[Federal Register Volume 59, Number 225 (Wednesday, November 23, 1994)]
[Unknown Section]
[Page 0]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 94-28821]


[[Page Unknown]]

[Federal Register: November 23, 1994]


_______________________________________________________________________

Part IV





Environmental Protection Agency





_______________________________________________________________________



Plant-Pesticides Subject to the Federal Insecticide, Fungicide, and 
Rodenticide Act (FIFRA) and the Federal Food, Drug, and Cosmetic Act 
(FFDCA); Proposed Policy; Notice



_______________________________________________________________________



40 CFR Parts 152, 174, and 180



Proposed Exemptions from the Requirement of a Tolerance for Plant-
Pesticides and Nucleic Acids and Viral Coat Proteins Produced in Plants 
Under FFDCA, and Plant-Pesticides Subject to FIFRA; Proposed Rules
ENVIRONMENTAL PROTECTION AGENCY

[OPP-300370; FRL-4755-2]
RIN 2070-AC02

 

Proposed Policy; Plant-Pesticides Subject to the Federal 
Insecticide, Fungicide, and Rodenticide Act and the Federal Food, Drug, 
and Cosmetic Act

AGENCY: Environmental Protection Agency (EPA).

ACTION: Statement of policy.

-----------------------------------------------------------------------

SUMMARY: This notice describes how EPA proposes to address pesticidal 
substances produced by plants under the Federal Insecticide, Fungicide, 
and Rodenticide Act (FIFRA) and the Federal Food, Drug, and Cosmetic 
Act (FFDCA). Substances that plants produce to protect themselves 
against pests and disease are pesticides under the definition of FIFRA 
section 2, (i.e., if they are ``. . .intended for preventing, 
destroying, repelling, or mitigating any pest. . .'') regardless of 
whether the pesticidal capabilities evolved in the plants or were 
introduced by breeding or through the techniques of modern 
biotechnology. These substances, along with the genetic material 
necessary to produce them, are designated ``plant-pesticides.'' This 
policy statement: (1) Clarifies the regulatory status under FIFRA and 
FFDCA of plants and plant-pesticides; (2) stipulates that EPA's 
regulatory attention will focus on plant-pesticides rather than on 
plants per se; (3) describes the criteria EPA is proposing to use in 
determining which plant-pesticides will be subject to regulation and 
which will be exempt; and (4) describes EPA's proposed procedures and 
information needs for the regulation of testing and commercial sale and 
distribution of plant-pesticides.

DATES: Comments identified by the docket control number [OPP-300370] 
must be received on or before January 23, 1995.

ADDRESSES: Submit written comments by mail to: Program Resources 
Section, Public Response and Program Resources Branch, Field Operations 
Division (7506C), Environmental Protection Agency, 401 M St., SW., 
Washington, DC 20460. In person, bring comments to Rm. 1132, Crystal 
Mall #2, 1921 Jefferson Davis Highway, Arlington, VA 22202.
    Information submitted as a comment concerning this document may be 
claimed confidential by marking any part or all of that information as 
``Confidential Business Information'' (CBI). Information so marked will 
not be disclosed except in accordance with procedures set forth in 40 
CFR part 2. A copy of the comment that does not contain CBI must be 
submitted for inclusion in the public record. Information not marked 
confidential may be disclosed publicly by EPA without prior notice. All 
written comments will be available for public inspection in Rm. 1132 at 
the Virginia address given above from 8 a.m. to 4 p.m., Monday through 
Friday, excluding legal holidays.

FOR FURTHER INFORMATION CONTACT: By mail: Bernice Slutsky, Science and 
Policy Staff, Office of Prevention, Pesticides and Toxic Substances 
(7101), Environmental Protection Agency, 401 M St., SW., Washington, DC 
20460. Office location and telephone number: Rm. E-627, 401 M St., SW., 
Washington, DC, (202-260-6900).
SUPPLEMENTARY INFORMATION:

I. Introduction and History of the Policy

A. Introduction

    EPA has received numerous inquiries concerning the regulation of 
plants that have been modified to produce pesticidal substances, 
particularly since modern biotechnology has provided the means of 
introducing novel pesticidal substances into plants. These inquiries 
have come from industry, public interest groups, and other government 
agencies. The principal focus of these inquiries has been requests for 
clarification of the regulatory status, under FIFRA (7 U.S.C. 136 et 
seq.) and FFDCA (21 U.S.C. 321 et seq.), of plants and the pesticidal 
substances that they produce. -
    Most plant varieties have the ability to resist pests and disease. 
The mechanisms of resistance can be varied, including structural 
characteristics of the plant, the production of general metabolites 
that have toxic properties, or the production of specific toxic 
substances in response to pest attack. A plant can be completely immune 
to a pest or can be partially resistant.
    Plant varieties with a greater ability to withstand pests have 
traditionally been bred from progenitor plants that have high levels of 
resistance to the target pest. It is now also possible to introduce 
into plants mechanisms of pest and disease resistance that are not 
found in the plant kingdom. For example, plants can be modified to 
express toxins from invertebrates and microorganisms. These toxins can 
confer plant resistance to insect attack and disease. Such pesticidal 
substances can be diverse and can potentially originate from any 
taxonomic kingdom.
    There are a number of types of substances produced in plants that 
enable plants to resist pest attack and disease. These substances 
include both those pesticidal substances that would be considered 
normally a component of a plant and those that would be considered new 
to a plant. Examples of plant-pesticides that would be considered 
normally a component of a plant are phytoalexins (plant-produced 
substances that act against phytopathogenic microorganisms). An example 
of a plant-pesticide that would not be considered normally a component 
of a plant is the insecticidal delta endotoxin that is produced in the 
bacterium, Bacillus thuringiensis.
    This policy statement clarifies the regulatory status, under FIFRA 
and FFDCA, of plants that act as biological control agents (and thus 
can be considered pesticides) and the plant-pesticides produced by 
plants. In doing so, it clarifies that plants continue to be exempt, 
and it defines the categories of plant-pesticides that would be 
regulated by EPA under FIFRA and FFDCA. This document outlines EPA's 
proposed procedures to assess plant-pesticides at different stages of 
testing and at sale or distribution. It also describes the information 
that EPA would need to evaluate those plant-pesticides that the Agency 
is proposing would be subject to EPA regulation under FIFRA and FFDCA.
    This policy statement is based upon the Agency's current knowledge 
of new plant varieties under development in agricultural research, 
particularly those developed through the new techniques of 
biotechnology. Accordingly, while this policy statement would apply to 
all plant-pesticides produced in plants (including bryophytes such as 
mosses, seedless vascular plants such as ferns, gymnosperms such as 
conifers, and angiosperms such as most major crop plants), it 
concentrates primarily on proteinaceous plant-pesticides produced in 
new varieties of terrestrial crops.
    In developing its policy on plant-pesticides under FIFRA and FFDCA, 
the Agency considered how the two statutes authorize EPA to regulate 
pesticides and pesticide residues, the differences in statutory 
criteria imposed by each statute, and how the statutes complement each 
other. Under the approach articulated in this policy statement, the 
Agency believes that most plant-pesticides would not require regulation 
under FIFRA. However, the Agency believes some type of oversight is 
appropriate for plant-pesticides that are new to the plant and have a 
toxic mechanism of action (see Unit IV.B. of this document). Similarly 
under FFDCA, the Agency believes that most plant-pesticides should be 
exempt from the requirement of a tolerance. However, the Agency 
believes that EPA review should take place for certain plant-pesticides 
that are used in food/feed (see Unit IV.C. of this document).
    This statement of policy is one of several documents published in 
today's Federal Register that address EPA's regulation of plant-
pesticides. The other documents are: (1) a proposed regulatory 
amendment that would describe categories of plant-pesticides that are 
subject to or exempt from regulation under FIFRA and clarifies the 
status of plants that produce plant-pesticides (``Plant-pesticides 
Subject to the Federal Insecticide, Fungicide, and Rodenticide Act; 
Proposed Policy''); (2) a proposed exemption from the requirement of a 
tolerance under FFDCA for categories of plant-pesticides that do not 
result in significantly different dietary exposures (``Plant-
pesticides; Proposed Exemption from the Requirement of a Tolerance 
Under the Federal Food, Drug, and Cosmetic Act''); (3) a proposed 
exemption from the requirement of a tolerance under FFDCA for viral 
coat proteins (``Plant-pesticides; Proposed Exemption from the 
Requirement of a Tolerance Under the Federal Food, Drug, and Cosmetic 
Act for Viral Coat Proteins Produced in Plants''); and (4) a proposed 
exemption from the requirement of a tolerance under FFDCA for nucleic 
acids, including deoxyribonucleic and ribonucleic acids (``Plant-
pesticides; Proposed Exemption from the Requirement of a Tolerance 
Under the Federal Food, Drug, and Cosmetic Act for Nucleic Acids 
Produced in Plants'').

B. History

    Since 1987, EPA has sponsored, or co-sponsored with other Federal 
agencies, three conferences that discussed whether transgenic plants 
(plant varieties developed through new biotechnology methodologies) 
producing pesticidal substances pose potential risks and the nature of 
those risks. In addition, EPA has requested advice on how best to 
address plant-pesticides from two scientific advisory committees at 
three meetings. On December 18, 1992, a Subpanel of the FIFRA 
Scientific Advisory Panel (SAP) was convened to review a draft proposed 
policy statement and to answer a series of scientific questions 
concerned primarily with EPA's proposed approach for plant-pesticides 
under FIFRA. On July 13, 1993, a Subcommittee of the EPA Biotechnology 
Science Advisory Committee (BSAC) was convened to address a series of 
scientific questions concerned primarily with EPA's proposed approach 
for plant-pesticides under FFDCA. On January 21, 1994, a joint SAP/BSAC 
Subpanel was convened to address a series of scientific questions 
concerned with the scope of regulation under FIFRA and FFDCA and 
guidance for data needs for the evaluation of plant-pesticides. For 
more detailed discussion of the reports from the three advisory 
committee meetings, refer to Unit VIII. of this document.

II. Summary of Proposed Policy Under FIFRA and FFDCA

A. Introduction

    On June 2, 1982, EPA promulgated a final regulation under FIFRA 
section 25(b) that exempted all biological control agents, except for 
certain microorganisms, from the requirements of FIFRA (47 FR 23928; 
see 40 CFR 152.20). EPA defines the term ``biological control agent'' 
as ``any living organism applied to or introduced into the environment 
that is intended to function as a pesticide against another organism 
declared to be a pest by the Administrator'' (40 CFR 152.3). The 
exemption of biological control agents was promulgated because EPA 
found that the risks posed by biological control agents other than 
microorganisms were adequately addressed by other Federal agencies such 
as the U.S. Department of Agriculture's (USDA) Animal and Plant Health 
Inspection Service (APHIS) and the U.S. Department of the Interior. 
Although plants used as biological control agents were not specifically 
addressed in the June 2, 1982, Federal Register notice, they have been 
excluded from regulation under FIFRA through this exemption. EPA 
continues to believe that plants used as biological control agents are 
adequately regulated by other Federal agencies. However, EPA believes 
that the status of pesticidal substances produced in plants (i.e., 
plant-pesticides) requires regulatory clarification.
    Although plants used as biological control agents were excluded 
from FIFRA regulation under 40 CFR 152.20, substances that are 
extracted from plants and used as pesticides are not similarly 
excluded. For example, chrysanthemums produce pyrethrum, a substance 
that has insecticidal activity. The chrysanthemum plants that produce 
pyrethrum have been exempted from regulation when used as biological 
control agents (i.e., living chrysanthemums), but pyrethrum itself, as 
the pesticide substance, has not been exempted when extracted from 
chrysanthemums and applied to other plants as an insecticide.
    This distinction is reasonable in light of the potential for 
increased and unique exposures due to large-scale application of 
extracted pyrethrum to plants that do not naturally produce it. The use 
of extracted pyrethrum as an insecticide can involve exposure to the 
pesticide over large acreages, whereas the exposure associated with 
pyrethrum in living chrysanthemum plants would not be expected to reach 
such proportions. In addition, application of pyrethrum beyond the 
environment in which it is normally produced (i.e., beyond the living 
chrysanthemum plant) could result in new or unique exposures of 
nontarget organisms, including humans.
    Although it has been EPA's policy under FIFRA to regulate 
pesticidal substances extracted from plants, EPA has not, thus far, 
clearly stated its policies for regulation of pesticidal substances 
that are produced in living plants but not extracted from the plants 
(plant-pesticides). This policy statement, and the companion document 
published elsewhere in today's issue of the Federal Register entitled 
``Plant-pesticides Subject to the Federal Insecticide, Fungicide, and 
Rodenticide Act; Proposed Rule,'' is designed to provide such 
clarification for pesticidal substances that have evolved in plants, 
for pesticidal substances introduced into plants by breeding, and for 
pesticidal substances introduced into plants through biotechnology.
    Similarly under FFDCA, EPA has regulated substances that are 
extracted from plants and used as pesticides on food or feed. For 
example, a tolerance has been set for pyrethrum that is extracted from 
plants and applied to food or feed. However, the Agency has not clearly 
explained how pesticidal substances produced in plants (plant-
pesticides) would be regulated under FFDCA. For example, if a food 
plant could be modified, for pesticidal purposes, to produce pyrethrum, 
EPA has not, thus far, explained how this pyrethrum would be regulated 
under FFDCA.
    It is the intent of this policy statement to give guidance as to 
the types of plant-pesticides that would be evaluated by the Agency 
under FFDCA. The considerations used to determine whether EPA review 
would be required will be set forth in this policy statement and the 
companion Federal Register documents (``Plant-pesticides; Proposed 
Exemption from the Requirement of a Tolerance Under the Federal Food, 
Drug, and Cosmetic Act''; ``Plant-pesticides; Proposed Exemption from 
the Requirement of a Tolerance Under the Federal Food, Drug, and 
Cosmetic Act for Viral Coat Proteins Produced in Plants''; ``Plant-
pesticides; Proposed Exemption from the Requirement of a Tolerance 
Under the Federal Food, Drug, and Cosmetic Act for Nucleic Acids 
Produced in Plants'').

B. Proposed Regulatory Scheme

    In order to establish an effective regulatory scheme appropriate to 
plant-pesticides, EPA proposes to take the following actions under 
FIFRA and FFDCA. The Agency makes clear that the substances plants 
produce to protect themselves against pests and disease are pesticides 
under the FIFRA section 2 definition of ``pesticide,'' i.e., if they 
are ``. . .intended for preventing, destroying, repelling or mitigating 
any pest.'' Pesticidal substances that are produced in the living plant 
along with the genetic material necessary for the production of those 
substances are designated by EPA as plant-pesticides (Unit IV.B. of 
this document). The definition of pesticide under FIFRA section 2 also 
includes ``plant regulators.'' The Agency provides criteria for 
determining when a substance produced in a living plant but not 
extracted from the plant is a plant regulator (Unit IV.D. of this 
document), and the rationales EPA employed in developing these 
criteria.
    EPA indicates that it proposes to focus its regulatory attention on 
the plant-pesticide and not on the plant per se. The Agency defines the 
categories of plant-pesticides that it proposes to regulate under FIFRA 
and FFDCA. In general, the Agency would regulate, under FIFRA, those 
plant-pesticides that have the greatest potential for new environmental 
exposures and adverse effects to nontarget organisms. To do this, EPA 
proposes to exempt from FIFRA requirements, certain classes of plant-
pesticides based upon the source from which the plant-pesticide is 
derived and the mechanism of action of the pesticidal substance. Also 
contained in the proposed FIFRA exemption are coat proteins from plant 
viruses. Plant-pesticides that do not fall within these exemptions 
would be subject to FIFRA regulation. In a proposal published elsewhere 
in todays issue of the Federal Register a new part in 40 CFR, part 174, 
would establish the scope of regulation for plant-pesticides under 
FIFRA (``Plant-pesticides Subject to the Federal Insecticide, 
Fungicide, and Rodenticide Act; Proposed Rule'').
    Under FFDCA, the Agency would regulate those plant-pesticides that 
have the greatest potential for new dietary exposures. To establish the 
FFDCA scope of coverage, EPA proposes three exemptions from the 
requirements of a tolerance for three categories of plant-pesticides: 
(1) Certain plant-pesticides commonly found in food; (2) coat proteins 
from plant viruses; and (3) nucleic acids (see documents published 
elsewhere in todays issue of the Federal Register entitled, ``Plant-
pesticides; Proposed Exemption from the Requirement of a Tolerance 
Under the Federal Food, Drug, and Cosmetic Act''; ``Plant-pesticides; 
Proposed Exemption from the Requirement of a Tolerance Under the 
Federal Food, Drug, and Cosmetic Act for Viral Coat Proteins Produced 
in Plants''; and ``Plant-pesticides; Proposed Exemption from the 
Requirement of a Tolerance Under the Federal Food, Drug, and Cosmetic 
Act for Nucleic Acids Produced in Plants''). Plant-pesticides that do 
not fall within these exemptions would be subject to the FFDCA 
tolerance requirements.
    Recognizing the unique characteristics of plant-pesticides, the 
Agency is proposing to establish a new part 174, in 40 CFR under FIFRA 
for plant-pesticides (``Plant-pesticides Subject to the Federal 
Insecticide, Fungicide, and Rodenticide Act; Proposed Rule''). In the 
future EPA will propose, under part 174, procedural requirements for 
plant pesticides.
    In this proposed policy statement, EPA provides information on how 
manufacturers, importers, and distributors of plant-pesticides subject 
to FIFRA and FFDCA requirements should interact with the Agency. This 
guidance contains (1) Information on when and how manufacturers should 
first consult with the Agency; (2) a set of ``points to consider'' to 
assist manufacturers in developing data for review; (3) descriptions of 
proposed Agency procedures for Experimental Use Permits (EUPs) and 
registration; and (4) descriptions of EPA's interaction with other 
agencies.

