposal (a), reduction in
bait concentration and application rate, USDA quoted from PD
2/3 that "The Agency has no direct data to specify an active
ingredient concentration of less than 0.11 percent." USDA,
therefore, recommended that the bait concentration of 1080
for prairie dog control remain the same until studies by the
Fish and Wildlife Service are completed. New Mexico also
took issue with the Agency's proposal and noted the inadequacy
of basing the bait reduction on LD 50 values rather than
field efficacy data. South Dakota made similar observations
stating that the only realistic way to determine the lowest
concentration level is by actually testing the bait under
conditions encountered in the field. Montana commented that
time and money are needed to insure adequate field testing of
the 0.02% bait and if it is efficacious it should be made
available to all states. Colorado, which has the only
permissible product for prairie dog control, commented that
based on their many years of field experience the 0.02% bait
concentration level and application rate of 1 teaspoonful per
burrow would not work. Colorado also questioned where the
Agency's supporting data are for justifying the 0.02% level.
The Board of County Commissioners in Ouray and Bent County,
Colorado, also expressed concern that lowering the bait
concentration would result in reduced efficacy and the need
for additional baiting.
-------
-48-
Concerning proposal (b), standardization of baiting and
post-baiting procedures, USDA commented that the same objections
that applied to ground squirrels apply to prairie dogs. They
also noted that the hazard with prairie dogs appears to be with
the black-footed ferret and few if any ferrets are known to
exist in Colorado.
USDA and New Mexico commented that proposal (c), limitation
of use to government agencies or persons under the direct
supervision of members of government agencies, was unnecessary
and not justified. Both recommended that use be allowed by
certified applicators. USDA noted that no government agency
has a sufficient number of trained people to carry out a
large program without assistance of the private landowner
and that such a requirement would add significant additional
costs with no evidence of increased efficacy or reduced
risk.
Proposal (d), the use of a precontrol survey for the black-
footed ferret, was commented on by USDA, New Mexico and South
Dakota. USDA recommended that the Colorado Division of
Wildlife or appropriate States be consulted and their historical
records of black-footed ferret presence or absence be considered.
USDA stated that they did not support a costly and needless
survey when there is no sign or record of the species in the
area for the last 20 to 30 years. They also recommended that
informal consultation or § 7 consultation under the Endangered
Species Act involving the appropriate State would suffice
rather than the more costly and restrictive modifications of
the survey.
New Mexico's comments and recommendations were similar,
i.e. reliance on State wildlife agencies and consultation with
FWS rather than the survey. South Dakota commented that until
procedures for the black-footed ferret survey are formalized,
they were unable to comment on their suitability or practicality,
however they note that such survey methods should not be so
complex and time consuming that they are neither practical or
economically feasible.
The Tri-River area Cooperative Extension Service (Colorado)
commented that prairie dogs are carriers of disease harmful
to humans and must be controlled. They noted that 1080 and
strychnine are the most effective means of control.
-------
-49-
3. Agency Response
The Agency agrees with commenters that a determination of
the most efficacious bait concentration and application rate
should be based on field test data. USDA's comment that EPA
noted in PD 2/3 that it did not have data to specify a bait
concentration of less than 0.11% is correct, but the Agency
goes on to explain in PD 2/3 that the Agency was convinced,
based on LD50 values that a concentration less than 0.11%
would be efficacious and 0.02% appeared to be the best estimate.
The Agency notes that the U.S. Forest Service/USDA is conducting
1080 field efficacy studies. These studies should provide an
indication of appropriate bait concentration for prairie dog
control. The Agency believes that the studies will support
the Agency's conclusions that a bait concentration lower than
the current .11% will be efficacious.
Based on the concerns raised by commenters about the
potential adverse effects from lowering the bait concentration,
the Agency has modified its position regarding the use of the
.11% bait for the 1985 control season. The details are
described in Chapter V.
The Agency response to comments on baiting and post
baiting procedures is the same as for ground squirrels.
These procedures will be modified accordingly for prairie dog
control.
The Agency believes that because of the endangered
status and rarity of the black-footed ferret, relying solely
on training provided as part of the applicator certification
requirements is not sufficient to provide an adequate level
of protection. The Agency's intent in PD 2/3 was to allow
use by farmers and ranchers under the direct supervision of
governmental agency personnel. A decision regarding the
distribution and possession of 1080 for prairie dog control
will depend on an assessment of the Colorado program for the
protection of the black-footed ferret. Clarification of
these and other issues, such as training of ranchers and
farmers, what constitutes "direct supervision", etc. will be
addressed following approval by EPA of a precontrol survey.
A precontrol survey for the black-footed ferret has been
the central concern of the Agency in determining whether to
allow the continued use of 1080 for prairie dog control. The
Agency has asked USDI/FWS, the experts in wildlife management
and protection, for an opinion and guidance on the use of a
precontrol survey. USDI's position is that a precontrol
survey prior to the use of 1080, strychnine and several other
toxicants can provide adequate protection for the ferret,
provided it is conducted in strict accordance with approved
-------
-50-
protocols. The Agency, USDI and several state and farm
organizations have been seeking to develop and agree upon an
adequate survey in connection with the use of strychnine for
prairie dog control. However, at this time, there are many
elements regarding the precontrol survey that have not been
adequately clarified or agreed upon.
The Agency notes in response to USDA's suggestion that
historical records of ferret presence or absence and consultation
with State wildlife agencies be relied upon to decide if 1080
treatment should proceed might well result in secondary
ferret exposure. The black-footed ferret currently has an
unknown range, and is very rarely spotted even in areas it is
believed to inhabit. In fact, the one known viable population
of black-footed ferrets, the group in the Meteetsee mountains
of Wyoming was discovered outside what was then considered
the range of the species. This argues for the continued use
of a precontrol survey in areas of potential black-footed
ferret habitat.