III. Statutory and Regulatory Background

    This policy was developed under the authority of FIFRA, as amended 
(7 U.S.C. 136 et seq.) and FFDCA (21 U.S.C. 321 et seq.). Under FIFRA, 
a pesticide may not be sold or distributed in the United States unless 
it is registered, or has been exempted from regulation. Under FFDCA, 
EPA has the authority to set tolerances or establish an exemption from 
the requirement of a tolerance for pesticide residues in or on raw 
agricultural commodities and to establish food additive regulations for 
pesticide residues in or on processed foods.

A. FIFRA

    FIFRA section 2(u) defines ``pesticide'' as: ``(1) any substance or 
mixture of substances intended for preventing, destroying, repelling, 
or mitigating any pest, and (2) any substance or mixture of substances 
intended for use as a plant regulator, defoliator, or desiccant. . . 
.''
    FIFRA section 3 provides that no person may distribute or sell in 
the United States any pesticide that is not registered under the Act. 
Before a product may be registered as a pesticide under FIFRA, it must 
be shown that when used in accordance with widespread and commonly 
recognized practice, it will not generally cause ``unreasonable adverse 
effects on the environment.'' FIFRA section 2(bb) defines the term 
``unreasonable adverse effects on the environment'' as any unreasonable 
risk to ``man or the environment, taking into account the economic, 
social, and environmental costs and benefits of the use of any 
pesticide.'' Thus, FIFRA involves a balancing of the risks presented by 
the use of the pesticide against the benefits associated with the use 
of that pesticide.
    In addition to the requirement for a registration, FIFRA authorizes 
EPA to issue EUPs under section 5 and to otherwise regulate the use of 
unregistered pesticides under FIFRA section 3(a). Section 5 of FIFRA 
and 40 CFR part 172 provide for issuance by the Agency of Experimental 
Use Permits (EUP's) for the testing of new, unregistered pesticides or 
registered pesticides being tested for new uses in which the purpose is 
only to determine its value for pesticide purposes or to determine its 
toxicity or other properties. Such permits are generally issued for 
large-scale testing of pesticides on more than 10 cumulative acres of 
land or 1 surface acre of water. Contained within the scope of the 
regulation, however, is the presumption that small-scale testing, i.e., 
on not more than 10 cumulative acres of land or 1 surface acre of 
water, does not require an EUP provided that the crops are destroyed or 
an appropriate tolerance is in place (40 CFR 172.3(a)). This 
presumption, however, is caveated not to preclude experimental testing 
on larger areas in certain circumstances where the purpose of the large 
acreage test is only to determine the substance's value for pesticidal 
purposes or to determine its toxicity or other properties, and no 
benefit from pest control is expected (40 CFR 172.3(b)). In the Federal 
Register of January 22, 1993 (58 FR 5878), EPA issued a proposed 
amendment to 40 CFR part 172. The proposed amendment would, among other 
things, modify section 172.3 to clarify that the determination of 
whether an EUP is required is based on risk/benefit considerations. The 
amendment would provide that tests conducted on not more than 10 acres 
of land and 1 surface acre of water are presumed not to involve 
unreasonable risks, and therefore, do not require an EUP.
    FIFRA also authorizes EPA to require data to be submitted to 
evaluate whether an EUP or registration will be granted. Moreover, 
under FIFRA, EPA can impose labeling restrictions and FIFRA requires 
that the pesticide be used in accordance with such labeling 
restrictions. Under FIFRA section 25(b), EPA may exempt, by regulation, 
any pesticide determined to be: (1) Adequately regulated by another 
Federal agency, or (2) of a character which is unnecessary to be 
subject to the Act in order to carry out the purposes of the Act.

B. FFDCA

    The Reorganization Plan of 1970 that created EPA reallocated the 
authority under FFDCA to regulate pesticide residues in foods and 
animal feeds to EPA. Pursuant to section 402 of FFDCA, foods that are 
raw agricultural commodities are deemed to be adulterated if they 
contain a pesticide chemical which is unsafe within the meaning of 
section 408(c) of FFDCA. Under FFDCA section 408, any poisonous or 
deleterious pesticide chemical added to a raw agricultural commodity, 
that is not ``generally recognized as safe'' (GRAS), is deemed to be 
unsafe unless a tolerance, or an exemption from the requirement of a 
tolerance, for such pesticide chemical is established and the pesticide 
chemical residue is within the tolerance limits. Section 408 of FFDCA 
applies to all ``pesticide chemicals'' which are defined in section 
201(q) of FFDCA as:

    any substance which, alone, in chemical combination or in 
formulation with one or more other substance, is ``a pesticide'' 
within the meaning of [FIFRA] . . .and which is used in the 
production, storage, or transportation of raw agricultural 
commodities.

Thus, pesticide chemicals subject to regulation under FFDCA are defined 
by reference to the definition of pesticide under FIFRA.
    Section 408 of FFDCA authorizes EPA to set tolerances for pesticide 
chemical residues on raw agricultural commodities to the extent 
necessary to protect the public health. In establishing a tolerance, 
EPA must give appropriate consideration to the following factors: (1) 
The necessity for the production of an adequate, wholesome, and 
economical food supply; (2) the other ways in which the consumer may be 
affected by the same pesticide chemical or by other related substances 
that are poisonous or deleterious; and (3) the opinion submitted with a 
certification of usefulness under the Act (FFDCA section 408(b)). Thus, 
as with FIFRA, the regulatory decisions EPA makes under FFDCA section 
408 involve a risk/benefit balance. Unlike FIFRA, however, FFDCA only 
addresses dietary risks to humans and other animals.
    Under FFDCA section 408(c), EPA can exempt, by regulation, any 
pesticide chemical from the necessity of a tolerance when such 
tolerance is not necessary to protect the public health. In the absence 
of such an exemption, any pesticide chemical used on raw agricultural 
commodities is deemed unsafe unless EPA establishes a tolerance for the 
pesticide chemical residue, and the pesticide chemical residue is 
within the tolerance limits, or the pesticide chemical is GRAS.
    Under FFDCA section 402, food is deemed to be adulterated if it 
contains any food additive not authorized by a food additive regulation 
under section 409. Because of the ``flowthrough'' provision in section 
402(a)(2), EPA has interpreted section 409 as applying to pesticide 
residues in processed food which result from use of the pesticide in or 
on raw food if the concentration of the pesticide in the processed food 
is greater than the level set under section 408 for the raw food 
tolerance. If EPA grants an exemption from the requirement of a 
tolerance under section 408 for the raw food, residues in the resulting 
processed food are also exempt even if they are higher than in the raw 
food. Section 409 also applies to pesticide residues in processed food 
resulting from direct application of the pesticide to processed food. 
However, since the plant-pesticides that EPA is addressing would all be 
present in the plant which would be a raw agricultural commodity, this 
aspect of section 409 would not come into play. In issuing a food 
additive regulation under section 409, EPA must determine that the 
proposed use of the food additive, under the conditions of use 
specified in the regulation, will be safe. In EPA's view, the 
determination of whether use of a pesticidal food additive is safe 
should take into account the net effects of use of the additive on the 
food supply. These net effects include the benefit of an adequate, 
wholesome, and economical food supply that may result from a 
pesticide's use as well as any harm to the food supply that may result 
from the pesticide's use.
    A section 409 food additive regulation is not required for any 
substance that is GRAS. A GRAS finding must be based either on a record 
of safe use in food prior to 1958 (when Congress modified FFDCA) or 
evidence of safety and widespread agreement in the appropriate 
scientific community (FFDCA section 201(s)).

IV. Rationale and Regulatory Status of Plant-pesticides

A. Introduction

    As are all pesticides, all plant-pesticides are potentially subject 
to EPA's regulatory authority under FIFRA. Since FFDCA defines 
pesticides in terms of the definition in FIFRA section 2, EPA also has 
the authority to regulate residues of plant-pesticides under FFDCA 
sections 408 and 409 (Unit III.). Both FIFRA and FFDCA give EPA the 
authority to exempt pesticides from regulation through notice and 
comment rulemaking.
    EPA has attempted to identify, for regulatory oversight, those 
types of plant-pesticides that appear to have greater potential for 
environmental and/or human health risks. Through FIFRA section 25(b), 
EPA proposes to exempt certain categories of plant-pesticides that do 
not warrant oversight. Those plant-pesticides, or categories of plant-
pesticides, not exempted would form the scope of EPA's regulatory 
scrutiny under FIFRA.
    FIFRA section 25(b) allows the Agency to exempt a pesticide if it 
is of a character unnecessary to be subject to the Act in order to 
carry out the purposes of the Act. For plant-pesticides, the Agency 
proposes to amend 40 CFR part 152 and to create a new part 174 that 
would exempt, from regulation under FIFRA, certain categories of plant-
pesticides that pose low probability of risk and will not cause 
unreasonable adverse effects on the environment (See the proposal 
published elsewhere in todays issue of the Federal Register entitled 
``Plant-pesticides Subject to the Federal Insecticide, Fungicide, and 
Rodenticide Act; Proposed Rule'').
    Under FFDCA section 408(c), EPA can establish an exemption from the 
requirement of a tolerance for a pesticide chemical if the Agency 
determines that a tolerance is not necessary to protect the public 
health. The Agency is proposing to exempt, on that basis, certain 
categories of plant-pesticides from the requirement from a tolerance 
(described in documents published elsewhere in todays issue of the 
Federal Register entitled, ``Plant-pesticides; Proposed Exemption from 
the Requirement of a Tolerance Under the Federal Food, Drug, and 
Cosmetic Act''; ``Plant-pesticides; Proposed Exemption from the 
Requirement of a Tolerance Under the Federal Food, Drug, and Cosmetic 
Act for Viral Coat Proteins Produced in Plants''; ``Plant-pesticides; 
Proposed Exemption from the Requirement of a Tolerance Under the 
Federal Food, Drug, and Cosmetic Act for Nucleic Acids Produced in 
Plants''). Those plant-pesticides not exempt would be subject to EPA 
review under the FFDCA authorities.
    The following unit of this statement describes the environmental 
and human health considerations that the Agency weighed in determining 
which plant-pesticides to propose for exemption from regulation under 
FIFRA and FFDCA. Those plant-pesticides that would not be exempt would 
be subject to regulation.

B. FIFRA

    1. Summary of proposed regulatory status for plant-pesticides. The 
following unit summarizes the proposed regulatory status of plant-
pesticides under FIFRA. Because of the unique nature of plant-
pesticides, the Agency is proposing regulatory definitions that would 
apply to plant-pesticides only. In addition, the Agency is proposing to 
exempt certain classes of plant-pesticides from regulation under FIFRA 
because the Agency believes that they pose low probability of risk and 
will not cause unreasonable adverse affects (see the proposal published 
elsewhere in todays issue of the Federal Register entitled, ``Plant-
pesticides Subject to the Federal Insecticide, Fungicide, and 
Rodenticide Act; Proposed Rule'').
    a. Definition of plant-pesticide. EPA is proposing to define 
``plant-pesticide'' under FIFRA as:

    A pesticidal substance that is produced in a living plant and 
the genetic material necessary for the production of the substance, 
where the substance is intended for use in the living plant.

    EPA is including the genetic material necessary to produce the 
substance in the proposed definition of plant-pesticide for a number of 
reasons. First, it is the genetic material that is introduced into the 
plant with the intent that it will ultimately result in a pesticidal 
effect. Additionally, EPA's regulation of pesticides is based on an 
evaluation of the potential for unreasonable adverse effects to humans 
and the environment associated with the use of the pesticidal 
substance, in this case, the pesticidal substance produced in the 
plant. Regulation also includes risk management considerations. A focus 
on the genetic material would permit the Agency to address the 
potential for the spread of the pesticidal substance in the environment 
through the spread of the genetic material necessary for the production 
of the substance. Moreover, the amount of pesticidal substance likely 
to be produced by the plant is also an important consideration that the 
Agency may, in some circumstances, be able to address through the 
inclusion of genetic material in the definition of plant-pesticide. In 
addition, including the genetic material in the definition of plant-
pesticide would permit the Agency to address plant-pesticides during 
stages of the plant's life cycle or in plant parts where the pesticidal 
substance itself is not produced or is produced in very small amounts 
(e.g., in pollen or seed). In these cases, it is technically easier to 
verify the presence of the genetic material than the pesticidal 
substance.
    b. Active and inert ingredients. The regulation of pesticides under 
FIFRA entails the identification of ``active ingredients'' and ``inert 
ingredients.'' Under FIFRA section 2, the term active ingredient means 
``. . .an ingredient which will prevent, destroy, repel, or mitigate 
any pest. . . [or acts as a plant regulator, defoliant or desiccant].'' 
The term inert ingredient means ``. . .an ingredient which is not 
active.'' EPA recognizes that plant-pesticides have certain 
characteristics that are different from those of more traditional 
chemical pesticides. EPA believes that the overall characteristics of 
plant-pesticides require specifically tailored active and inert 
ingredient definitions.
    In light of this consideration, EPA proposes to use the following 
definitions for active and inert ingredients for plant-pesticides.
    ``Active ingredient,'' when referring to plant-pesticides only, 
means a pesticidal substance that is produced in a living plant and the 
genetic material necessary for the production of the substance, where 
the substance is intended for use in the living plant.
    ``Inert ingredient,'' when referring to plant-pesticides only, 
means any substance, such as a selectable marker, other than the active 
ingredient, and the genetic material necessary for the production of 
the substance, that is intentionally introduced into a living plant 
along with the active ingredient, where the substance is used to 
confirm or ensure the presence of the active ingredient.
    Note that the plant-pesticide active ingredient is the plant-
pesticide and therefore the proposed definition of active ingredient 
for plant-pesticides is the same as the definition of plant-pesticide. 
The plant-pesticide product includes both the active and inert 
ingredients.
    The definition of plant-pesticide and the active and inert 
ingredient definitions would include all of the genetic material 
``necessary for the production'' of the pesticidal and inert substance. 
The following genetic regions are considered ``necessary for the 
production'' of the plant-pesticide, active and inert substances: (1) 
The genetic material that encodes for a pesticidal substance or leads 
to the production of a pesticidal substance and (2) regulatory regions 
such as promoters, enhancers, and terminators.
    The genetic material can either directly encode for the pesticidal 
substance or may encode for enzymes that lead to the production of a 
pesticidal substance (e.g., phenylalanine ammonia-lyase (PAL) catalyzes 
the first reaction in the synthesis of such phytoalexins as 
pterocarpans in Leguminosae and furanocoumarins in Solanaceae and 
Umbelliferae; Ref. 6). It might also include genetic regions encoding 
for RNA that acts as the pesticidal substance or leads to the 
production of the pesticidal substance (e.g., antisense mRNA). The 
active and inert ingredients would also include any regulatory regions, 
such as promoters, that control the expression of the genetic material 
encoding for the pesticidal or inert substance or leading to the 
production of the pesticidal or inert substance and are introduced into 
the plant along with that gene.
    The genetic material ``necessary for the production'' of the plant-
pesticide, active and inert substances does not include genetic regions 
that are not involved in DNA expression (i.e., noncoding, nonexpressed 
sequences such as linkers, adapters, homopolymers and sequences of 
restriction enzyme recognition sites). However, the Agency would 
require information concerning these sequences if it determines that 
such information is necessary for the evaluation of the active or inert 
ingredient.
    There may be genetic material encoding other functions (e.g., 
genetic material intended to alter the amount of carbohydrate in the 
plant) that are introduced into the plant along with the active and 
inert ingredients. These functions would be subject to Food and Drug 
Administration (FDA) authorities.
    c. Exemptions under FIFRA. EPA has attempted to identify those 
types of plant-pesticides that have greater potential for environmental 
and/or human health risks and to focus its regulatory scrutiny on these 
plant-pesticides. To exempt from regulation those plant-pesticides 
having less potential for risk, EPA is proposing to employ its 
exemption authority under FIFRA section 25(b). FIFRA section 25(b)(2) 
allows the Agency to exempt a pesticide from FIFRA regulation if it is 
of a character unnecessary to be subject to the Act in order to carry 
out the purposes of the Act. Through FIFRA section 25(b)(2), EPA 
proposes to exempt, from FIFRA regulation, certain categories of plant-
pesticides that EPA believes pose low probability of risk and are not 
likely to cause unreasonable adverse effects even in the absence of any 
regulatory oversight under FIFRA and, thus, are of a character 
unnecessary to be subject to the Act. Those plant-pesticides not 
exempted would form the scope of EPA's regulatory scrutiny under FIFRA.
    EPA finds that the plant-pesticides it is proposing to exempt have 
a low probability of risk and have potential benefits associated with 
them (e.g., economic benefit to farmers and reducing the need for 
chemical pesticides) that outweigh any potential risks associated with 
them, and that the low probability of risk does not justify the cost of 
regulation. For a detailed description and analysis of the proposed 
exemptions under FIFRA, see the proposal published elsewhere in todays 
issue of the Federal Register entitled ``Plant-pesticides Subject to 
the Federal Insecticide Fungicide and Rodenticide Act; Proposed Rule.''
    (i) Exemption of plant-pesticides derived from closely related 
plants. The Agency is proposing to concentrate its regulatory efforts 
under FIFRA on those plant-pesticides that are new to the plant and, 
thus, have the greatest potential for exposing nontarget organisms to a 
new pesticidal substance. The Agency is proposing to exempt from FIFRA 
regulation those plant-pesticides that are normally a component of (not 
new to) the plant. The approach EPA is proposing to use to capture the 
concept of ``normally a component'' is based on the concept of sexual 
compatibility. The standard of sexual compatibility is embodied in the 
following language from the proposed regulatory text:

    [Plant-pesticides are exempt from FIFRA requirements if:]
    . . .The genetic material that encodes for a pesticidal 
substance or leads to the production of a pesticidal substance is 
derived from plants that are sexually compatible with the recipient 
plant and has never been derived from a source that is not sexually 
compatible with the recipient plant;. . .