The potential human health hazard from diseases carried
by the prairie dog was not addressed by EPA. No data or
information supporting this contention was submitted by any
commenter or from health officials in Colorado or other states.
C. Rangeland, Cropland and Non-Agricultural Sites Except
Around Ships and Buildings; Other Rodents
1. Proposed Actions in PD 2/3
In the PD 2/3, the Agency proposed the denial of appli-
cations for the control of the following rodents on rangeland:
chipmunks, cotton rats, deer mice, kangaroo rats, and meadow
mice. The Agency also proposed the denial of applications for
the control of Norway rats, cotton rats, and wood rats on
croplands as well as denial of applications for chipmunks,
cotton rats, Norway rats, kangaroo rats, wood rats, deer
mice, and meadow mice on non-agricultural sites.
The Agency proposed denial of these uses because it
determined that the uses potentially put nontarget animals at
risk and safer alternatives were available. The Agency also
noted that because of low usage and the availability of
efficacious, economically viable alternatives, the impact on
overall benefits was expected to be low.
-------
PLEASE RETURN TO;
-51- NCIC/OTS CHEMICAL LIBRARY
401 M ST., S.W., TS-793
WASHINGTON, D.C. 20460
The Agency proposed to modify the terms and conditions
of, use for meadow mice control on cropland because this use
accounted for the majority of 1080 used on croplands and
denial could result in locally severe impacts on benefits
because of the increased cost of alternatives. The Agency
also concluded that risk reduction measures were available
that would enable benefits to outweigh risks. These measures
included:
i. Standardize the bait active ingredient concentration
at 0.02% and a dosage of one teaspoon for hand
baiting or four to five pounds per acre for aerial
baiting.
ii. Standardize baiting and post-baiting procedures for
hand and aerial application in accordance with the
proposed action under "Rangeland and Cropland Rodents:
Ground Squirrels."
iii. Prohibit use in the range of the Aleutian Canada
goose in accordance with the proposed action under
"Rangeland and Cropland Rodents: Ground Squirrels."
The Agency proposed no action for the use of 1080 to
control pocket gophers below ground on rangeland, cropland or
nonagricultural sites because the Agency believed a potential
for substantial risk to nontarget animals was unlikely.
2. Comments on PD 2/3
USDA noted that the PD 2/3 pointed out that no risk to
nontarget animals had been substantiated and only minimal
amounts of 1080 were used to control the rodents whose product
applications were to be denied. They noted that only a single
incident had been reported from any of these uses; one involving
geese mortality where both 1080 and zinc phosphide were used
for control of meadow voles. USDA noted that 1080 was principally
used to control highly irruptive populations of rodents that
have proven difficult to control with alternative rodenticides
and recommended that the uses be maintained. USDA stated
that they did not see how the potential risks at these use
sites was any greater than for use in and around ships and
buildings. USDA also cited the Agency's Preliminary Benefit
Analysis conclusion that 1080 is the most efficacious and
cost effective single dose rodenticide available for control
of field and commensal rodents.
-------
-52-
CDFA commented that the risks to nontargets from these
uses were minimal and the uses should be retained. New Mexico
also stated that cancelling these uses was not justified.
Another commenter, Nevada, stated that this proposal should
be reconsidered for meadow mice and kangaroo rats, since the
alternative pesticides, zinc phosphide and strychnine, were
more hazardous to birds than 1080.
3. Agency Response
The Agency maintains that these uses pose some risk to
nontarget species, that they have very low benefits, and
alternative pesticides are available. However, the Agency
is concerned about the the control of irruptive populations
of rodents and the possibility of increased risk to avian
species from alternative chemicals. Therefore, the Agency
will allow the continued use of 1080 to control all of the
rodent species proposed for denial in PD 2/3 provided registrants
agree to generate the data necessary for Federal registration,
and make certain use modifications and labeling changes
described in Chapter V.
Although no comments were received on the meadow mouse
use on cropland, the Agency will permit the continued use of
1080 for this use at current bait concentration and application
rates provided registrants make certain changes in baiting
and post baiting procedures and agree to generate the data
necessary for Federal registration.
Although no action was proposed for the pocket gopher
use and no comments were received, the Agency will call in
intrastate products bearing this use for Federal registration.
Registrants will be able to continue to use 1080 for this use
provided they agree to generate the data necessary for Federal
registration.
D. Use In and Around Ships and Buildings; Norway Rat, Roof
Rat and House Mice
The Agency proposed no action for the use of 1080 in and
around ships and buildings to control these three rodent
species. No comments were submitted on the proposal. Although
these uses are contained on a Federally registered product,
the Agency will require data for these uses.
E. General Comments on Proposed Actions
As a general comment, USDA suggested that no final decision
to further restrict or limit the use of 1080 be rendered
unless significant risks are demonstrated. While noting the
need to monitor and restrict the use of materials and techniques
-------
-53-
which present a real and significant threat to humans, domestic
animals, and non-target wildlife species, USDA commented that
we cannot afford to eliminate or regulate to an ineffective status,
needed, efficacious, cost effective control tools based on
assumptions, presumptions, conjecture or noncomparative analogies
and a lack of sufficient field data as presented in PD 2/3.
Several Agencies, most notably CDFA and USDA, have
emphasized the economic importance of 1080 and the significant
adverse effects on the agricultural economy that would result
if 1080 was denied or restricted in the manner proposed in PD
2/3.
In its response to comments, the Agency has noted throughout
this document its support of scientific decisions on the basis
of field data rather than extrapolations from LD50 data.
However, the kinds of data referred to by USDA and other
commenters do not exist. The Agency also notes that the stated
economic importance of 1080 should result in the interested
parties generating data to support the continued use of 1080.
-------
-54-
V. Final Regulatory Decision
A. General
There is only one 1080 product that is registered under
§ 3 of FIFRA. This product, which is a technical grade
material, bears directions for use to control certain rodents
in and around buildings and ships. Its label, however, does
not contain any use directions for control of field rodents.
All 1080 products that bear use directions for field rodent
control are intrastate products, except for a "special local
needs" registration in Oregon.
Intrastate products are distributed in only three states,
California, Colorado and Nevada. In other states where 1080
was used at one time for field rodent control, users relied
on the U.S. Department of the Interior's 1080 product. The
registration of this product was withdrawn following an
Executive Order prohibiting the use of 1080 on Federal lands.
In order for EPA to implement the regulatory position of
this document, intrastate products must be "called in" for
Federal registration. Intrastate producers who wish to
continue to sell and distribute their products must submit an
application for Federal registration. In addition, intrastate
producers must agree to: 1) provide the data necessary for
Federal registration, 2) modify procedures and product labeling
as described in this chapter, and 3) adhere to the schedule
found in Appendix A for submittal of data. The Agency will
consider the submittal of the application for Federal registration
as demonstrating the intent to provide the necessary data.
Failure to submit an application in a timely manner or to
adhere to the other reguirements of continued use may result
in the denial of the application for Federal registration and
a prohibition against the sale, distribution and use of the
1080 product in guestion.
Oregon's § 24(c) registrant and the registrant of technical
1080 will be reguired to provide a full complement of data to
support their registrations. A data call in under FIFRA
§ 3(c)(2)(B) will be issued to these registrants. Failure to
respond in a timely manner or commit to the generation of
data may result in the Agency suspending these registrations.
The Oregon § 24(c) registrant will have to modify his
product label in accordance with this document. The technical
registrant will also be regured to make certain label changes.
Specific changes will be detailed in notices sent to each
registrant.
-------
-55-
The Agency will continue to allow the use of 1080 to control
all field rodent species identified on current intrastate
and Federal products at current bait concentrations, except
for the control of ground squirrels in the range of the
California condor and for the control of prairie dogs in
Colorado. These uses will be permitted at current bait
concentrations during 1985. Beginning in 1986, however, bait
concentrations for prairie dog control must be reduced to .02%
unless efficacy data are generated. If efficacy data show
that .02% is not effective, the Agency will consider changes
in the bait concentration provided data are available that
establish the lowest effective concentration level and the
Agency can determine that use at that level will not impact
the black-footed ferret.
If 1080 use is permitted in the range of the condor, after
1985, such use will be at a maximum bait concentration of .02%
unless field efficiacy data are submitted demonstrating that
the .02% level is not effective as well as data establishing
the lowest effective bait concentration. The Agency must also
determine if higher bait concentration will impact the condor.
A decision on the use of 1080 in the range of the condor has
been delayed pending discussions with the Office of Endangered
Species (OES).
The Agency recently received a biological opinion from
the Office of Endangered Species (OES) stating that use of
the 1080 livestock protection collar in the range of the
California condor would jeopardize the continued existence
of this species. The OES opinion emphasized that there are
only a few condors in existence and there is uncertainty about
the effects of the collar on the condor. Although there are
substantial differences in potential 1080 exposure from the
livestock protection collar and treated grain bait, the drastic
decline in the condor population (OES reported one source
estimated 9 to 11 condors in the wild) may reguire prohibition
of 1080 use in the range of the condor until safety data are
evaluated by the Agency. A decision on the use of 1080 in
the range of the condor will be made following discussion
with OES. The decision will be made by January 1, 1986.
Compound 1080 has been classified as a restricted use
pesticide, i.e. it can only be used by a certified applicator
or someone under the direct supervision of a certified applicator.
Restricted use classification does not apply to intrastate products
although generally states have limited the sale and use of
intrastate products to certified applicators if EPA has
restricted a corresponding Federal product or active ingredient.
As a condition of continued use, intrastate products will now
be reguired to bear the restricted use label statements
contained in 40 CFR 162.10(j)(2 ). The addition of the restricted
-------
-56-
use labeling on intrastate products does not negate the
necessity of maintaining current limitations on supervision,
use, possession, storage and disposal of 1080 required on
current labels.
The Agency has concluded that all 1080 grain baits must
be dyed a unigue standardized color. Available information
indicates that Auramine 0, a yellow dye used in California,
does not appear to impact on avian species. This may be the
result of repellency caused by the dye. Therefore, until
data are submitted supporting the use of another dye color,
the Agency is reguiring that all 1080 grain baits used for
field rodent control be dyed yellow. The producer of technical
1080 will be required to add a dye statement to the product
label. This statement is contained in section (F)(l)(d) of
this chapter.
Other applicants seeking Federal registration of any field
rodent use of 1080 must submit the data to obtain a Federal
registration prior to using 1080. The Agency believes it is
necessary that any expansion of the use of 1080 to other
states be predicated on an adeguate data base to support
such expansion.
Registrants and intrastate producers will have 30 days
to submit an application for amended registration or regis-
tration respectively, following receipt of notification from
EPA. Each registrant or intrastate producer will be notified
of the reguirements for compliance with this document by
certified mail. All 1080 intrastate products and the 24(c)
product that are sold, used, distributed, or released for
shipment after December 31, 1985, must comply with the
applicable parts of this chapter. Any 1080 product used for
reformulation that is released for shipment after December
31, 1985, must comply with the applicable parts of this section,
Existing stocks, except where specified otherwise, may be
used with a modified label after December 31, 1985.