    Key definitions associated with this language are:
    ``Bridging crosses'' between plants means the utilization of an 
intermediate plant in a cross to produce a viable zygote between the 
intermediate plant and a first plant, in order to cross the plant 
resulting from that zygote with a third plant that would not otherwise 
be able to produce viable zygotes from the fusion of its gametes with 
those of the first plant. The result of the bridging cross is the 
mixing of genetic material of the first and third plant through the 
formation of an intermediate zygote.
    ``Genetic material that encodes for a pesticidal substance or leads 
to the production of a pesticidal substance'' does not include 
regulatory regions or noncoding, nonexpressed nucleotide sequences.
    ``Regulatory region'' means genetic material that controls the 
expression of the genetic material that encodes for a pesticidal 
substance or leads to the production of a pesticidal substance. 
Examples of regulatory regions include promoters, enhancers, and 
terminators.
    ``Sexually compatible,'' when referring to plants, means capable of 
forming a viable zygote through the fusion of two gametes, including 
the use of bridging crosses or wide crosses between plants.
    ``Source'' means the donor of the genetic material that encodes for 
a pesticidal substance or leads to the production of a pesticidal 
substance.
    ``Wide crosses'' between plants means to facilitate the formation 
of viable zygotes through the use of surgical alteration of the plant 
pistil, bud pollination, mentor pollen, immunosuppressants, in vitro 
fertilization, pre- and post- pollination hormone treatments, 
manipulation of chromosome numbers, embryo culture, or ovary and ovule 
cultures or any other technique that the Administrator determines meets 
this definition.
    EPA is also proposing for discussion two alternative options for 
describing this category of plant-pesticides in plants (see Unit 
IV.B.2. of this document and the proposal published elsewhere in 
today's issue of the Federal Register entitled ``Plant-pesticides 
Subject to the Federal Insecticide, Fungicide, and Rodenticide Act; 
Proposed Rule'').
    (ii) Exemption of plant-pesticides that act primarily by affecting 
the plant. One of EPA's primary goals in regulating pesticides is to 
control the potential for adverse effects of pesticides on nontarget 
organisms. An important component in the evaluation of this potential 
is the way in which the pesticidal substance acts on the target pest 
since it would also likely affect nontarget organisms through the same 
mechanism. Based on this rationale, the Agency is proposing to exempt 
from FIFRA regulation plant-pesticides that are not directly toxic to 
the target pest. This proposed exemption is embodied in the following 
language from the proposed regulatory text:
[Plant-pesticides are exempt from FIFRA requirements if:]
     . . .The pesticidal substance acts primarily by affecting the 
plant so that the target pest is inhibited from attaching to the 
plant, penetrating the plant, or invading the plant's tissue in at 
least one of the following ways:
    (1) The pesticidal substance acts as a structural barrier to 
attachment of the pest to a host plant, a structural barrier to 
penetration of the pest into a host plant, or a structural barrier 
to spread of the pest in a host plant, for example, through the 
production of wax or lignin, or length of trichomes (plant hairs); 
or
    (2) The pesticidal substance acts in the host plant to 
inactivate or resist toxins or other disease-causing substances 
produced by the target pest; or
    (3) The pesticidal substance acts by creating a deficiency of a 
plant nutrient or chemical component essential for pest growth on/in 
the host plant.

    (iii) Exemption of coat proteins from plant viruses. Coat proteins 
are those substances that viruses produce to encapsulate and protect 
their genetic material. When the genetic material encoding the coat 
protein from a plant virus is introduced into a plant's genome, the 
plant is able to resist infections by the virus (termed viral coat 
protein mediated resistance or vcp-mediated resistance). The Agency 
proposes to exempt the genetic material encoding the coat protein and 
the coat protein itself when these are introduced into a plant to 
effectuate vcp-mediated resistance. This proposed exemption is embodied 
in the following regulatory text:
[Plant-pesticides are exempt from FIFRA requirements if:] . . .The 
pesticidal substance is a coat protein from a plant virus. . .
    EPA is also proposing for discussion an alternative, more 
restrictive exemption for coat proteins from plant viruses used in 
viral coat protein mediated resistance (see Unit IV.B.2. of this 
document and the proposal published elsewhere in todays issue of the 
Federal Register entitled ``Plant-pesticides Subject to the Federal 
Insecticide, Fungicide, and Rodenticide Act; Proposed Rule'').
    2. Rationale for Proposed Exemptions under FIFRA. As with 
traditional pesticides, the underlying considerations in analyzing 
risks posed by plant-pesticides are the potential for exposure to the 
pesticidal substance and hazards of the pesticidal substance to humans 
and other nontarget organisms. For plant pesticides, exposure and 
hazard will be determined by the chemical and toxicological properties 
of the pesticidal substance and the biological characteristics of the 
plant that is producing the substance.
    The properties of the plant-pesticide, including the mechanism by 
which it affects the target pest, will determine the potential for 
hazards to nontarget organisms. The type of organism exposed to the 
plant-pesticide will be determined by the characteristic of the plants 
that produce the substance and the environment where the plants are 
grown; e.g., whether the production of the substance is limited to 
particular plant parts, the organisms that normally associate with the 
plant, and the acreage and location planted. An important consideration 
not seen with traditional pesticides is the potential for spread of the 
plant's genetic material. Because plants can reproduce sexually and/or 
asexually, unintentional exposure to the plant-pesticide could occur in 
both the agro- or natural ecosystems, particularly if wild relatives 
acquire the ability to produce the plant-pesticide through successful 
hybridization.
    Such hazard and exposure considerations form the bases of the three 
exemptions that the Agency is proposing for plant-pesticides under 
FIFRA (see the proposal published elsewhere in todays issue of the 
Federal Register entitled ``Plant-pesticides Subject to the Federal 
Insecticide, Fungicide, and Rodenticide Act; Proposed Rule'' for more 
detailed analyses of these exemptions).
    The benefits associated with use of some categories of plant-
pesticides include the economic benefit to farmers for use of plant-
pesticides in circumstances where traditional pesticides may not be as 
effective (e.g., for some systemic plant pests) or may be more 
expensive, thus increasing crop yield and/or reducing farmers' costs. 
An additional benefit is the environmental benefit associated with 
potential reduced use of pesticides that may be less environmentally 
benign than these plant-pesticides.
    a. Exemption of plant-pesticides derived from closely related 
plants. A primary consideration in evaluating plant-pesticides is the 
potential for new exposures of nontarget organisms to the pesticide. If 
a plant normally produces a pesticidal substance, organisms that 
normally come in contact with the plant have likely been exposed to 
that substance in the past, perhaps over long periods of time. No new 
exposures would be expected to occur.
    In contrast, if a plant-pesticide is new to a plant, the organisms 
that come in contact with the plant may never have been exposed to the 
substance. For instance, certain spiders produce a toxin that is 
targeted for their insect prey. Plants are not known to produce this 
toxin in nature nor in cultivation. If this toxin were to enter the 
gene pool of specific plants, organisms that had never previously been 
exposed to the toxin could now be exposed. Prior to the introduction of 
the toxin into these plants, only the insect prey of the spider would 
potentially be exposed to the toxin. If plants could now express the 
toxin, a different or larger group of organisms could be exposed to it, 
possibly resulting in adverse effects to these organisms. For instance, 
insects or animals that feed on the plant could be exposed to the 
toxin. If the toxin is found in pollen, pollinators could also be 
exposed.
    EPA proposes to concentrate its regulatory efforts under FIFRA on 
those plant-pesticides that are new to the plant and thus have the 
greatest potential for exposing nontarget organisms to a new pesticidal 
substance. The Agency would consider plant-pesticides produced in 
sexually compatible plants to be least likely to result in these new 
exposures (see the proposal published elsewhere in todays issue of the 
Federal Register entitled ``Plant-pesticides Subject to the Federal 
Insecticide, Fungicide, and Rodenticide Act; Proposed Rule''). Sexually 
compatible plants are more apt to share common traits than are 
unrelated plants. It is a common expectation that similarity is 
associated with the degree of relatedness. Natural hybridization and 
selection have produced groups of plants which have a common gene pool. 
Generations of artificial hybridization practiced to produce improved 
crops for cultivation have tended to increase the extent of relatedness 
among elements of a broader segment of agricultural plants. Since 
traits can be passed through a plant population by sexual 
recombination, it is reasonable to predict that, in a sexually 
compatible population, new exposures of organisms that associate with 
plants in the population to the pesticidal substance is unlikely.
    The practice of saving seed from desirable plants has been going on 
for thousands of years and controlled crosses to produce plant hybrids 
have been documented since the eighteenth century. Since the 
rediscovery of Mendel's work on the inheritance of traits, there is a 
base of experience of 50 to 100 years of breeding for most major crops. 
During that time, it has been common agricultural practice to cross 
sexually compatible wild relatives with crop plants to develop crop 
varieties with better pest resistance. Techniques such as genetic 
mapping reveal the presence of genetic loci in cultivated plants that 
previously were considered to be present only in the wild species. 
Sexually compatible crop varieties are also crossed with each other to 
achieve better pest resistance in their progeny. Because of these 
common practices, the potential for significantly different 
environmental exposures is likely to be low.
    EPA proposes to extend the concept of sexual compatibility to 
include wide crosses because wide crosses are commonly used to expand 
the gene pool for varietal improvement. EPA believes that the fact that 
a wide cross produces a viable zygote indicates a fairly high degree of 
relatedness between the parental plants. However, for regulatory 
purposes it is somewhat difficult to define what constitutes a wide 
cross in a definitive way since techniques may change over time. EPA is 
thus proposing to define, for the purposes of this rulemaking, wide 
crosses based on existing techniques with the provision that new 
techniques can be added if they meet the definition.
    A second approach that EPA is considering for defining when a 
plant-pesticide is new to the plant is a standard based on taxonomy. 
Under this approach, the standard would rely on the taxonomic grouping 
of genus; plant-pesticides moved between plants in the same genus would 
be exempt. The assumption under this approach is that the genus 
grouping correlates with a relatively high degree of relatedness among 
plants even though not all plants in a genus are sexually compatible.
    A third approach EPA is considering combines the above two 
standards of taxonomy and sexual compatibility. The standard under this 
option would rely primarily on the taxonomic grouping of genus as a 
measure of relatedness. Recognizing that some plants that are sexually 
compatible are classified in different genera and assuming that sexual 
compatibility is correlated with a high degree of relatedness, EPA also 
includes a provision extending the exemption to include plant-
pesticides moved between sexually compatible plants even if the plants 
are classified in different taxonomic genera.
    For all of the approaches presented in this unit, the Agency has 
evaluated whether changes in the levels of plant-pesticides that plants 
normally produce would warrant regulation under FIFRA. (Ref. 1 and 
Federal Register document entitled ``Plant-pesticides Subject to the 
Federal Insecticide Fungicide and Rodenticide Act; Proposed Rule'' for 
a more thorough analysis of this issue.) The Agency's analysis 
indicates that changes in the levels of such plant-pesticides expressed 
by a plant could result in increased or decreased exposures of 
nontarget organisms to a plant-pesticide. However, EPA believes, for 
the reasons outlined below, that the potential for unreasonable adverse 
effects from these exposures is low and these types of plant-pesticides 
do not warrant regulation under FIFRA.
    In deciding whether and how to regulate such plant-pesticides, EPA 
first considered whether an increase in the levels of such plant-
pesticides is likely to exceed the ranges normally found within and 
between plant varieties (both cultivated and uncultivated). EPA 
believes that increases in the levels of such plant-pesticides are not 
likely to result in overall significantly different exposures of 
nontarget organisms to the pesticide. The level of production of 
pesticidal substances normally produced by plants varies among related 
plants because of differences in genetic makeup and environmental 
conditions. EPA also considered the extent to which any substance can 
be increased in cultivated plants without unwanted effects on other, 
desirable characteristics of the plant (e.g., yield or palatability of 
fruit). In general, breeders balance all of these characteristics in 
developing marketable plant varieties. Considerations of 
characteristics such as yield could serve to mitigate against exceeding 
certain ranges of pesticide levels. EPA anticipates that the majority 
of plants with modified levels of plant-pesticides will fall within 
existing ranges of pesticide levels and does not anticipate that 
increasing the level of a plant-pesticide that is normally a component 
of a plant would lead to significantly different spectrum of exposure 
to the plant-pesticide.
    b. Exemption of plant-pesticides that act primarily by affecting 
the plant. As discussed previously, an important component in 
evaluating the potential for adverse affects on nontarget organisms is 
the way in which the pesticidal substance acts on the target pest. A 
pesticidal substance that acts directly on the target pest through a 
toxic mechanism of action might also exert a similar effect on other 
organisms. For example, a substance that acts by inhibiting DNA 
synthesis of the pest could inhibit DNA synthesis in other nontarget 
organisms.Toxic mechanisms of action include, but are not limited to, 
those that affect: (i) membrane permeability, (ii) cell division, (iii) 
gene expression, (iv) DNA replication, or (v) other metabolic functions 
(Ref. 4).
    Pesticidal substances can also act through mechanisms that are less 
likely to be directly toxic. Although it is possible for these 
substances to adversely affect nontarget organisms, the Agency believes 
that, in most cases, they pose significantly lower levels of 
environmental risk than plant-pesticides with a generalized toxic 
mechanism of action. For example, if a plant is modified so that it can 
counter specific disease-producing compounds by inactivating them, it 
is less likely that organisms that interact with the plant in other, 
more beneficial ways will be affected. Similarly, a plant may produce 
defense structures such as layers of cork cells in response to 
microbial infections. These structures form a barrier to further 
penetration by the pests and may block the spread of any toxins. Those 
organisms that do not stimulate this response are not likely to be 
adversely affected.
    Plant-pesticides that are less directly toxic generally act 
primarily by affecting the plant so that the pest is inhibited from 
attaching to the plant, penetrating the plant's surface, or invading 
the plant's tissue. The Agency believes that it would be appropriate to 
exempt from regulation, under FIFRA, plant-pesticides that act through 
mechanisms such as these. (See the proposal published elsewhere in 
todays issue of the Federal Register entitled ``Plant-pesticides 
Subject to the Federal Insecticide, Fungicide, and Rodenticide Act; 
Proposed Rule'' for a more detailed discussion of this exemption.) EPA 
believes that by focusing its regulatory attention on plant-pesticides 
that act through toxic mechanisms, it will be able to focus on those 
plant-pesticides presenting higher levels of risk potential.
    c. Exemption of coat proteins from plant viruses. The Agency is 
proposing to exempt the genetic material encoding the viral coat 
protein and the coat protein itself when these are introduced into a 
plant to effectuate viral coat protein mediated resistance. A more 
detailed discussion of the Agency's assessment of the risks and 
benefits of viral coat proteins can be found in the Federal Register 
document entitled, ``Plant-pesticides Subject to the Federal 
Insecticide, Fungicide, and Rodenticide Act; Proposed Rule.'' Major 
points of the analysis are summarized here.
    The Agency's proposal is made in light of a number of 
considerations which, when taken together, bring EPA to the conclusion 
that coat proteins from plant viruses generally pose a low probability 
of risk and would not pose unreasonable adverse effects even in the 
absence of any regulation under FIFRA. These considerations include the 
low potential for adverse affects to nontarget organisms and the 
potential benefits of utilizing vcp-mediated resistance.
    Environmental benefits associated with the use of viral coat 
proteins include the reduction of the use of chemical pesticides for 
viruses that are spread by vectors (usually insects). Chemical 
pesticides are used for those crop plants where the most effective 
method of protection against viral attack is by controlling the vector. 
These pesticides may not be environmentally benign. The expression of 
viral coat proteins by plants for protection from viral infection would 
likely reduce the amount of chemical pesticide used to control the 
vectors.
    In addition to environmental benefits associated with the use of 
viral coat proteins, an effective method for controlling virus 
infection will have economic benefits. Plant viruses create economic 
losses for a vast variety of crops by reducing yields and negatively 
affecting the quality of the crop. Yield losses and quality effects for 
a specific crop may vary depending on the host plant and strains of the 
virus present, the incidence and activity of vectors, timing of the 
infection, health and nutritional state of the plant, and weather (Ref. 
8).
    Presently, growers may need to use several control methods during a 
crop season in an attempt to prevent viral infection and dissemination, 
primarily by planting virus free material (for mechanically transmitted 
viruses) and by controlling plant virus vectors, such as insect 
populations (for vector transmitted viruses). Insecticides, 
nematicides, and fungicides are all used for vector control with 
varying success, depending upon the virus/vector relationship and 
vector efficiency. Plants developed through conventional breeding 
techniques offer some degree of virus resistance. Such resistance may 
not be uniform or the virus may develop new strains. However, breeding 
for resistance has not been successful for the majority of field crops 
and, in particular, vegetable crops that are severely affected by 
viruses (Ref. 8).
    In enabling plants to resist viral attack, viral coat proteins act 
in a very specific fashion, apparently adversely affecting only viruses 
by blocking or limiting their ability to infect, replicate, and/or 
translocate within the plant. This specificity minimizes the potential 
for viral coat proteins produced in plants to adversely affect nonviral 
organisms. In addition, plants in nature and in the agro-ecosystem 
frequently exhibit viral infections; nontarget organisms, including 
humans, have been and continue to be exposed to the viral coat proteins 
with no observed adverse effects.
    The possibility that environmental risk might be associated with 
the use of vcp-mediated resistance was discussed at the December 18, 
1992, FIFRA SAP Subpanel meeting. EPA agrees with the conclusions of 
the SAP Subpanel and in developing its proposal has utilized the advice 
of the Subpanel to supplement EPA's own evaluation of the scientific 
literature (Ref. 2). The considerations discussed at this meeting 
included: (1) The potential for new viruses to be formed through 
transcapsidation (also called heterologous encapsidation) and 
recombination; (2) the potential for synergistic infections; (3) the 
potential for seed transmission; and (4) the potential for the 
development of selective advantage in wild relatives through successful 
hybridization with the plant producing the viral coat protein. (See 
Unit VIII. of this document for a more detailed discussion of the SAP 
report and the proposal published elsewhere in todays issue of the 
Federal Register entitled ``Plant-pesticides Subject to the Federal 
Insecticide, Fungicide, and Rodenticide Act; Proposed Rule'' for 
discussion of the Agency's preferred proposed exemption and the 
alternative approach to viral coat proteins).
    With regard to selective advantage, the Subpanel report noted that 
while the series of events that must occur for the wild plant to 
acquire a selective advantage from vcp-mediated resistance coat 
proteins is rather improbable, such a series of events is not 
impossible. An alternative option presented by the Agency offers a more 
limited exemption of vcp-mediated resistance coat proteins to address 
the possibility that plants acquiring the vcp-mediated resistance genes 
might also acquire a selective advantage. With regard to the 
alternative option, the Agency has defined a set of criteria that would 
be used to identify those viral coat protein/plant combinations that 
have the greatest potential for outcrossing to wild, free living 
relatives and thus have the possibility endowing these wild relatives 
with a competitive advantage. Viral coat proteins that potentially 
could be outcrossed to wild relatives would be subject to regulation 
while those viral coat protein/plant combinations with a lesser or no 
probability of outcrossing would be exempt from regulation. The 
language covering this alternative is as follows.