As the data for Federal registration are submitted and
reviewed by the Agency, additional modifications in the terms
and conditions of use of 1080 products may be required. After
reviewing the data, the Agency may also determine that some
or all uses of a particular product should not be permitted and
will cancel the registration or deny the application for
Federal registration, which ever is appropriate, for those
uses in guestion.
The following sections in this chapter describe the
reguirements necessary for the continued use of existing 1080
products. Section F of this chapter details all of the
requirements.
-------
-57-
B. Rangeland, Cropland and Non-Agricultural Sites;
Ground Squirrels
1. Reduction and Standardization of Bait Concentration
and Application Rate's" ~~~~
The Agency has decided to allow use at current label
bait concentrations until efficacy data are generated,
except in the range of the California condor.
The Agency is concerned that use of ineffective baits
might result in overbaiting, increase costs to producers,
and might lead to the development of more 1080 resistant
populations of ground sguirrel. Because of these concerns,
the Agency believes, where appropriate, it is prudent to
base bait concentration levels on efficacy data. However,
the precarious nature of the condor population requires, at
a minimum, imposition of lower bait concentrations. It is
possible for reasons discussed earlier in this chapter that
the Agency may prohibit the use of 1080 in the range of the
condor until data are submitted showing that the use of 1080
will not jeopardize the continued existence of the species.
Because of the timing of the release of the PD 4, reduction
of bait concentrations for 1985 baiting programs cannot be
implemented. However, beginning in 1986 if 1080 use in the
range of the condor is permitted, baits must be at a maximum
concentration of .02% unless field efficacy data are submitted
demonstrating that the .02% level is not effective as well as
data establishing the lowest effective bait concentration.
The Agency must also determine if higher bait concentrations
will impact the California condor. A label statement regarding
use of 1080 at the .02% level will be reguired if use in the
range of the condor is permitted. This statement is presented
in section F of this chapter.
The Agency believes that its calculation of an efficacious
bait concentration level of .02%, which was based on available
LD 50 data and is consistent with Spencer's laboratory
observations (1946), will be substantiated by field efficacy
testing.
The generation of efficacy data establishing the lowest
efficacious bait concentration will be one of the first
requirements of applicants in support of Federal registration
of their product.
2. Standardization of Hand Baiting and Post-Baiting
Procedures
Concerning the standardization of hand baiting and
post-baiting procedures, the Agency concludes that the requirement
-------
-58-
to keep pets and domestic animals away is overly restrictive,
and would, as a practical matter, prohibit use since there
are very few places where free-roaming pets and livestock do
not occasionally occur. Therefore the label statement has
been revised to read:
Do not expose baits in manner which presents a
likely hazard to pets, poultry, or livestock.
The Agency has concluded that the proposed requirement
to bury or burn all visible carcasses cannot be reasonably
implemented in some situations and has revised this requirement.
Therefore the label statement has been revised to read:
Where possible, pick up and burn or bury deeply all
visible carcasses of animals killed by 1080.
3. Standardization of Aerial Baiting and Post-Baiting
Procedures
The Agency has revised the proposed procedures for
disposal of carcasses to be consistent with those for hand
baiting procedures. Specific requirements for aerial baiting
are presented in section (F)(2)(b) of this chapter.
4. Special Restrictions for Endangered Species Protection
As noted in Chapter IV, the Agency has withdrawn the
seasonal and geographic restrictions proposed for continued use
of 1080 in California. Compound 1080 bait products used in
California will be required to bear the following endangered
species statement:
Notice: The killing of a member of an endangered species
during compound 1080 baiting operations may result in a
fine and/or imprisonment under the Endangered Species
Act. Before baiting, the user is advised to contact the
the local Fish and Game Office for specific information
on endangered species. Do not use compound 1080 in the
geographic ranges of the following species except under
programs and procedures approved by the USEPA: California
condor, San Joaquin kit fox, Aleutian Canada goose, Morro
Bay kangaroo rat, and salt marsh harvest mouse. Users
must contact the local fish and Game Office for specific
information on EPA approved programs and procedures re-
garding endangered species.
Other states, as appropriate, will be required to incorporate
a statement similar to the one above on products used to
control ground squirrels. This statement is contained in
section (F)(2)(c) of this chapter.
-------
-59-
In PD 2/3 the Agency's proposed actions concerning the black-
footed ferret were limited to risks associated with the control of
prairie dogs. However, the black-footed ferret is also
potentially at risk from ground squirrel control operations.
The Agency is requiring that the following use restriction be
placed on the Colorado 1080 product label for prairie dog
control:
Do not use for ground squirrel control within 200 yards
of prairie dog colonies unless a precontrol survey for
the black-footed ferret has been performed prior to
control and produces no evidence that a black-footed
ferret is present in the survey area. Do not use within
five miles of a prairie dog colony where a black-footed
ferret has been confirmed to be present.
5. Other Label Statements
Additional label statements concerning hazard to nontarget
wildlife, restricted use classification, and bait dyes are
contained in section F of this chapter.
C. Rangeland and Cropland; Prairie Dogs
The Agency has concluded that the bait concentration for
prairie dog control should be lowered to .02% in the absence
of any data that demonstrates this level does not provide an
effective level of control. As a practical matter, the
timing of the release of this document precludes Colorado
from reducing bait concentrations for the 1985 season.
However, beginning in 1986 the use of 1080 for prairie dog
control will be at a maximum bait concentration of .02%
unless field efficacy data are submitted demonstrating that
the .02% level is not effective as well as data establishing
the lowest effective bait concentration. The Agency must
also determine if higher bait concentrations will impact the
endangered black-footed ferret. A label statement regarding
use of 1080 at the .02% level for prairie dog control is
required and is presented in section (F)(3)(a) of this chapter.