    Coat proteins from plant viruses [would be exempt] if the 
genetic material necessary to produce a coat protein is introduced 
into a plant's genome and the plant has at least one of the 
following characteristics:
    (1) The plant has no wild relatives in the United States with 
which it can successfully exchange genetic material, i.e., corn, 
tomato, potato, soybean, or any other plant species that EPA has 
determined has no sexually compatible wild relatives in the United 
States.
    (2) It has been demonstrated to EPA that the plant is incapable 
of successful genetic exchange with any existing wild relatives 
(e.g., through male sterility, self-pollination).
    (3) If the plant can successfully exchange genetic material with 
wild relatives, it has been empirically demonstrated to EPA that 
existing wild relatives are resistant or tolerant to the virus from 
which the coat protein is derived or that no selective pressure is 
exerted by the virus in natural populations.
    For the purposes of this option, ``introduced into the plant's 
genome'' would mean movement of nucleotide sequences into the genetic 
material in a plant cell's nucleus, mitochondria, chloroplasts and any 
other plastids. ``Successful genetic exchange'' would mean capable of 
forming zygotes viable in the laboratory and/or field through the 
fusion of two gametes.

C. FFDCA

    1. Summary of regulatory status. As indicated previously, the 
Agency has available, under FFDCA section 408(c), the authority to 
exempt plant-pesticides from the requirement of a tolerance if such 
tolerance is not necessary to protect the public health (Unit III). The 
Agency is proposing that such a finding is appropriate for three 
classes of plant-pesticides: (1) Categories of pesticidal substances 
produced in plants that do not result in new dietary exposures based 
upon the source from which the pesticidal substance is derived; (2) 
nucleic acids produced in plants as part of a plant-pesticide; and (3) 
coat proteins from plant viruses when they are produced in plants. (For 
more detail on these proposed exemptions refer to the following Federal 
Register documents: ``Plant-pesticides; Proposed Exemption from the 
Requirement of a Tolerance Under the Federal Food, Drug, and Cosmetic 
Act''; ``Plant-pesticides; Proposed Exemption from the Requirement of a 
Tolerance Under the Federal Food, Drug, and Cosmetic Act for Viral Coat 
Proteins Produced in Plants''; ``Plant-pesticides; Proposed Exemption 
from the Requirement of a Tolerance Under the Federal Food, Drug, and 
Cosmetic Act for Nucleic Acids Produced in Plants'').
    a. Exemption from the requirement of a tolerance for categories of 
plant-pesticides that would not result in new dietary exposures. Under 
this exemption, the Agency is proposing to exempt from the requirement 
of a tolerance two categories of plant-pesticides: (1) Plant-pesticides 
produced in food plants and derived from closely related food or non-
food plants and (2) plant-pesticides produced in food plants and 
derived from food plants that are not closely related to the recipient 
food plant and would not result in significantly different dietary 
exposures when produced in the recipient food plant. The proposed 
exemption is as follows:

    (a) Residues of pesticidal substances produced in living plants 
as plant-pesticides are exempt from the requirement of a tolerance 
if the genetic material that encodes for a pesticidal substance or 
leads to the production of a pesticidal substance is derived from 
plants that are sexually compatible with the recipient plant and has 
never been derived from a source that is not sexually compatible 
with the recipient plant;
    (b) Residues of pesticidal substances produced in living plants 
as plant-pesticides are exempt from the requirement of a tolerance 
when the genetic material that encodes for a pesticidal substance or 
leads to the production of a pesticidal substance is derived from 
plants that are not sexually compatible with the recipient plants 
if:
    (1) The genetic material that encodes for a pesticidal substance 
or leads to the production of a pesticidal substance is derived from 
food plants; and
    (2) The pesticidal substances would not result in significantly 
different dietary exposures.
    For the purposes of this exemption, the following definitions 
apply:
    ``Bridging crosses'' between plants means the utilization of an 
intermediate plant in a cross to produce a viable zygote between the 
intermediate plant and a first plant, in order to cross the plant 
resulting from that zygote with a third plant that would not otherwise 
be able to produce viable zygotes from the fusion of its gametes with 
those of the first plant. The result of the bridging cross is the 
mixing of genetic material of the first and third plant through the 
formation of an intermediate zygote.
    ``Food plant'' means a plant which, either in part or in toto is 
used as food by humans.
    ``Genetic material that encodes for a pesticidal substance or leads 
to the production of a pesticidal substance'' does not include 
regulatory regions or noncoding, nonexpressed nucleotide sequences.
    ``Living plant'' means a plant that is alive, including periods of 
dormancy, and all viable plant parts/organs involved in the plant's 
life cycle.
    ``Major crops for human dietary consumption'' means wheat, corn, 
soybeans, potatoes, oranges, tomatoes, grapes, apples, peanuts, rice, 
beans, and any other crops that the Agency has determined is a major 
crop for human dietary consumption.
    ``Noncoding, nonexpressed nucleotide sequences'' means the 
nucleotide sequences are not transcribed and are not involved in gene 
expression. Examples of noncoding, nonexpressed nucleotide sequences 
include linkers, adapters, homopolymers, and sequences of restriction 
enzyme recognition sites.
    ``Recipient plant'' means the plant into which the plant-pesticide 
is introduced and in which the plant-pesticide is produced.
    ``Regulatory region'' means genetic material that controls the 
expression of the genetic material that encodes for a pesticidal 
substance or leads to the production of a pesticidal substance. 
Examples of regulatory regions include promoters, enhancers, and 
terminators.
    ``Result in significantly different dietary exposure'' means:
    (1) The pesticidal substance is produced in inedible portions of 
the source food plant, but, in the recipient plant, the pesticidal 
substance is present in the plant's edible portions.
    (2) The pesticidal substance is produced in the immature, but not 
in the mature, edible portions of the source food plant, but, in the 
recipient plant, the pesticidal substance is present in the mature, 
edible portions.
    (3) The pesticidal substance is from a source food plant normally 
cooked or processed prior to consumption and is produced in a recipient 
plant that is not normally cooked or processed prior to consumption.
    (4) The pesticidal substance is derived from a source food plant 
that is not a major crop for human dietary consumption and is 
introduced into a recipient plant that is a major crop for human 
dietary consumption.
    ``Sexually compatible,'' when referring to plants, means capable of 
forming a viable zygote through the fusion of two gametes, including 
the use of bridging crosses or wide crosses between plants.
    ``Source food plant'' means the donor of the genetic material that 
encodes for a pesticidal substance or leads to the production of a 
pesticidal substance.
    ``Wide crosses,'' between plants, means to facilitate the formation 
of viable zygotes through the use of surgical alteration of the plant 
pistil, bud pollination, mentor pollen, immunosuppressants, in vitro 
fertilization, pre- and post-pollination hormone treatments, 
manipulation of chromosome numbers, embryo culture, or ovary and ovule 
cultures, or any other technique that the Administrator determines 
meets this definition.
    The Agency is also proposing an alternative approach that uses a 
standard that relies primarily on the taxonomic grouping of genus as a 
measure of relatedness between plants (for more discussion of this 
alternative approach, see Unit IV.C.2. of this document and the 
proposal published elsewhere in todays issue of the Federal Register 
entitled ``Plant-pesticides; Proposed Exemption from the Requirement of 
a Tolerance Under the Federal Food, Drug, and Cosmetic Act.''
    b. Exemption from the requirement of a tolerance for nucleic acids 
produced in plants as part of a plant-pesticide. This proposed 
exemption would exempt nucleic acids (i.e., deoxyribonucleic acid (DNA) 
and ribonucleic acid (RNA)) from the requirement for a food tolerance 
when produced in plants as part of a plant-pesticide active or inert 
ingredient (refer to Unit IV.B.1. of this document for a discussion of 
the definition of plant-pesticide active and inert ingredients). The 
proposed exemption is as follows:

    Residues of nucleic acids produced in living plants as part of a 
plant-pesticide active or inert ingredient, including both 
deoxyribonucleic and ribonucleic acids, are exempt from the 
requirement of a tolerance.

    For the purposes of this exemption, ``nucleic acids'' means 
ribosides or deoxyribosides of adenine, thymine, guanine, cytosine, and 
uracil and the polymers of these ribosides and deoxyribosides and does 
not apply to nucleic acid analogues.
    c. Exemption from the requirement of a tolerance for viral coat 
proteins produced in plants. EPA proposes to exempt from the 
requirement of a tolerance coat proteins from plant viruses when they 
are produced by plants to enable the plants to resist viral infection. 
The proposed exemption is as follows:

    Residues of coat proteins from plant viruses, or segments of the 
coat proteins, produced in living plants as plant-pesticides are 
exempt from the requirement of a tolerance.

    2. Rationale for exemptions under FFDCA. There are circumstances 
where EPA believes that plant-pesticides should be reviewed by EPA 
either to set a tolerance or issue an exemption from the requirement of 
a tolerance. In general, EPA believes that plant-pesticides resulting 
in significantly different dietary exposures than those that already 
occur should be subject to EPA review under FFDCA tolerance procedures.
    a. Exemption from the requirement of a tolerance for categories of 
plant-pesticides that would not result in new dietary exposures. Many 
substances having pesticidal activity occur naturally in the edible 
parts of plants (i.e., they are inherent to the plant) and have long 
been accepted as part of the human diet. The safety of foods containing 
these substances is demonstrated by extensive consumption and 
experience. For many foods, the inherent toxicants they may contain, 
including pesticidal substances, are known (Ref. 5). Also, the 
established practices that plant breeders employ in selecting and 
developing new plant varieties, such as chemical analyses, taste-
testing and visual analyses, have historically proven to be reliable 
for ensuring food safety. That there are few examples of new plant 
varieties causing food safety concerns, despite the large numbers of 
new varieties introduced into commerce each year, is a reflection of 
the effectiveness of this process. Moreover, consumer experience with 
the handling and preparation of food from these plants contributes to 
the safety of food from these plants.
    (i) Plant-pesticides from closely related plants. This proposed 
exemption is based upon the premise that new dietary exposures would 
not likely arise for plant-pesticides produced in food plants if the 
genetic material leading to the production of the plant-pesticide is 
derived from closely related plants. In establishing this exemption, 
EPA is proposing to use a standard that is similar to the standard 
proposed for an exemption for plant-pesticides under FIFRA (see Unit 
IV.B. of this document). That standard would be sexual compatibility. 
Under both statutes, this standard would be used as a measure of 
relatedness between plants. However, under FFDCA, the standard of 
sexual compatibility must be examined specifically within the context 
of the food supply and dietary consumption. The Agency believes, based 
on the experience with sexually compatible plants (see Unit IV.B. of 
this document), that most plant varieties developed by plant breeders 
using genetic material from plants that meet the sexually compatible 
standard produce food that is safe for human consumption and/or that 
appropriate processing procedures are widely known and routinely used 
by consumers in preparation of food from such sources.
    As under FIFRA, EPA proposes to extend the concept of sexual 
compatibility to include wide crosses because wide crosses are commonly 
used to expand the gene pool for varietal improvement and, as discussed 
earlier, EPA believes that the fact that wide crosses can produce a 
viable zygote indicates a fairly high degree of relatedness between 
parental plants.
    The Agency is considering, for this exemption, a second option 
based primarily on the taxonomic standard of genus rather than on 
sexual compatibility. Under this approach, a plant-pesticide would be 
exempt if it were derived from a plant within the same genus as the 
recipient plant. The Agency recognizes that some plants that are 
closely related (as evidenced by sexual compatibility) are not 
classified in the same genus. Under this alternative option, the Agency 
would extend the exemption to plant-pesticides derived from plants in 
these populations, as well as to intrageneric plant-pesticides. 
Therefore, if a plant-pesticide is derived from a plant outside of the 
same genus, sexual compatibility between the two plants would need to 
be demonstrated. The language defining this option would be as follows.
[Residues of pesticidal substances produced in living plants as 
plant-pesticides are exempt from the requirement of a tolerance if:]

    The genetic material that encodes for a pesticidal substance or 
leads to the production of a pesticidal substance:
    (1) Is derived from plants that are within the same genus as the 
recipient plant [regardless of sexual compatibility] or, is derived 
from plants that are sexually compatible with the recipient plant; 
and
    (2) Has never been derived from a source outside the same genus 
that is not sexually compatible with the recipient plant.