The Agency has made changes in the proposed baiting and
post-baiting procedures consistent with those procedures_for
ground squirrels. These procedures are detailed in Section F
of this chapter. Other label statements concerning endangered
species, hazard to nontarget wildlife, restricted use
classification, and bait dyes are contained in section F of
this chapter.
The Agency has concluded that the use of 1080 for prairie
dog control in Colorado should be by government agency
personnel or under the direct supervision of such personnel.
As with the reduction in bait concentration, the Agency will
-------
-60-
not require changes in the 1985 Colorado program. Specific
details regarding use supervision, possession and storage of
1080, etc. will be resolved if a precontrol survey is
approved for the black-footed ferret.
Use after December 31, 1985, will require the use of an
EPA approved program including a precontrol survey for the
black-footed ferret. Colorado should submit the precontrol
survey by January 1986, in order to allow the Agency adequate
review time.
Other states seeking to use 1080 for prairie dog control
must obtain a § 3 registration prior to use.
D. Rangeland, Cropland and Non-Agricultural Sites Except
Around Ships and Buildings: Other Rodents
The Agency has concluded that the uses proposed for
denial in PD 2/3 can be continued provided registrants agree
to generate the data necessary for Federal registration, submit
this data on schedule, and make certain modifications in
baiting and post-baiting procedures, add endangered species
precautions and restricted use label statements and comply
with the requirements of this document. These same conditions
apply to meadow mice on cropland, which was proposed for
continued use in PD 2/3. The same limitations on use in the
range of the condor as required for ground sguirrels apply
to these rodent species. The uses proposed for denial in PD
2/3 were:
(1) Rangeland: chipmunks, cotton rats, deer mice,
kangaroo rats and meadow mice.
(2) Cropland: Norway rats,cotton rats and wood rats.
(3) Non-Agricultural Sites: chipmunks, cotton rats,
Norway rats, kangaroo rats, wood rats, deer mice
and meadow mice.
The specific requirements of continued use are detailed in
section F of this chapter.
No action was proposed in PD 2/3 for pocket gophers.
Registrants of this use will be "called in" and will
have to agree to generate data for Federal registration.
E. Use In and Around Ships and Buildings: Norway Rat, Roof
Rat and House Mice
Although these uses are contained on a Federally registered
product, the Agency will require the full complement of data
necessary for registration under $ 3 of FIFRA.
-------
-61-
F. Regulatory Requirements
1• General Requirements
All intrastate products will be "called in" for Federal
registration. A § 3(c)(2)(B) data call in notice will be
issued for the § 24(c) product in Oregon and for the technical
1080 product. The § 24(c) registrant and the technical
registrant will be required to submit a full complement of
data identified in Appendix A to support the continued
registration of their products.
All applicants for Federal registration and current
Federal registrants will be allowed to continue selling,
distributing and using their products provided they:
0 Commit to generate the data necessary for Federal
registration under § 3 of FIFRA and submit the data
on schedule as outlined in Appendix A.
0 Make certain changes in baiting and post-baiting
procedures, add endangered species and nontarget
precautions and restricted use statements to product
labels and comply with the other requirements
specified in this chapter.
The following requirements apply to all field rodenc
use products containing 1080:
a. Hand Baiting and Post-Baiting Procedures
The following procedures are to be placed on all 1080
product labels:
i. Baiting should not be done unless tests indicate
satisfactory bait acceptance will occur in areas
to be treated.
ii. Do not expose baits in a manner which presents a
likely hazard to pets, poultry, or livestock.
iii. Clean up all accidentally spilled bait immediately.
iv. Do not place bait in piles.
v. Where possible, pick up and burn or bury deeply
all visible carcasses of animals killed by 1080.
vi. Do not use within 1/4 mile of a dwelling without
first notifying the occupants.
b. Nontarget Species Label Precaution
The following statement must be placed on all 1080
product labels:
-------
-62-
This product is very highly toxic to wildlife.
Birds and mammals feeding on target organisms
or treated bait may be killed. Keep out of any
body of water. Apply this product only as
specified on this label.
c. Restricted Use Classification Label Statement
All 1080 products may be used only by certified applicators
or persons under their direct supervision in addition to any current
use supervision requirements or restrictions. All 1080 products
must bear the restricted use label statements specified in
40 CFR 162.10(j)(2). The statement "Restricted Use Pesticide"
must appear on the front panel of the label. Directly below
"Restricted Use Pesticide" the following statement is required:
For retail sale to and use only by Certified Applicators
or persons under their direct supervision and only
for those uses covered by the Certified Applicator's
certification.
Modifications in this statement will be permitted to reflect
state requirements limiting sale, distribution and possesion of
1080.
d. Bait Dyes
Any bait containing directions for use to control field
rodents must be dyed yellow. The label of any 1080 product
that contains directions for formulating end-use baits for
field rodent control must contain the following statement:
Formulators must ensure that all compound 1080 grain
baits used for field rodent control are dyed yellow.
The label of any 1080 product that contains directions
for formulating end-use baits for commensal rodent control
must contain the following statement:
Compound 1080 grain baits used for commensal rodent
control must be dyed in accordance with USDA
regulations.
2. Rangeland, Cropland and Non-Agricultural Sites;
Ground Sguirrels
In addition to the reguirements cited in section 1 of
this chapter, 1080 products bearing directions for use for
the control of ground sguirrels must comply with the
following requirements.