    For a more detailed analysis of EPA's preferred approach and Option 
2, refer to the Federal Register document published elsewhere in 
today's issue of the Federal Register entitled, ``Plant-pesticides; 
Proposed Exemption from the Requirement of a Tolerance under the 
Federal Food, Drug, and Cosmetic Act.''
    (ii) Plant-pesticides derived from food plants that are not closely 
related with the recipient plant. There are circumstances where 
experience with exposure can be inferred for plant-pesticides 
introduced into food plants from other food plants that are not closely 
related to the recipient plant. For plant-pesticides derived from a 
food plant that is not sexually compatible with the recipient food 
plant, there is experience with exposure because both plants have 
contributed to the food supply. Thus, the Agency is proposing to exempt 
from the requirement of a tolerance plant-pesticides derived from food 
plants that are not closely related to the recipient plant, if there 
would not be significantly different dietary exposures when the plant-
pesticide is produced in the recipient food plant.
    The Agency has defined a set of criteria to determine whether 
significantly different dietary exposures from the these plant-
pesticides will occur. For example, if a pesticidal substance is 
normally only produced in inedible portions or immature fruit of the 
food plant, the Agency would require a tolerance review if the modified 
food plant were to produce that substance in its mature fruit or edible 
portions. For example, tomatine is a toxicant produced in much higher 
amounts in immature tomato fruit (that is normally eaten) than it is in 
the fruit. If the genetic material leading to the production of 
tomatine were introduced into a plant for pesticidal purposes such that 
the tomatine were produced in the mature fruit as it is in the immature 
fruit, the Agency would need to conduct a tolerance review to determine 
whether a tolerance is necessary to protect the public health. 
Similarly, if a pesticidal substance is produced in a food that is 
almost always cooked or processed prior to consumption, the Agency 
would want to conduct a tolerance review if another food plant that is 
not cooked or processed prior to consumption is modified to produce the 
substance. For example, some beans are rich in lectins, glycoproteins 
that are natural toxicants. Soaking and cooking the beans destroys the 
lectins. If the genetic material encoding lectins were transferred, for 
pesticidal purposes, from beans to a plant which is not normally cooked 
(e.g., lettuce), the Agency would need to conduct a tolerance review. A 
significantly different dietary exposure could also result if a widely 
consumed food staple such as corn is modified to produce a pesticidal 
substance from a food crop with minor consumption such as eggplant.
    EPA is also considering adding another criterion to the exemption 
from the requirement of a tolerance for categories of plant-pesticides 
that would not result in new dietary exposures (Unit IV.C.1.a. of this 
document). This criterion would address the potential for allergenicity 
of plant-pesticides in food. Under this criterion, if a plant-pesticide 
is derived from a commonly allergenic food, the plant-pesticide would 
not be exempt from tolerance requirements and the Agency would conduct 
a tolerance review on a case-by-case basis to determine whether to 
establish an exemption from the requirement of a tolerance, establish a 
tolerance, or deny a tolerance. Some examples of foods that commonly 
cause an allergenic response are milk, eggs, fish, crustacea, molluscs, 
tree nuts, wheat, and legumes (particularly peanuts and soybeans).
    b. Nucleic acids in plants. The Agency is also proposing to exempt 
from the requirement of a tolerance nucleic acids produced in plants as 
part of plant-pesticide active or inert ingredients. Nucleic acids 
(deoxyribonucleic and ribonucleic acid) are present in the cells of 
every living organism, including plants, microorganisms, and animals. 
Because nucleic acids are ubiquitous in the food supply and lack any 
toxicity when they are consumed in food, EPA believes that a tolerance 
for nucleic acids is not necessary to protect the public health (see 
the proposal published elsewhere in todays issue of the Federal 
Register entitled ``Plant-pesticides; Proposed Exemption from the 
Requirement of a Tolerance Under the Federal Food, Drug, and Cosmetic 
Act for Nucleic Acids Produced in Plants'' for a more detailed 
discussion).
    c. Coat proteins from plant viruses. The Agency's proposal for 
exempting coat proteins from plant viruses is based on virus-infected 
plants having always been a part of the human and domestic animal food 
supply without detectable adverse human health effects. There is no 
evidence of any plant virus being able to replicate in mammals or other 
vertebrates. In addition, the exemption will only apply when the 
portion of the viral genome coding for the whole coat protein or a sub-
component of the coat protein will be expressed in the plant. This 
portion of the viral genome by itself is incapable of forming 
infectious particles. Since whole, intact plant viruses are not known 
to cause untoward human health effects, it is reasonable to assume that 
a subunit of these viruses will not be harmful (see the proposal 
published elsewhere in todays issue of the Federal Register entitled 
``Plant-pesticides; Proposed Exemption from the Requirement of a 
Tolerance Under the Federal Food, Drug, and Cosmetic Act for Viral Coat 
Proteins Produced in Plants'' for a more detailed discussion).

D. Plant-pesticides and Plant Regulators

    As discussed in Unit III. of this document, FIFRA section 2(u) 
defines ``pesticide'' as: ``. . .(2) any substance or mixture of 
substances intended for use as a plant regulator, defoliator, or 
desiccant. . . .'' FIFRA section 2 also defines ``plant regulator'' as 
``. . .any substance or mixture of substances intended, through 
physiological action, for accelerating or retarding the rate of growth 
or rate of maturation, or for otherwise altering the behavior of the 
plants or the produce thereof. . . .'' With regard to substances 
applied to plants, this definition has been interpreted to include, for 
example, plant hormones (e.g., auxin, gibberellin, cytokinin and 
ethylene). In the area of plant-pesticides, the issue arises as to 
which substances produced by a plant should be considered plant 
regulators, as a result of a change in the physiology of the plant, and 
therefore, pesticides, subject to regulation.
    At the time that the term ``plant regulator'' was added to the 
definition of pesticide, in 1959, Congress addressed substances applied 
to plants but did not address how the definition applied to substances 
produced in plants. EPA believes this is because the technology to 
develop plant varieties expressing substances using genetic information 
derived from diverse sources (e.g., outside the plant kingdom) was not 
in existence, and thus, Congress did not provide direct guidance on the 
implications of the definition of plant regulator for substances 
produced in plants.
    EPA, therefore, believes that it has the discretion to develop a 
reasonable approach to defining what constitutes a plant regulator for 
substances produced in plants. In developing this interpretation, EPA 
looked at previous Congressional action relating to ``plant 
regulators,'' plant science, the traditional roles of EPA and FDA in 
this area, and the extent to which risk concerns would go unaddressed 
if EPA did not include certain plant substances in the definition of 
``plant regulator.''
    While Congress has not spoken on the full extent of the definition 
of ``plant regulator,'' it has given some guidance on what it does not 
consider to be plant regulators through exclusions to the definition in 
FIFRA itself. For example, Congress specifically excluded ``substances 
to the extent that they are intended as plant nutrients, trace 
elements, nutritional chemicals, plant inoculants, and soil 
amendments.'' Arguably, Congress recognized that no purpose would be 
served by requiring substances such as plant nutrients to be regulated 
as ``pesticides.''
    In 1972, Congress added the vitamin-hormone horticultural product 
exclusion to the definition of plant regulator. This exclusion provides 
that ``the term plant regulator shall not be required to include any of 
such of those nutrient mixtures or soil amendments as are commonly 
known as vitamin-hormone horticultural products, intended for 
improvement, maintenance, survival, health, and propagation of plants, 
and as are not for pest destruction and are non-toxic, nonpoisonous in 
the undiluted packaged concentration'' (FIFRA section 2(v) emphasis 
added). This exclusion can be used as a reasonable starting point for 
EPA's identification of the types of substances that fall within its 
jurisdiction.
    Another factor that the Agency considered in developing its 
approach to ``plant regulators'' was harmonization with the traditional 
split between EPA and FDA jurisdiction under FFDCA. The reorganization 
plan of 1970 generally placed the FFDCA responsibility for setting 
tolerances for pesticide residues in food under EPA's jurisdiction and 
the responsibility for regulating all other substances in food under 
FDA's jurisdiction. FDA traditionally regulates, for example, 
substances in food that are used for improved food processing or 
improved nutritional content. The issue here, therefore, is not whether 
or under what statutory authority a substance will be regulated. 
Rather, the issue is who will regulate. If a substance is defined by 
FIFRA as a pesticide, it is subject to EPA's regulatory authority. If 
the substance is not a pesticide under FIFRA, FDA has regulatory 
responsibility. EPA believes it is reasonable to develop a plant 
regulator interpretation for plant-pesticides that provides for FDA to 
regulate the types of substances that it has experience and expertise 
in regulating and that avoids regulation by EPA as ``pesticides,'' 
substances that relate to nutrition and food quality. Accordingly, EPA 
is proposing the following interpretation of ``plant regulator'' for 
the purpose of determining which substances produced by a plant as a 
result of changes in the plant's physiology should be considered to be 
``pesticides,'' subject to EPA's regulatory jurisdiction. EPA believes 
that this interpretation is consistent with Congress' intent of 
including a category of substances they termed ``plant regulators,'' or 
at least certain types of growth regulators, within the definition of 
pesticide.
    A substance that is produced in a plant as a result of a change in 
the plant's physiology would be considered a plant regulator if:
    It is intended to accelerate or retard the rate of growth or rate 
of maturation, or alter the behavior of the plants and meets one of the 
following criteria:
    (1) Is a plant hormone.
    (2) Acts to prevent, destroy, repel, or mitigate a pest.
    (3) Is toxic in concentrations found in the plant (undiluted 
package).
    Plant hormones that are produced in plants as the result of an 
intentional change in the plants' physiology would be considered plant 
regulators. As plant regulators, they would also be considered a plant-
pesticide and under EPA's authority. However, EPA believes that some of 
these substances would be candidates for exemption from regulation 
under FIFRA under the exemption of plant-pesticides derived from 
closely related plants (see Unit IV.B.1.c.i. of this document and the 
proposal published elsewhere in today's issue of the Federal Register 
entitled ``Plant-pesticides Subject to the Federal Insecticide, 
Fungicide, and Rodenticide Act; Proposed Rule''). EPA is also 
considering extending the exemption of plant-pesticides that act 
primarily by affecting the plant to include substances such as plant 
hormones (see Unit IV.B.1.c.i. of this preamble and the proposal 
published elsewhere in today's issue of the Federal Register entitled 
``Plant-pesticides Subject to the Federal Insecticide, Fungicide, and 
Rodenticide Act; Proposed Rule'').
    A few examples of substances expressed in plants that would not be 
considered plant regulators and therefore not under EPA's authority 
are: (1) Substances intended to alter the nutritional composition of 
the plant; (2) substances intended to enhance the plant's resistance to 
chemical herbicides; and (3) substances intended to alter the flavor or 
the texture of the food.

V. Regulatory Process Under FIFRA and FFDCA

    This section outlines the process EPA proposes to follow for plant-
pesticides subject to FIFRA and/or FFDCA requirements. It describes the 
options that EPA is evaluating for its approach to the regulation of 
testing of plant-pesticides and includes current proposed Agency's 
thinking for sale or distribution of a plant-pesticide. In the future, 
EPA will propose regulations concerning the procedures for plant-
pesticides under FIFRA (e.g., EUPs and labeling).
    In the period before procedures that are specific for plant-
pesticides are finalized, existing regulations (e.g., 40 CFR parts 156, 
158, and 172) will be used as the basis for plant-pesticide regulatory 
procedures. However, the Agency is aware that many of these existing 
procedures may not be appropriate for plant-pesticides, and encourages 
producers to contact EPA on a case-by-case basis.
    Producers of plant-pesticides should be aware that having no 
obligations under one of the statutes (FIFRA or FFDCA) does not 
necessarily mean that there are no obligations under the other statute. 
Producers should therefore evaluate the requirements of both statutes 
before reaching a determination on their responsibilities.
    Producers should also be aware that if certain plant-pesticides 
present unreasonable risk and there are limited or no risk mitigation 
options available to the Agency for certain plant-pesticides submitted 
for registration, the Agency would not be able to register the plant-
pesticides. Potential registrants are again encouraged to consult with 
EPA for guidance early in the product development cycle.

A. Testing of Plant-pesticides

    The following unit outlines EPA's preferred approach to when a 
producer would contact EPA under FIFRA and FFDCA during testing of a 
plant-pesticide. The procedures are laid out in terms of whether the 
crop would be used as food and/or feed since this question is among the 
first considerations in determining when to contact EPA.
    1. Food and/or feed use. This unit gives guidance as to when a 
producer would have obligations under FFDCA and FIFRA if a plant-
pesticide is tested in a crop/plant used for food and/or feed or if any 
adjacent food/feed plants would produce the plant-pesticide.
    a. FFDCA requirements. The FFDCA requirements for a plant-pesticide 
would be based on the following considerations.
    (i) Existing exemption from the requirement of a tolerance. If 
there is an existing exemption from the requirement of a tolerance for 
the plant-pesticide in the crop/plant to be tested (e.g., if EPA's 
proposed tolerance exemptions published elsewhere in todays issue of 
the Federal Register), there would be no further obligations under 
FFDCA.
    (ii) Containment. A tolerance review would not be required for the 
test if the crops/plants containing the plant-pesticide are tested in a 
way that ensures adequate containment within the test site of the 
genetic material necessary for the production of the pesticidal 
substance. ``Adequate containment'' in this context would include 
ensuring any adjacent crops, to be used as food, do not express the 
plant-pesticide as a result of successful pollination by the plant used 
to test the plant-pesticide.
    (iii) Crop destruct. A tolerance review would not be required for 
the test if all crops within the test site are destroyed or used for 
experimental purposes only.
    (iv) Petition for temporary tolerance or exemption. If there is no 
existing exemption from the requirement of a tolerance for the plant-
pesticide, if containment is not adequate or the crop is not destroyed, 
a petition for a temporary tolerance, a full tolerance, or an exemption 
from the requirement of a tolerance must be submitted to the Agency, as 
described at 40 CFR 180.7 and 180.31. Such a crop cannot be sold or 
distributed for use as food or feed unless a tolerance or exemption 
from the requirement of a tolerance has been obtained for the plant-
pesticide in that crop and the plant-pesticide residues fall within the 
tolerance limits.
    b. FIFRA requirements. Prior to registration, if a plant- pesticide 
is produced in a crop/plant that is to be used as food or feed 
(condition (iv) above) at any field test acreage, producers would apply 
for an Experimental Use Permit (EUP) for that plant-pesticide under 
FIFRA.
    2. Nonfood and nonfeed use. This unit describes when a producer 
would have obligations under FFDCA and FIFRA if the plant-pesticide is 
used in a crop or plant that will not be used for food and/or feed.
    a. FFDCA requirements. If a plant-pesticide is produced in a crop/
plant that will not be used as food and/or feed, there would be no 
requirements under FFDCA as long as the genetic material is adequately 
contained to avoid successful transfer and expression of the plant-
pesticide in adjacent crops. (Refer to Unit V.A.1.a.(ii) and (iii). 
above for conditions of containment and crop destruction under which 
there are no FFDCA requirements.)
    b. FIFRA requirements. If a plant-pesticide is produced in a crop/
plant that will not be used as food and/or feed, an Experimental Use 
Permit would be required under the following conditions.
    (i) Not subject to the Plant Pest Act. If a plant-pesticide is 
produced in a plant that is not subject to the authority of the Plant 
Pest Act, an EUP would be required at first field introduction. This 
EUP requirement would extend to plant-pesticides in plants that are not 
within the statutory jurisdiction of the Plant Pest Act, including 
those for which APHIS has made a determination of nonregulated article 
status (58 FR 17044).
    (ii) Acreage requirements for plant-pesticides in plants subject to 
the Plant Pest Act. For plant-pesticides produced in plants that are 
(1) subject to the authority of the Plant Pest Act and (2) not used as 
food or feed, an EUP would be required when one of the following two 
conditions is met.
    (A) Acreage limit for individual field test. An EUP generally would 
be required if an individual field test for a plant-pesticide in a 
particular crop will be on greater than 10 acres of land. Once a 
requirement for an EUP has been triggered, if there are other field 
test site locations, for that plant-pesticide, of less than 10 acres 
that would also be planted in the same year, they would be included in 
the EUP. If a field test site of greater than 10 acres is planted for a 
crop that is producing more than one plant-pesticide, all field test 
sites for each of the plant-pesticides would be included in the EUP if 
these field tests would occur in the same year.
    (B) Upper cumulative acreage limit for field tests. Notwithstanding 
(A) above, an EUP would be required if the cumulative acreage of all 
field tests for the plant-pesticide in a particular crop exceeds 50 
acres of land regardless of the acreage of individual field sites.
    (iii) Other conditions for EUPs. The Agency could grant multi-year 
EUPs for plant-pesticides under certain conditions. For example, a 
multi-year EUP could be appropriate if the acreage of the field sites 
will increase yearly but the field test design and containment measures 
for the field sites remain the same such that there is not an increased 
risk for nontarget effects or outcrossing. In addition, producers 
should be aware that the data requirements for an EUP may not be 
different from the data requirements for a registration of the plant-
pesticide (see Unit VI. of this document).