-------
-63-
a. Bait Concentration Levels
All products may continue to be used at current bait
concentration levels until efficacy data are generated to
establish the lowest efficacious concentrations, except in
the range of the California condor. Through December 31, 1985,
use of 1080 in the range of the condor may"continue under
present label conditions. However, as noted in section A of
this chapter, a recent jeopardy opinion from OES on the livestock
protection collar necessitates further evaluation of the use of
1080 grain baits in the range of the condor. Beginning in
1986, if permitted, the use of 1080 in the range" of the
condor will be at a maximum bait concentration of .02% unless
field efficacy data are submitted demonstrating that the .02%
level is not effective as well as data establishing the lowest
effective bait concentration. The Agency must also determine
if higher bait concentrations will impact the California condor.
The following statement must be placed on 1080 product
labels used for ground squirrel control in California if use
in the range of the condor is permitted:
Baits at concentrations greater than .02% may not
be used in the range of the California condor.
b. Aerial Baiting and Post-Baiting Procedures
The following statement must be placed on 1080 product
labels used for ground squirrel control in California:
i. Use in accordance with the Guidelines for Applying
Rodent Baits by Aircraft for Control of Ground
Squirrels in the California Vertebrate Pest Control
Handbook.
ii. Where possible, pick up and burn or bury deeply
all visible carcasses of animals killed by 1080.
Aerial application of 1080 baits may continue in other
states as permitted by current labels. However, applicants
for Federal registration must submit aerial application
procedures and guidelines for Agency review at the time they
submit the application for Federal registration. A reference
to these procedures will be incorporated on the appropriate
product label.
c. Endangered Species Label Precautions
The following statement must be placed on 1080 product
labels used for ground squirrel control in California:
-------
-64-
Notice: The killing of a member of an endangered species
during compound 1080 baiting operations may result in a
fine and/or imprisonment under the Endangered Species Act.
Before baiting, the user must contact the local Fish
and Game Office for specific information on endangered
species. Do not use compound 1080 baits in the geographic
ranges of the following species except under programs and
procedures approved by the USEPA: California condor, San
Joaguin kit fox, Aleutian Canada goose, Morro Bay kangaroo
rat, and salt marsh harvest mouse.
The following statement must be placed on 1080 product
labels used for ground squirrel control in Colorado, Nevada
and Oregon.
Notice: The killing of a member of an endangered species
during compound 1080 baiting operations may result in a
fine and/or imprisonment under the Endangered Species Act.
Before baiting, the user must contact the local Fish
and Game Office or the Regional Office of the Fish and
Wildlife Service for specific information on endangered
species.
The following statement must be placed on 1080 product
labels used for ground sguirrel control in and around prairie
dog colonies:
Do not use for ground sguirrel control within 200 yards
of prairie dog colonies unless a precontrol survey for
the black-footed ferret has been performed prior to
control and the use of such survey produces no evidence
that a black-footed ferret is present in the survey area.
Do not use within five miles of a prairie dog colony
where the presence of a black-footed ferret has been
confirmed to be present.
3. Rangeland and Cropland; Prairie Dogs
In addition to the requirements cited in section 1 of
this chapter, the 1080 product bearing directions for use for
the control of prairie dogs must comply with the following
requirements:
a. Bait Concentration Levels
This product may continue to be used at current bait
concentration levels and label conditions through December
31, 1985. Beginning in 1986, the use of 1080 for prairie dog
control will be at a maximum bait concentration of .02%
unless field efficacy data are submitted demonstrating that
the .02% level is not effective as well as data establishing
-------
-65-
the lowest effective bait concentration. The Agency roust
also determine if higher bait concentrations will impact the
endangered black-footed ferret.
The following statement must be placed on any 1080 product
used to control prairie dogs:
Baits at concentrations greater than .02% may
not be used for the control of prairie dogs.
b. Use Supervision
The following statement must be placed on any 1080 product
used to control prairie dogs:
The use of 1080 for prairie dog control is restricted to
governmental agencies or persons under the direct
supervision of members of governmental agencies.
c. Precontrol Survey for the Black-footed Ferret
The following statement must be placed on any 1080 product
used to control prairie dogs:
Compound 1080 can be used for the control of prairie
dogs only if an EPA-approved survey for the black-footed
ferret has been performed prior to control and the use
of such a survey produces no evidence that a black-footed
ferret is present in the survey area.
d. Endangered Species Label Precautions
The following statement must be placed on any 1080 product
used to control prairie dogs:
Notice: The killing of a member of an endangered species
during compound 1080 baiting operations may result in a
fine and/or imprisonment under the Endangered Species
Act. Before baiting, the user must contact the local
Fish and Game Office or the Regional Office of the Fish
and Wildlife Service for specific information on endangered
species.
4. Rangeland, Cropland and Non-Agricultural Sites Except
Around Ships and Buildings; Other Rodents
0 Rangeland: chipmunks, cotton rats, deer mice,
kangaroo rats and meadow mice.
0 Cropland: meadow mice, Norway rats, cotton rats and
wood rats.
-------
-66-
0 Non-Agricultural Sites: chipmunks, cotton rats,
Norway rats, kangaroo rats, wood rats, deer mice
and meadow mice.
In addition to the reguirements cited in section 1 of
this chapter, 1080 products bearing directions for use for
the control of the above rodents must comply with the
following reguirements:
a. Bait Concentration Levels
All products may continue to be used at current bait
concentration levels until efficacy data are generated to
establish the lowest efficacious concentrations, except in
the range of the California condor. Through December 31,
1985, use of 1080 in the range of the condor may continue
under present label conditions. However, beginning in 1986
the use of 1080 in the range of the condor will be at a
maximum bait concentration of .02% unless field efficacy data
are submitted demonstrating that the .02% level is not effective,
that another higher level is effective, and that the higher
bait concentration will not adversely affect the condor.