B. Sale or Distribution

    1. FFDCA. If a plant-pesticide is produced in a crop to be sold or 
distributed as food and/or feed and is not already exempt from the 
requirement of a tolerance, the Agency must establish a tolerance or 
exemption from the requirement of a tolerance before sale or 
distribution.
    2. FIFRA. Before sale or distribution of a plant-pesticide, a 
producer would have to obtain a registration for the plant-pesticide 
product, unless it is otherwise exempt, as described in this document 
(Unit VI.B. of this document). The plant-pesticide product consists of 
the active ingredients and inert ingredients, as defined in Unit IV.B. 
of this document, EPA anticipates that the plant-pesticide product 
would be registered for use in a particular plant or crop (e.g., field 
corn). While the Agency anticipates that most registrations of plant-
pesticides would be for use in a particular plant or crop, there may be 
instances where a registration of a plant-pesticide could be limited to 
a particular variety or be restricted in some other way if risk 
considerations warrant such a restriction.
    In terms of shipping, EPA does not intend to change the status of 
the exemption under section 12(b)(5) of FIFRA which allows the shipping 
of a pesticidal substance under the conditions of section 12(b)(5) 
without being subject to penalty for failure to have a registration or 
EUP.

C. Agency Considerations Underlying Regulatory Procedures and 
Alternative Approaches for Testing of Plant-pesticides

    1. Agency considerations underlying FFDCA regulatory procedures. 
Under FFDCA at the field testing stage, the Agency will generally treat 
plant-pesticides as it treats other pesticides in terms of when EPA 
regulatory oversight begins. Thus, producers will be subject to FFDCA 
requirements when the crop plant containing the plant-pesticide is to 
be used as food or feed. EPA recognizes that many of its requirements 
for addressing feed under FFDCA arise because some pesticidal 
substances may be metabolized or stored by domesticated animals in ways 
that expose humans to these pesticides and/or their residues through 
consumption of meat or other animal products (e.g., milk and eggs). EPA 
also recognizes that the possibility that consumers might be exposed to 
proteinaceous plant-pesticides through such animal products is 
extremely low. Proteinaceous plant-pesticides are likely to be composed 
of the same constituents as animal protein or other animal cellular 
components, and, thus, would readily enter the metabolic cycles of the 
animal cell. For proteins, it is not anticipated that recalcitrant 
residues will be generated or accumulated in animals used as sources of 
meat. However, EPA cannot predict that all plant-pesticides will behave 
as proteinaceous plant-pesticides. Thus, EPA will require plant-
pesticides to be subject to FFDCA when the plants producing the plant-
pesticide are used as feed. Data requirements associated with the 
tolerance review will, however, be imposed recognizing the 
characteristics of proteinaceous plant-pesticides.
    2. Agency considerations underlying FIFRA regulatory procedures. 
The following unit describes EPA's rationale for its preferred trigger 
for EUP's and describes alternative approaches to the EUP trigger EPA 
is considering. The preferred approach and the alternative approaches 
address testing of plant-pesticides that are subject to the Plant Pest 
Act and are not used as food and/or feed (conditions described in Unit 
V.A.2.b.ii. of this document).
    EPA believes some type of oversight of plant-pesticides at the 
field testing stage is appropriate. Plant-pesticides are, by 
definition, part of a living organism. Thus, plant-pesticides present 
unique mechanisms by which they can be produced and spread in the 
environment. Because of the different risk considerations and risk 
mitigation measures associated with plant-pesticides than with 
traditional chemical pesticides, EPA is examining whether the 10 
cumulative acre presumption in 40 CFR 172.3 is appropriate with regard 
to plant-pesticides.
    Potential risks associated with tests of plant-pesticides will 
depend upon a number of variables, including the size of the test plot, 
the biology of the plant and the properties of the pesticidal 
substance. At smaller acreages, tests of plant-pesticides can usually 
be designed with containment that is adequate to minimize the spread of 
the plants' genetic material beyond the test site, thereby limiting the 
spread of the active ingredient from the field test site. Successful 
containment of the genetic material would result in minimal exposure of 
humans and other nontarget organisms to the active ingredient beyond 
the test site.
    During the development of plant-pesticides, depending upon the 
biology of the plant tested and the location of the field test sites, 
there will be a point at which it will be impractical to try to contain 
the spread of the genetic material. In terms of risk, the Agency 
believes its oversight of plant-pesticides under FIFRA should begin at 
the point when containment is impractical and the potential for 
significant exposure to nontarget organisms begins to increase.
    As tests of plant-pesticides progress to larger acreages, the 
potential hazards to nontarget organisms will generally be increased 
because of the potential for significant exposure to nontarget 
organisms on the test site. In addition, lack of adequate containment 
may mean the spread of the active ingredient and subsequent possible 
environmental exposure beyond the test site. Moreover, if the active 
ingredient is spread to neighboring crops used for food and/or feed, 
human dietary exposure could occur.
    Under any of the options put forth by EPA for EUP thresholds for 
plant-pesticides, EPA would retain the authority to rebut the 
presumption that an EUP is not required for certain small-scale 
testing. Such a rebuttal would be based on risk/benefit considerations. 
Thus, EPA may, on a case-by-case basis require EUP's for testing 
conducted with plant-pesticides at acreages smaller than those 
described in the options. EPA does not anticipate requiring EUP's at 
acreages below the threshold triggering EUP requirements (regardless of 
which option is chosen) very often, and when EPA determines that such 
an EUP is warranted, EPA will provide notice to the producer of the 
plant-pesticide being tested.
    In addition to risks in terms of hazard and exposure, EPA 
considered the following in developing its options for the threshold 
for EUPs for plant-pesticides: (1) Differences in the traits of plant-
pesticides (e.g., the delivery system of the pesticide in that plant-
pesticides are produced and used in a living plant) in comparison to 
more traditional chemical pesticides (e.g., they are applied to a 
plant); (2) varietal development procedures used in the plant breeding 
industry and the impact this has on the design of field testing; (3) 
USDA's activities under the Plant Pest Act for transgenic plants (some 
of which will be engineered to express plant-pesticides); (4) clarity 
to the regulated and other communities; (5) EPA's traditional approach 
to oversight of field testing in terms of establishing acreage cutoffs 
so as to provide regulatory consistency; and (6) costs to potential 
registrants and efficient utilization of Agency resources.
    a. EPA's preferred approach: single-site acreage threshold. Under 
its preferred approach, the Agency would use the concept that larger 
acreage for individual field sites leads, in general, to greater 
exposures and greater potential for escape from biological containment. 
Therefore, the Agency would link the concept of greater exposure to an 
acreage cutoff (i.e., 10 acres for a single field test site, 50 
cumulative acres; see Unit V.A.2.b.ii of this document). EPA recognizes 
that the acreage cutoff in its preferred approach may be more closely 
correlated with the potential for larger exposures for some crops than 
for others. However, it believes that the advantage of clarity and 
predictability associated with the acreage cutoff would outweigh this 
disadvantage. In addition, EPA would put an upper limit on the number 
of cumulative acres that could be planted without an EUP because 
multiple sites of larger acreages would lead to an increased potential 
for exposure to nontarget organisms on the cumulative acreage.
    b. Alternative approaches for testing of plant-pesticides. EPA is 
also considering four alternative approaches to triggers for Agency 
oversight for testing of plant-pesticides, under FIFRA section 5, for 
plant-pesticides that are subject to the Plant Pest Act and are not 
used as food/feed.
    (i) Containment as trigger for EUP. EPA seriously considered, and 
may yet choose to use as its trigger for EUP requirements, the concept 
of containment. Essentially, once a plant can no longer be effectively 
isolated biologically, EPA's EUP requirements would be triggered. To be 
isolated biologically means that the genetic material of the plants on 
the test site does not have a significant potential for successfully 
spreading to neighboring plants (including neighboring crop plants) 
through sexual recombination. The concept of biological isolation is 
basic to USDA's approach and use of such a standard by both agencies 
could permit a smooth transfer of oversight for plant-pesticides from 
USDA to EPA. The disadvantage of using this approach for EPA and for 
potential registrants, is that what constitutes ``appropriate 
containment'' varies from crop to crop and test to test. EPA is 
concerned that the lack of a clear line as a standard may result in 
numerous consultations between EPA and potential registrants over what 
constitutes appropriate containment. In addition, other groups, such as 
the public and public interest groups, may not be able to readily 
determine whether a potential registrant is in compliance with EPA 
requirements. Finally, there are exposure issues (i.e., larger 
potential for exposure on individual field sites at larger acreages) 
that are not addressed by this alternative.
    (ii) USDA/APHIS determination of nonregulated status as trigger for 
EUP. A second alternative approach EPA is considering is based on 
USDA's determination of nonregulated article status (58 FR 17044). 
Under this alternative, producers would apply to EPA for an EUP (or 
registration) at the time that they apply to APHIS for a determination 
of nonregulated article status. Although this approach could 
potentially minimize duplicative efforts by the two agencies, it lacks 
regulatory clarity and consistency as to when a producer would have to 
comply with EPA requirements. This approach may result in producers 
being uncertain as to when they should contact EPA. The result of this 
lack of regulatory clarity could be producers applying to EPA too late 
in their product development cycle.
    (iii) Cumulative acreage in a single state as trigger for EUP. The 
third alternative would be based on an acreage trigger, as is EPA's 
preferred approach. Under this approach, an EUP would be required when 
the cumulative acreage in any one state exceeds 10 acres. This 
alternative would allow producers to test a plant-pesticide in a number 
of different locales yet limit exposure to the plant-pesticide in any 
one locale. However, this approach also lacks regulatory clarity as to 
when producers must contact EPA if a number of different states are 
involved. Other groups, such as the public and public interest groups 
may not be able to readily determine whether a producer is in 
compliance with EPA requirements. In addition, fairly large total 
acreages of testing might occur since up to 10 acres could be tested in 
each of 50 states. The potential for exposure to nontarget organisms 
and outcrossing to wild relatives could be greater with this 
alternative than for the preferred or other alternative approaches.
    (iv) Cumulative acreage as trigger for EUP. A fourth alternative 
EPA is considering is to utilize for plant-pesticides the acreage 
presumption in the current EUP regulations; i.e., an EUP is presumed 
not to be required at less than 10 cumulative acres of land. The 
advantage this alternative presents is consistency in EPA's EUP 
regulations for all pesticides. The disadvantages to this approach are 
that it may not correlate well with current plant breeding procedures 
(varietal testing) that can require a number of field sites of smaller 
acreages and it may result in more duplicative efforts by EPA and USDA.
    It should be noted that EPA has not addressed in this document how 
it will approach aquatic testing of plant-pesticides. Producers 
anticipating research and commercialization activities with aquatic 
plants are encouraged to contact the Agency.

D. Labeling Requirements under FIFRA

    Labeling is required for pesticides that are regulated by EPA under 
FIFRA. Labeling includes both written material accompanying the 
pesticide and labels on or attached to the pesticide, its container, or 
wrapper. Labeling thus may have different forms. A pesticide which does 
not meet labeling requirements is considered to be misbranded and 
enforcement action can be taken.
    The Agency recognizes that certain types of labeling which are 
appropriate for chemical pesticides will not be practical for plant-
pesticides. For example, it would be impractical to require labels to 
be physically attached to the plant-pesticide itself (i.e., the 
pesticidal substance and the genetic material necessary for the 
production of the substance) at any point in the regulatory process 
(i.e., EUPs or registration). Therefore, the Agency is considering the 
types of labeling that would be appropriate for plant-pesticides during 
testing and at the time of registration.
    Labeling can include both prescriptive and informational 
components. The Agency is considering utilizing both these types of 
labeling for plant-pesticides. For example, plant-pesticides that are 
regulated by EPA but are not yet registered (e.g., are under an EUP) 
may have prescriptive labeling that would set forth the appropriate 
conditions for field testing, such as geographic location, field test 
design, and other limitations. The Agency believes that this type of 
labeling would be appropriate at this point in the regulatory process 
because the Agency would not have yet made a determination that the 
plant-pesticide generally will not result in ``unreasonable adverse 
effects'' without the restrictions specified on the label. Thus, the 
prescriptive labeling would be necessary to assure there would not be 
unreasonable adverse effects during the test of the plant-pesticide.
    Plant-pesticides that are registered would also have prescriptive 
labeling that would accompany the plant-pesticide throughout the 
process of developing and producing the commercial plant variety that 
contains the plant-pesticide. Such prescriptive labeling would, for 
example, specify the EPA registration number, the ingredient statement, 
and the plants/crops in which the plant-pesticide may be produced. It 
is unlikely, however, to include many of the limitations, discussed 
above, likely to be used for unregistered plant-pesticides because 
these types of limitations would not be practical for registered plant-
pesticides. For example, limitations such as conditions for planting, 
field design, or certain geographical restrictions may not be practical 
given the nature of the use and distribution patterns of the plants 
that produce the plant-pesticide (e.g., farmers saving seed for 
replanting, the potential for spread of the plants' genetic material). 
In some cases it will be appropriate for the Agency to assume that, at 
sale or distribution, the plant-pesticide will be introduced into all 
plants/crops (and their varieties) that are included in the plant-
pesticide registration (e.g., all corn varieties). The appropriateness 
of this assumption will be considered, on a case-by-case basis, by EPA 
in its assessment of whether a plant-pesticide produced in a particular 
crop/plant presents ``no unreasonable adverse effects.''
    The prescriptive label for a registered plant-pesticide may include 
a provision requiring informational labeling on plants/seeds containing 
the plant-pesticide to give information or notice to farmers and 
growers of the plant-pesticide. For example, informational labeling of 
this type could be attached to bags of seeds and could inform farmers 
of the type of pesticide that the plants will produce and against which 
pest it is active. EPA believes that such informational labeling will 
help to prevent unnecessary application of additional pesticides to the 
plants. The prescriptive label may require that the informational 
labeling accompanying plants or seeds bear other necessary statements. 
For example, the informational label could be required to have a 
statement that informs farmers and growers that they should report any 
adverse effects to the registrant through an address or telephone 
number provided on the label.

E. Import/Export of Plant-pesticides

    Unregistered, exported plant-pesticides that are not exempt from 
FIFRA regulation will be subject to the Final Export Policy Statement; 
Rule (40 CFR parts 168 and 169, 58 FR 9062). However, the Agency 
recognizes that some of the labeling requirements for exports may not 
be applicable to plant-pesticides (e.g., net weight of the plant-
pesticide may be difficult to determine in some cases). EPA is also 
considering whether to require an informational label on exported 
seeds/plants containing a registered plant-pesticide that contains: (1) 
Information about the type of plant-pesticide produced by the plants 
and the target pest, and (2) a statement that EPA's determination for 
registration of the plant-pesticide is based solely on consideration of 
risks in the United States and that this determination does not extend 
to use in other countries.
    If an imported plant-pesticide will be used for pesticidal purposes 
(e.g., if the seeds are sold to be planted and the resulting plants are 
intended to produce the plant-pesticide), the producer must apply for a 
registration of the plant-pesticide under FIFRA. As with labeling for 
exports, certain provisions in the ``Notice of Arrival of Pesticides 
and Devices'' (EPA form 3540-1) may not be applicable to plant-
pesticides. If a producer plans to import seeds/plants that contain a 
plant-pesticide, a tolerance or exemption from the requirement of a 
tolerance must be obtained if the plants/seeds will be used for food 
and/or feed. The Agency encourages producers to contact EPA with 
questions concerning procedures for both import and export of plant-
pesticides.