The following statement must be nlaced on 1080 product
labels used for rodent control in California:
Baits at concentrations greater than .0?% may not
be used in the range of the California condor.
b. Endangered Species Label Precautions
The following statement is to be placed on all 1080 products
used in California for control of rodents identified in
section (F)(4):
Notice: The killing of a member of an endangered species
during compound 1080 baiting operations may result in a
fine and/or imprisonment under the Endangered Species
Act. Before baiting, the user must contact the local
Fish and Game Office for specific information on
endangered species. Do not use compound 1080 baits in
the geographic ranges of the following species except
under programs and procedures approved by the
USEPA: California condor, San Joaguin kit fox, Aleutian
Canada goose, Morro Bay kangaroo rat, and salt marsh
harvest mouse.
The following statement is to be placed on all 1080 products
used in states other than California for control of rodents
identified in section (F)(4):
-------
-67-
Notice: The killing of a member of an endangered species
during compound 1080 baiting operations may result in a
fine and/or imprisonment under the Endangered Species
Act. Before baiting, the user must contact the local
Fish and Game Office or the Regional Office of the Fish
and Wildlife Service for specific information on endangered
species.
5. Pocket Gophers
No additional label modifications are required for this
use at this time except the restricted use statement contained
in section (F)(l)(c) and the bait dye statement contained in
(F)(l)(d)-of this chapter.
6. Use In and Around Ships and Buildings; Norway Rat, Roof
Rat and House Mice
No additional label modifications are required for this
use at this time except the dye color statement contained in
section (F)(l)(d) of this chapter.
-------
-68-
VI. BIBLIOGRAPHY
Fish and Wildlife Service correspondence. July 21, 1978.
Hagen, H., 1972. A review of the use of toxic chemicals for
mammalian animal control in California. California
Department of Fish and Game, Wildlife Management
Administration Report No. 72-75.
Hegdal, P.L., Gatz, T.A., Fite, E.G., 1980. Secondary effects
of rodenticides on mammalian predators. Denver Wildlife
Research Center. (Unpublished).
Hegdal, P.L., Gatz, T.A., Fagerstone, K.A., Glahn, J.F.,
Matsche, G.H., 1979. Hazards to wildlife associated
with 1080 baiting for California ground squirrels.
Final report of the U.S. Fish and Wildlife Service under
Interagency Agreement EPA-IGA-D7-0449.
Ketron, Inc., 1979. Assessment of the environmental effects
of predator and rodent control programs in Wyoming using
strychnine and 1080. KFR 232-79.
Marsh, R.F., 1967. Aircraft as a means of baiting ground
squirrels. Proc. 3rd Vert Control Conf., San Francisco,
March 7-9, 1967. pp 2-6.
Myers, K., 1980. Zoology Department, University of Guelph,
Ontario. Personal communication.
Pearson, O.P., 1966. The prey of carnivores during one cycle
of mouse abundance. J. Animal Ecology., 35:217-233.
Pearson, O.P., 1971. Additional measurements of the impact
of carnivores on California voles. (Microtus californicus)
J. Mammal., 52:41-49.
Schitoskey, F., 1975. Primary and secondary hazards of three
rodenticides to kit fox. J_. Wildl. Mgmt. , 39 ( 2 ): 416-418 .
Sibley,F.C., 1966. 1080 squirrel poisoning operation. Bureau
of Sport Fish and Wildlife.
Spencer, D.A., 1946. Compound 1080 - Sodium Flouroacetate,
As a control agent for field rodents. National Research
Council Insect Control Committee Report No. 161. Partial
interim report.
Swick, C.D., 1973. Determination of San Joaquin kit fox
range in Contra Costa, Alameda, San Joaquin, and Tulare
Counties. State of California Department of Fish and
Game.
-------
-69-
Swick, C.D., 1973. San Joaguin kit fox - an impact report of
secondary hazards of aerial application of 1080 grain
baits for ground squirrel control in San Luis Obispo
County. State of California Department of Fish and Game.
Tucker, R.K.f 1965-1972. Internal report series in pharmacology.
United States Department of the Interior, Fish and
Wildlife Service, Denver Wildlife Research Center (Unpublished)
United States Environmental Protection Agency. Position
Document 1. Rebuttable Presumption Against Registration:
Compound 1080.
United States Environmental Protection Agency. Position
Document 2/3. Sodium Monoflouroacetate (Compound 1080).
United States Environmental Protection Agency, Science Advisory
Panel. Report of the Ad Hoc Study Group on Sodium
Monoflouroacetate. Washington D.C. July 3, 1984.
Wagner, F.H. and L.C. Stoddart, 1972. Influence of coyote
predation on black-tailed jackrabbit population in
Utah. Journal of Wildlife Management., 36(2):329-342.
-------
APPENDIX A
DATA REQUIREMENTS* AND SUBMISSION SCHEDULE
FOR COMPOUND 1080 REGISTRATION
158.120 PRODUCT CHEMISTRY DATA REQUIREMENTS:
Product Identity:
Guideline
Reference
Time Allowed
To Submit Study
Results to EPA
Identity of Ingredients
61-1
6 months
Statement of Composition
61-2
1
year
Discussion of Formation of Ingredients
61-3
year
Analysis & Certification of Product
Ingredients;
Certification of Limits
Analytical Method for Enforcement of
limits
62-2
62-3
6 months
1 year
Physical & Chemical Properties;
Color
Physical State
Odor
Melting Point
Boiling Point
Density, Bulk Density, or Specific Gravity
Solubility
Vapor Pressure
Dissociation Constant
Octanol/Water Partition Coefficient
PH
Stability
Oxidizing or Reducing Action
Flammability
Explodability
Storage Stability
Viscosity^
Miscibility
Corrosion Characteristics
Dielectric Breakdown Voltage
63-2
63-3
63-4
63-5
63-6
63-7
63-8
63-9
63-10
63-11
63-12
63-13
63-14
63-15
63-16
63-17
63-18
63-19
63-20
63-21
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
year
year
year
year
year
year
year
year
year
year
year
year
year
year
year
year
year
year
year
year
*Agency review of these data may reguire the development of additional
safety data or benefits data. The specific data reguirements that
apply to basic producers and end users will be identified
in the Notice "Calling In" intrastate products for registration arid in
the 3(c)(2)(B) data call in notice sent to Federal registrants. If the
basic producer does not agree to generate the reguired data, end users
must agree to submit such data. Some of the data reguirements of this
Appendix may already have been satisfied in conjunction with the regis-
tration of the livestock protection collar or other registration
activities.
-------
- 2 -
158.125 RESIDUE CHEMISTRY DATA REQUIREMENTS;
Registrants must submit a detailed description of each
use with their application for Federal registration.
This information will be used by the Agency to determine if a
particular use is a food or non-food use. The information
needed for each use site includes:
(1) How the bait is applied, e.g. bait box, broadcast ground
application, aerial application.
(2) Rate of application.
(3) The number or freguency of applications.
(4) The minimum interval between applications.
If EPA determines what a particular use is a food use,
applicants will be notified of which data reguirements
listed below will be applicable.
Time Allowed
to Submit Study
Results to EPA*
Chemical Identity^
Directions
for
Use
17
17
1-
1 —
2
3
1
1
year
year
Nature of Residue:
Plants
Livestock
171-4
171-4
3 years
2 years
Residue Analytical
Magnitude of the
Method and
Residue:
Crop Field Trials
Processed Food/Feed
Meat/Milk/Poultry/Eggs
Potable Water
171-4
171-4
171-4
171-4
2 years
2 years
2 years
2 years
Reduction of Residue
Proposed Tolerance
(includes $40,000 filing fee)
Reasonable Grounds in Support of Petition
Submittal of Analytical Reference Standards
171-5
171-6
171-7
171-13
2 years
2 years
2 years
* Time begins following determination of food use.
-------
- 3 -
158.130 ENVIRONMENTAL FATE DATA REQUIREMENTS:
Degradation Studies (Lab):
Metabolism Studies (Lab):
Mobility Studies:
Time Allowed
to Submit Study
Results to EPA
Hydrolysis
Phot ©degradation:
On Soil
161-1
161-3
1 year
1 year
Aerobic Soil
Anaerobic Aquatic
161-1
162-3
3 years
3 years
Leaching (Absorption/Desorption 163-1 [ 1 year
-------
- 4 -
15R.135 TOXICOLOGY DATA REQUIREMENTS:
Time Allowed
to Submit Study
Results to EPA
Acute Testing:
Dermal LD50 - Rabbit (Preferred Species)
Primary Eye Irritation - Rabbit
Primary Dermal Irritation - Rabbit
Dermal Sensitization - Guinea Pig
81-2
81-4
81-5
81-6
1 year
1 year
1 year
1 year
The following tests will be necessary if 1080 tolerances are required.
Subchronic Testing:
90-Day Feeding:
Rodent
Non-Rodent (Dog)
21-Day Dermal - Rabbit
90-Day Dermal
90-Day Inhalation
90-Day Neurotoxicity - Hen, Mammal
82-1
82-1
82-2
82-3
82-4
82-5
9
15
6
9
9
9
months
months
months
months
months
months
Chronic Testing:
Chronic Feeding;
2 spp. Rodent & Non-Rodent (Dog)
83-1
Oncogenicity Study:
2 spp. Rat & Mouse (Preferred)
Teratogenicity (2 Species) - Rat
mouse, hamster, Rabbit
Reproduction ( 2-Generation) Rat or
other species
Mutagenicity Testing:
Gene Mutation (Ames Test)
Chromosomal Aberration
Other Mechanisms of Mutagenicity
Special Testing:
General Metabolism - Rat
Dermal Penetration
83-2
83-3
83-4
84-2
84-2
84-4
4
vears
4
9
2
6
6
6
85-1 1
85-2
Special Requirement:
Domestic Animal Safety
86-1
6
years
months
1/2 years
months
months
months
jyear
months
*
Variable depending on species and type of study basis
-------
_ c _
158.145 WILDLIFE AND AQUATIC ORGANISM DATA REQUIREMENTS:
Time Allowed
to Submit Study
Register to EPA
Avian and Mammalian Testing;
Field efficacy data to establish the lowest
effective bait concentration.
Ferret Mortality Study
Ferret Survey Study
Protocol for Ferret Survey Study
Avian Oral LD50 71-1
Avian Dietary LC50 71-2
Avian Reproduction 71-4
Avian and Mammalian Secondary Hazard Studies
*Simulated & Actual Field Testing- 71-5
mammals and birds
1 year
1 jfear
1 year
4 months
1 year
1 year
3 years
1 year
1.5-3 years
depends upon
number of years
to be examined
Aquatic Organism Testing;
Freshwater Fish LC50
Acute LC50 Freshwater Invertebrates
72-1
72-2
*Fish Early Life Stage and Aguatic
Invertebrate Lifecycle
*Fish - Lifecycle
*Aguatic Organism Accumulation
*Simulated or Actual Field Testing -
Aguatic Organisms
72-5
72-6
72-7
1 year
1 year
1 year
1.5 year
9 months
1.5-3 years
depends upon
number of years
* Reserved - depends on results of above tests.
------- |