VI. Information Needs Under FIFRA and FFDCA

A. Introduction

    This unit describes the types of information EPA would need, in 
general, to evaluate a plant-pesticide during product development and 
before the pesticide could be registered for sale or distribution. The 
types of information the Agency would need for its evaluation of a 
plant-pesticide will depend upon a number of variables, including the 
biology of the plant and the properties of the pesticidal substance. 
The specifics of EPA's evaluation of testing of a plant-pesticide 
(e.g., evaluation of containment) will also depend, to a certain 
extent, on which of the options for EUP triggers apply in a final 
policy statement (i.e., the time during product development that a 
producer is required to contact the Agency varies depending on the 
option selected (see Unit V. of this document).
    This unit is designed to give general guidance while maintaining an 
appropriate flexibility to data needs for individual cases. It covers 
the areas of product analysis, environmental fate, ecological effects, 
and human health effects. Producers are encouraged to consult with the 
Agency on the types of information relevant to EPA's evaluation of 
their particular product.
    Although plant-pesticides present new considerations for risk 
assessment, the framework that is used to evaluate risks for chemical 
and biological pesticides can be used as a model if the unique aspects 
of the plant-pesticides are taken into account. The major 
characteristic of plant-pesticides that is different from traditional 
pesticides is that the plant itself produces the pesticidal substance 
rather than the pesticide being applied to the plant.
    Thus, the exposure pattern may be very different for plant-
pesticides than for traditional pesticides, both because of how the 
pesticide is produced and the biology of plants. This different 
exposure pattern has implications for the evaluation of field tests 
(e.g., the analysis of containment of the test) and for exposures 
resulting from sale or distribution. In turn, the unique exposure 
potential may involve different non-target or endangered species than 
for a traditional pesticide. However, the potential for causing adverse 
health effects may be more circumscribed than for traditional 
pesticides because, in many cases, the only significant route of human 
exposure may be oral.

B. General Considerations for Product Development and Sale or 
Distribution

    1. Product development. As a plant-pesticide product is developed, 
it is usually necessary to test, in the field, plants producing the 
pesticide. At this point, EPA may not have extensive information on 
ecological and human health effects for these products. Unreasonable 
adverse effects associated with field tests can be minimized by 
conducting these tests under conditions appropriate to ensure 
containment. Adequate containment would ensure that exposure of humans 
and nontarget organisms to the active ingredient beyond the field test 
site is minimal. Thus, for field testing during product development the 
Agency would need information to determine whether containment is 
sufficient. If containment is not sufficient, information to determine 
the potential effects of the uncontained testing would be needed.
    Any human dietary exposure will require a tolerance, or exemption 
from requirement for a tolerance, regardless of the stage of product 
development or the size of the field tests. Even if the crop is 
destroyed, a tolerance, or exemption from requirement for a tolerance, 
may be necessary if there is genetic outcrossing to other crops that 
would be used as food or feed (See Unit V. of this document). 
Information addressing health effects thus may be necessary to allow a 
tolerance assessment.
    2. Sale or distribution. In order to assess the potential exposure 
when a plant-pesticide is sold or distributed, EPA will consider 
whether all varieties of a crop in which the plant-pesticide is 
registered (see Unit V.B. of this document) will be able to express the 
pesticidal substance. Potential exposure could also result if the 
plant-pesticide is expressed by plant relatives to which the genetic 
material encoding the plant-pesticide could be transferred. This 
analysis of exposure is a component in the possible identification of 
the types of nontarget organisms that associate with particular types 
of crops/plants and therefore may be exposed to the pesticidal 
substance. This type of analysis would be considered depending on the 
plant-pesticide/plant combination. In addition, sale or distribution 
will often involve human dietary exposure and may thus require 
information to address those risks and to support a tolerance or 
tolerance exemption.

C. Product Analysis

    Product analysis/characterization data and information are critical 
for assessing potential risks to humans and the environment. Product 
characterization embraces five basic areas: (1) Identification of the 
donor organisms and the nucleotide sequences that are inserted into the 
recipient plant; (2) identification and description of the vector or 
delivery system used to move the nucleotide sequences into the 
recipient plant; (3) identification of the recipient organism, 
including information on the insertion of the nucleotide sequences 
(e.g., stability of insertion); (4) chemical characterization of the 
plant pesticide products; and (5) data and information on the levels of 
the pesticidal substances in the recipient plant, including any tissue 
specificity of expression.

D. Environmental Fate Analysis

    1. Environmental fate risk issues. An accurate assessment of the 
fate, transport, and persistence of the pesticidal gene product 
(pesticidal substance) in the environment is essential to the entire 
risk assessment process. This information will determine if there is 
adequate containment during product development and will support the 
ecological non-target species and the health effects risk assessments 
for sale or distribution of the products. The following environmental 
fate risk issues are associated with field tests and sale or 
distribution of a plant-pesticide:
    (1) Increasing the ability of the modified plant to survive outside 
of cultivation through the introduction of a specific trait.
    (2) Gene capture and expression of the introduced trait by a wild 
or weedy relative.
    (3) Potential for a trait conferring a selective advantage to a 
plant in a natural plant community with the result of increasing the 
``weediness'' of that species.
    (4) Environmental fate of the pesticidal substance. The dosage to 
soils after plant senescence and incorporation into the soil, rate of 
degradation or dissipation and transport in the environment. Also 
whether or not the pesticidal substance is either exuded or volatilized 
from the plant during the growing season, resulting in a continuous 
application to the environment.
    The environmental fate of plant-pesticides introduced into the 
environment is composed of two facets: the movement of the gene 
encoding for the pesticidal substance (biological fate) and the fate of 
the pesticidal substance itself (chemical fate). Some of the 
information about the biology of the plant producing the plant-
pesticide may be available in the published literature and this 
information should be used to address the biological fate risk issues. 
There are several points of information that a producer should consider 
when developing a plant-pesticide for sale or distribution in 
agriculture. These points of information are presented in the Units 
VI.D.2. through VI.D.4. of this document. The points are arranged in a 
tiered framework that allows for resolution of the risk issue as early 
as possible in the review process.
    2. Biological fate analysis. The first consideration in the 
biological fate analysis is whether or not the plant producing the 
plant-pesticide can exist under other than cultivated conditions. This 
consideration results from risk issue (1) in Unit V.D.1. above.
    The next consideration is whether or not the plant producing the 
plant-pesticide has weedy or wild relatives and whether the relatives 
are distributed in or near the areas where the plant will be grown. 
This consideration results from risk issue (2) in Unit VI.D.1. above. 
If the plant producing the plant-pesticide does have weedy or wild 
relatives of concern, the next consideration is whether, based on its 
life cycle, pollinator requirements, and genetic limitations, it can 
take part in a successful outcrossing event.
    If the plant producing the plant-pesticide has the ability to 
outcross, the next considerations are its outcrossing rate and pollen 
longevity under laboratory conditions. If it is determined that either 
or both of these factors are high, outcrossing rates should be 
determined under field conditions. This consideration addresses risk 
issue (2) in Unit VI.D.1. above.
    If significant outcrossing is achieved under field conditions a 
determination of whether the plant-pesticide confers a selective 
advantage to the relative would be made. This consideration addresses 
risk issue (3) in Unit VI.D.1. above.
    3. Exposure to the pesticidal substance produced by the plant. If 
there is a toxicological concern for the plant pesticidal substance, an 
assessment of expression levels in all or some parts of the plant may 
be required. This consideration addresses risk issue (4) in Unit 
VI.D.1. above.
    4. Chemical fate analysis. Determining the persistence and movement 
of the pesticidal substance in the environment would be required if 
there is a toxicological concern for that pesticidal substance. Points 
that should be considered are the pesticidal substance's persistence 
and mobility in all environmental media (soil, water, and air). If the 
substance is persistent and another crop is grown in rotation with the 
transformed crop, a crop rotation study would be required. Similarly, 
if the pesticidal substance is stable in the environment and is 
expected to reach aquatic environments, a fish accumulation study would 
be required. This consideration addresses risk issue (4) in Unit 
VI.D.1. above.

E. Ecological Effects

    Sale or distribution of a plant-pesticide may ultimately lead to 
the plant-pesticide being expressed in the entire agricultural crop of 
the plant that is producing the plant-pesticide and in any other crops 
and/or relatives with which the plant can cross-breed. A careful 
analysis of all potential nontarget species (including threatened or 
endangered species) that may be susceptible to the pesticidal substance 
may thus be needed. Some plant/plant-pesticide combinations may not be 
acceptable for registration if use could be expected to result in 
unreasonable environmental effects if traditional risk mitigation 
restrictions are not appropriate for the plant-pesticide. In these 
cases, genetic limitations on the expression of the pesticidal 
substance in plants, or on the potential for gene transfer to other 
plants, may result in sufficient risk reduction to allow registration. 
Also, if the presence of the plant-pesticide is limited to the actual 
plant material, the number and kinds of species exposed may also be 
limited.
    EPA has, for traditional pesticides, relied on single species 
testing to evaluate potential effects on nontarget species, and this 
approach will continue to be of value. However, the standard test 
species and the standard acute exposure protocols used for chemical 
pesticides may not be sufficient to evaluate plant-pesticides due to 
their unique exposure scenario, e.g., the presence of the pesticidal 
substance as part of the plant and the potential for gene flow to other 
plants.
    In addition, the substance to be tested (whole plants, extracts of 
plants, pure pesticidal-substance, etc.) may have to be determined on a 
case-by-case basis, but the substance should be tested in an 
ecologically relevant manner. Unlike traditional chemical pesticides 
where direct contact is the predominant form of exposure, exposure to 
plant-pesticides will primarily be from ingestion of, or contact with, 
plant tissues that contain plant-pesticides.
    Traditional test protocols rely on a maximum-hazard dose to ensure 
that potential adverse effects are detected. However, it may be 
difficult to obtain a maximum-hazard dose using plant-pesticide 
materials if there are low expression levels of the pesticidal 
substance in the plant. Therefore, in some cases, chronic exposure 
testing may be more appropriate.
    The following points may be of particular value in performing an 
ecological risk assessment:
    (1) An analysis of which non-target species feed on, or contact, 
plant parts that will contain the pesticidal substance will be 
particularly helpful in identifying the species to be tested. 
Consideration should be given to which species are most representative 
of those likely exposed to the plant-pesticide.
    (2) If the pollen of a plant contains the pesticidal substance, the 
pesticide substance may be airborne beyond the immediate field 
location. The effect on aquatic invertebrates may need to be 
determined. Plant pollen enters the aquatic environment quite readily 
through wind movement. In addition, honeybees, and particularly 
honeybee larvae, are likely to be exposed to pollen. Honeybee larvae 
may be susceptible to these plant-pesticides, especially those intended 
to control insect larvae, even if the adult honeybees are resistant.
    (3) If the pesticidal substance is expressed in the seed or fruit, 
a different range of non-target organisms will be exposed than if the 
pesticidal substance is produced in the pollen. In this instance, 
possible effects on birds and mammals should be considered.
    (4) The duration of testing is a factor. If the pesticidal 
substance is expressed by the plant throughout the entire plant life-
cycle, some nontarget species may be exposed to a chronic dose of the 
substance as compared to traditional pesticide usage which often 
results in an acute dose. Chronic exposure in terms of the duration of 
expression of the plant-pesticide by the plant could be measured in 
terms relative to the life cycles of nontarget organisms likely to be 
exposed.
    (5) If, after harvest at the end of the growing season, the plant 
tissue containing the plant-pesticide is tilled into the soil or left 
in the field to decompose, soil organisms (i.e., Colembola and other 
soil arthropods, nematodes, mollusks, and annelids) may receive a low 
level chronic exposure, depending on the stability of the pesticidal 
substance. This may affect decomposition processes which occur 
naturally in the soil.
    (6) Threatened or endangered species may be at risk from widespread 
or uncontained use of the plant-pesticide. Since the pesticidal 
substance may have the potential to be expressed in the entire crop and 
related plants, it may be difficult to limit the exposure of threatened 
or endangered species. There is particular cause for concern if there 
are any threatened or endangered species related to the target species 
that feed on the plants or if there are any threatened or endangered 
species related to species susceptible to the pesticidal substance. 
Information on the feeding habits and preferred food sources of any 
potentially affected threatened or endangered species will be needed to 
address this issue.
    (7) Secondary feeding effects may increase the possibility of non-
target exposure, e.g., the possibility that species feeding on the 
plant would accumulate enough pesticide to affect predatory species 
feeding on them.
    (8) The possibility of transfer of a disease- or insect- resistant 
trait to wild or weedy relatives or the presence of the trait in the 
crop, itself, may create a weed or may increase the competitiveness of 
a known weed. There may be cause for concern if related, naturally-
occurring, pest-resistant plants are weeds, particularly if this 
particular pest-resistant trait is found in the most aggressive 
varieties of the weeds. If the pesticidal trait results in 
significantly greater pest-resistance, analysis of the competitiveness 
with naturally occurring weeds may be a practical way to address this 
issue.

F. Human Health Affects

    Plant-pesticides are likely to present a limited exposure of the 
pesticidal substance to humans. In most cases, the predominant, if not 
the only, exposure route of concern will be dietary. Significant 
respiratory and dermal exposures would be unlikely. A full assessment 
to be made from a specific, limited, data set (as compared to 
traditional pesticides) can thus be made. Information on the presence 
of the pesticidal substance in edible portions of the crop will help 
determine the degree of human dietary exposure. In cases where the 
plant-pesticide containing crop is used as animal feed, domestic animal 
safety information may be necessary.
    The types and numbers of mammalian toxicology studies needed for 
human health effects risk assessment will depend on whether the plant 
pesticide is a protein; and, if not, whether the plant pesticide is 
analogous to a traditional chemical pesticide or a biochemical 
pesticide.
    For plant-pesticides that are proteins, an acute oral toxicity 
study in the rodent may be sufficient to address health issues and/or 
questions. Although not yet validated, an in vitro digestibility assay 
may answer questions about allergenicity or about the potential for 
toxicity of proteins including those with a deliberately altered amino 
acid sequence.
    Naturally occurring non-proteinaceous plant pesticides with a non-
toxic mode of action against the target pest could be addressed in a 
manner analogous to biochemical pesticides (although the Agency is 
proposing to exempt this category of plant-pesticide under FIFRA, they 
would not necessarily be exempt under FFDCA). Non-proteinaceous plant-
pesticides with a toxic mode of action could be addressed as for 
traditional chemical pesticides. The focus on data requirements for 
non-proteinaceous plant-pesticides will also primarily be on the oral 
route of exposure.

G. Development of Resistance to Plant-pesticides

    EPA recognizes that there is a potential for the development of 
resistance to plant-pesticides. At present, the issue has been raised 
particularly in the case of the Bacillus thuringiensis insecticidal 
delta-endotoxin. Field resistance to the delta-endotoxin has been 
documented for foliar applications of the microbial pesticide, Bacillus 
thuringiensis (Ref. 7). It is postulated that resistance to the delta-
endotoxin could develop when it is produced by plants. The development 
of insect resistance to the delta-endotoxin could lead to a loss in the 
effectiveness of this valuable pesticide. EPA is committed to the 
development of pesticides that are viable alternatives to more toxic 
and persistent chemical pesticides. The Agency is considering how it 
can best encourage the development of agricultural practices that will 
minimize the development of resistance to plant-pesticides. Toward this 
end, the Agency has begun to analyze the regulatory and nonregulatory 
tools it could use to address resistance to all pesticides, including 
plant-pesticides.

VII. Interactions With Other Agencies

    EPA is the Federal agency primarily responsible for the regulation 
of pesticides. However, in fulfilling this mission EPA works closely 
with the U.S. Department of Agriculture (USDA) which has 
responsibilities under the Plant Pest Act and the Plant Quarantine Act 
and the U.S. Food and Drug Administration (FDA) which has 
responsibilities under FFDCA.

A. USDA

    USDA has authority to prevent the introduction and dissemination of 
plant pests under the Plant Pest Act and the Plant Quarantine Act. An 
introduction at any acreage of a plant that is under the jurisdiction 
of the Plant Pest Act requires either that a permit be obtained from 
USDA's Animal and Plant Health Inspection Service (USDA/APHIS) or for 
certain plants that a notification be submitted to USDA/APHIS, unless 
it has been exempted from those requirements.
    EPA and USDA/APHIS have consulted and exchanged information on 
plants and plant-pesticides and intend to continue to do so in the 
coordination of their regulatory activities. The two agencies also have 
and intend to continue to consult closely on scientific issues related 
to the safety considerations associated with the environmental impact 
of field tests of plant-pesticides.

B. FDA

    Pursuant to FFDCA and the reorganization that created EPA, 
pesticides as defined by FIFRA are subject to EPA's regulatory 
authority under FFDCA. However, FDA's authority under FFDCA extends to 
any nonpesticidal substance that may be introduced into a new plant 
variety and that is expected to become a component of food.
    FDA and EPA have and intend to continue to consult closely on any 
jurisdictional questions, as well as on scientific matters where 
consultation will be helpful in resolving safety questions. Both 
agencies agree that EPA will address under its regulatory jurisdiction 
the food safety issues associated with the plant-pesticide, including 
marker genes used to confirm the presence of the DNA necessary for the 
production of the pesticidal substance. Any food safety questions 
beyond those associated with the plant-pesticide, such as those 
involving changes to food quality or raised by unexpected or unintended 
compositional changes, are under FDA's jurisdiction (Ref. 3). 
Similarly, food safety issues associated with alterations in levels of 
a substance with pesticidal properties, or the appearance of a 
substance with pesticidal properties, that occur as an unintended 
consequence of modifications to a non-pesticidal trait would also fall 
under FDA's authority..

VIII. External Review

    In developing its approach to regulating plant-pesticides, EPA has 
requested the advice of two scientific advisory committees in three 
meetings. On December 19, 1992, pursuant to section 25 of FIFRA, a 
Subpanel of the FIFRA SAP was convened to review a draft policy 
statement on plant-pesticides and respond to a series of scientific 
questions posed by the Agency primarily on EPA's approach under FIFRA. 
On July 13, 1993, a Subcommittee of the EPA Biotechnology Science 
Advisory Committee (BSAC) was convened to address a series of 
scientific questions primarily on EPA's approach under FFDCA. On 
January 21, 1994, a joint meeting of the SAP/BSAC Subpanel on plant-
pesticides was held. The issues raised at these meetings are discussed 
below, together with the Agency's response. (Full reports from these 
meetings are available in the public docket.)

A. Substances New to the Plant

    Questions on how best to describe ``substances that are new to the 
plant'' were posed at all three science advisory meetings. At its 
December 1992 meeting, the FIFRA Subpanel was asked whether the 
taxonomic demarcation of ``genus'' was appropriate, or whether some 
other demarcation would be more appropriate. The Subpanel expressed 
concern over an exemption based on a taxonomic definition and suggested 
the Agency evaluate a series of considerations involving the potential 
for quantitative and qualitative differences in exposure to a plant-
pesticide.
    The SAP Subpanel suggested that the Agency would ``need to create a 
workable balance between effective regulatory oversight and 
encouragement of the development of plant-produced pesticides.''
    At its July 13, 1993 meeting, the BSAC Subcommittee addressed a 
related issue with regard to the regulation of plant-pesticides under 
the FFDCA and human dietary exposures to plant pesticides.
    Included in questions to the Subcommittee were queries on the 
availability of information on current levels of exposure in the diet 
to plant-pesticides in raw agricultural commodities and on which plant-
pesticides might be of concern should their levels be significantly 
increased.
    The BSAC Subcommittee in their report stated that no formal, 
complete data base for such information exists. Rather most of this 
knowledge is part of breeders' experience, with breeders depending 
primarily on familiarity with food crops (e.g., knowledge of which crop 
plants have the ability to produce which toxicants) to ensure consumers 
are not exposed to deleterious levels of such substances. In general, 
little information exists on the range of levels of plant-pesticides in 
plants, including ranges within the most studied grouping, food plants. 
The mechanisms through which plants display resistance to pests, 
moreover, have not been worked out very well. Based on experience, 
however, the BSAC Subcommittee suggested EPA consider a scheme based on 
sexual compatibility to identify those groupings wherein plant-
pesticides might present new and novel dietary exposures and those that 
would not.
    The use of sexual compatibility and/or taxonomy as a standard for 
the potential for significantly different environmental exposures was 
discussed at the January 21, 1994, joint SAP/BSAC Subpanel meeting. The 
joint Subpanel members were supplied with the reports of the previous 
meetings and drafts of proposals analyzing the strengths and weaknesses 
of approaches based on sexual compatibility and/or taxonomy. In 
response to the question of whether plants in a sexually compatible 
population are likely to share substances or traits, the joint Subpanel 
agreed that sexually compatible plants are more likely to have a common 
constitution than unrelated plants and thus are less likely to lead to 
novel exposures.
    The report of the January 21, 1994, joint SAP/BSAC Subpanel 
meeting, indicates that the joint Subpanel agrees that both the concept 
of sexual compatibility and the concept of taxonomy should restrict the 
occurrence of significantly different exposures and finds that Option 3 
is a reasonable approach for agricultural plants. However, the Subpanel 
questioned whether under the taxonomic criterion of Option 3 (and 
Option 2) a low probability of novel exposures can be extended to wild 
or semi-wild plants. For these types of plants, the genus standard may 
result in the exemption from regulation of plant-pesticides that may 
present novel exposures.
    The Agency also included a question, at the January 21, 1994, joint 
BSAC/SAP meeting, concerning an approach using a criterion based on the 
process used to modify the plant, e.g., recombinant DNA methodologies. 
As described in the report of the joint BSAC/SAP Subpanel meeting, if 
the Agency were to use this approach, it would first exempt plant-
pesticides developed through techniques other than those of modern 
biotechnology from its regulatory scope. For those plant-pesticides 
that are not exempted because they were developed through techniques of 
modern biotechnology, the exemptions proposed by the Agency would apply 
(see Units IV.B. and IV.C. of this document).
    The joint Subpanel supported the inclusion of a criterion based on 
methodologies such as rDNA as a rational approach to making the first 
cut as to which plant-pesticides would be regulated. However, the joint 
Subpanel cautioned that further exemptions such as those proposed by 
EPA should be used in conjunction with the criterion based on 
methodology. In addition, the joint Subpanel recommended that the 
Agency define methodologies in a way that clearly delineates to the 
scientific community and the public what is and is not included in the 
regulatory scope, based on current state-of-the-science.
    EPA Response: The Agency has chosen to propose to use under both 
statutes, an approach based on sexual compatibility. First, this 
approach would exempt under both FIFRA and FFDCA, plant-pesticides 
having a high probability of being derived from plants having high 
numbers of genes in common. Under such circumstances, the likelihood of 
new or novel exposures both to the environment and in terms of human 
consumption is low.
    Second, use of the standard of sexual compatibility will allow EPA 
to use its authorities under FIFRA and FFDCA in concert to regulate 
plant-pesticides, and thus to utilize, to the extent possible in light 
of the different statutory standards, similar approaches to oversight 
under each of the two statutes.
    Third, the Agency believes that its proposed approach would be 
consistent with the SAP Subpanel's concern that EPA ``. . .create a 
workable balance between effective regulatory oversight and 
encouragement of the development of plant-produced pesticides.'' Under 
the preferred approach, novel exposures are not likely to occur with 
plant-pesticides exchanged between plants that are sexually compatible 
(See also Unit V. of this document for additional discussion).
    With regard to the advice of the January 21, 1994, joint SAP/BSAC 
Subpanel concerning the use of a process-based criterion in the scope, 
if the Agency were to use this approach, plant-pesticides developed 
through techniques other than those involving in vitro manipulation of 
genetic material would be exempt. In order to meet the recommendations 
of the joint Subpanel, the Agency would define this category of plant-
pesticides in the following way: The genetic material that encodes for 
the pesticidal substance or leads to the production of the pesticidal 
substance is extracted from an organism and introduced into the genome 
of the recipient plant or is synthesized in vitro and introduced into 
the genome of the recipient plant. The exemptions proposed by the 
Agency in Unit IV. of this document would be used in concert with this 
criterion. The Agency believes this approach would meet the 
recommendations of the SAP/BSAC joint Subpanel. The Agency is 
soliciting comment on this approach (see the proposal published 
elsewhere in today's issue of the Federal Register entitled ``Plant-
pesticides Subject to the Federal Insecticide, Fungicide, and 
Rodenticide Act; Proposed Rule'').

B. Plant-pesticides That Act Primarily by Affecting the Plant

    The SAP Subpanel at its December 1992 meeting considered whether 
EPA's language clearly and sufficiently identified plant resistance 
mechanisms that do not involve substances whose mode of action produces 
a direct toxic effect on the pest. The SAP Subpanel stated that for the 
most part the language EPA was proposing was clear and appropriately 
identified plant resistance mechanisms whose mode of action was not 
directly toxic. The Subpanel noted, however, that the issue of 
resistance to toxins produced by the pests was not addressed by that 
language. The Subpanel recommended insertion of the following statement 
into EPA's proposed language: ``Acts in the host plant to produce 
target(s) of the toxin that are resistant to the toxin's deleterious 
action.''
    EPA Response: EPA accepted this recommendation and modified the 
language of its approach to incorporate the issue of resistance to 
toxins.

C. Viral Coat Proteins

    1. FIFRA. The December 18, 1992, SAP Subpanel meeting and the 
January 21, 1994, joint SAP/BSAC Subpanel meeting addressed the use of 
viral coat protein genes to modify plants to protect the plant from 
damage from viral infection. In the discussion at the December 18, 
1992, SAP Subpanel meeting, several risk considerations were identified 
and the probability of occurrence of each addressed in the SAP Subpanel 
report. The SAP Subpanel report stated that the probability of 
occurrence of the risks examined is very low (see Unit IV.B.1.c.iii. of 
this document and the proposal published elsewhere in todays issue of 
the Federal Register entitled, ``Plant-pesticides Subject to the 
Federal Insecticide, Fungicide, and Rodenticide Act; Proposed Rule,'' 
for a more complete discussion). The January 21, 1994, joint SAP/BSAC 
Subpanel meeting discussed the alternative option for the exemption of 
viral coat proteins from FIFRA regulation. The joint Subpanel 
reiterated the statement of the December 18, 1992, SAP Subpanel report 
that the potential risks associated with the use of vcp-mediated 
resistance coat proteins are low. The SAP/BSAC joint Subpanel, at the 
January 21, 1994 meeting, did not support the inclusion of the 
alternative partial exemption option. Unit IV.B.1.c. of this document 
describes how the SAP and joint SAP/BSAC discussion of vcp-mediated 
resistance viral coat proteins supplements and influences EPA's 
analysis.
    EPA Response: EPA agrees that the probability of risks from the 
introduction of viral coat protein genes into plant genomes is low, and 
as its preferred option proposes to exempt these plant-pesticides from 
FIFRA oversight. Because of public comments received at the December 
18, 1992, SAP Subpanel meeting and the January 21, 1994, joint SAP/BSAC 
meeting, however, concerning viral coat proteins and selective 
advantage to wild relatives of managed plants, EPA is offering for 
comment in this document the alternative approach to viral coat 
proteins to allow the fullest discussion possible.
    2. FFDCA. The December 18, 1992, SAP Subpanel also addressed the 
question of whether viral coat proteins might present a dietary risk. 
It stated that ``[s]ince viruses are ubiquitous in the agricultural 
environment at levels higher than will be present in transgenic plants, 
and there has been a long history of `contamination' of the food supply 
by virus coat protein, there is scientific rational for exempting 
transgenic plants expressing virus coat protein from the require of a 
tolerance.''
    EPA Response: EPA agrees with this position and is proposing an 
exemption from the requirement of a tolerance for viral coat proteins.

D. Nucleic Acids

    The BSAC Subcommittee on July 13, 1993, confirmed that nucleic 
acids (DNA and RNA), are present in the cells of every living organism, 
including plants, microorganisms and animals, used for food, and do not 
raise safety concerns as a component of food.
    EPA Response: EPA agrees with the BSAC Subcommittee and is 
proposing to exempt nucleic acids (DNA and RNA) as a class from the 
requirement of a tolerance under FFDCA.

E. Pest Resistance

    The SAP Subpanel on December 18, 1992, addressed the issue of 
development of pest resistance to a pesticidal substance produced by 
plants. The Subpanel stated that ``[d]elaying the evolution of 
resistance is clearly important if we are to realize the full potential 
of effective and safe crop protection that could be obtained by 
application of biotechnology.'' The SAP Subpanel urged EPA to actively 
assess the problem of pesticide resistance, especially when the 
pesticide is part of the progression toward use of ``safer 
pesticides.''
    EPA Response: EPA agrees that the issue of the development of 
resistance is an important one and is examining what it might do to 
address this issue under its regulatory authorities.

F. Allergenicity

    The December 18, 1992, SAP Subpanel, in its discussion of EPA's 
approach for oversight of plant-pesticides under FFDCA, raised the 
question of the potential allergenicity of certain food components.
    EPA Response: EPA is aware that there are food components that can 
induce food allergies and that the issue of allergenicity in novel 
foods is important. EPA hosted, along with FDA and USDA/APHIS, a 
scientific conference on food allergens on April 18-19, 1994. The 
agencies are now reviewing the discussions that were held at the 
conference.

G. In Vitro Digestibility Assay

    EPA asked the July 13, 1993, BSAC Subcommittee to consider the 
scientific merits and limitations of an in vitro digestibility assay to 
predict toxicity from dietary exposure to proteinaceous transgenic 
plant pesticides. This in vitro digestibility assay would use features 
of the test, 701- Disintegration for Plain Coated Tablets or Enteric-
coated Tablets, of the US Pharmacopeia.
    The Subcommittee responded to the following specific questions on 
the in vitro digestibility assay. What are the appropriate assay 
endpoints (i.e., free amino acids, peptides) to conclude that a protein 
is digested to harmless components? What are the best methods to 
quantify digestibility? With plant associated proteinaceous pesticides, 
what is the most relevant test material? How predictive is the in vitro 
assay for addressing the variations in luminal adsorption that occur 
with maturation, (i.e., infant versus adult versus geriatric uptake)? 
What is the significance of digestive disorders like hypochlorhydria 
and enzyme deficiencies with respect to the in vitro digestibility 
assay and interpretation of assay results? To what extent can the amino 
acid sequence of known toxic proteins and known sequences of non-toxic 
proteins be used to predict toxicity or lack of toxicity in other 
proteins? Would fragments of non-toxic proteins also be expected to be 
non-toxic?
    In response, the Subcommittee indicated that it could not endorse 
use of an in vitro digestibility assay as the sole test for determining 
toxicity. An in vitro digestibility assay might provide useful 
information if employed in conjunction with other tests.
    The Subcommittee suggested that if an in vitro digestibility test 
assay is to be used as part of a more comprehensive toxicological 
evaluation, the test should utilize a range of gastric and intestinal 
phases in order to address variations in luminal absorption that occur 
with maturation.
    In response to the question concerning the appropriate test 
material, the Subcommittee suggested the test be conducted with 
purified plant-pesticide. Purified plant-pesticide may be tested in the 
presence or absence of standardized mixtures, (e.g., protein/agar). The 
standardized mixture would approximate the condition of a plant-
pesticide ingested in food.
    With regard to the question concerning how to best quantify 
digestibility, the Subcommittee suggested a Western blot could be used 
to determine the degree of digestion of the plant-pesticide.
    In response to the question on what constitutes the appropriate 
test material and assay endpoint and the question of whether fragments 
of non-toxic proteins could be expected to be non-toxic, the 
Subcommittee noted that partially digested proteins and peptides may be 
toxic even if the parent protein is not. They suggested it may 
therefore be difficult to declare anything short of complete digestion 
to amino acids as non-toxic.
    The Subcommittee suggested the Agency consider exploring other 
methods of addressing the issue of toxicity from dietary exposure, 
including: in vitro cell and culture systems to address questions of 
digestibility, translocation/transport across the lumen, and binding to 
cells of the GI tract; feeding studies involving whole animal systems, 
specifically those involving ``failure to thrive'' assays; cell binding 
assays to test for the presence of receptors for toxins; an in vivo 
assay using brine shrimp. The Subcommittee cautioned, however, that the 
brine shrimp or other nonmammalian test systems need to be evaluated 
carefully to ensure that the data can be extrapolated to mammalian 
species and to determine that the test responds appropriately to 
proteinaceous plant-pesticides. The Subcommittee indicated that EPA may 
have to rely on a case-by-case approach to assessing toxicity for the 
near future.
    EPA Response: EPA is exploring the suggestions of the BSAC 
Subcommittee as it develops its data requirements for plant pesticides.

H. Points to Consider for Data Needs

    EPA asked the January 21, 1994, joint SAP/BSAC Subpanel to consider 
the ``points to consider'' EPA had developed for plant-pesticides in 
terms of product analysis, ecological effects, environmental fate, and 
human health effects.
    1. Product analysis. In terms of product analysis, the joint 
Subpanel endorsed the Agency position that the evaluation of the risks 
posed by plant-pesticides demands that methods be available for product 
analysis and characterization of plant-pesticides in terms of: (1) 
Identification of the donor organisms and gene sequences that are 
inserted into the recipient plant; (2) identification and description 
of the vector or gene delivery systems; (3) identification of the 
recipient organisms, including information on the insertion of the gene 
sequences; (4) chemical characterization of the plant-pesticide 
products; and (5) quantification of the plant-pesticides in recipient 
plants.
    The joint Subpanel suggested some additional information may be 
useful: (1) Are the gene sequences stable in the recipient organism or 
are they prone to deletion or mutation?; (2) The chemical 
characterization of the plant-pesticides should be equally rigorous 
whether the products are proteinaceous or nonproteinaceous; (3) Methods 
for quantification of plant-pesticides are essential for assessing 
exposure to target and nontarget organisms, for assessing human 
exposure, for determining environmental fate, transport, and 
persistence, and for determining the distribution (i.e., in edible and 
nonedible portions) of plant-pesticides within plants.
    2. Ecological effects. In terms of ecological effects, the joint 
Subpanel stated that EPA in its proposed policy statement addresses the 
major issues of concern in assessing the ecological risks associated 
with the testing and commercialization of plant-pesticide products. 
They suggested, however, that there are several points that could be 
expanded to better address potential ecological risks. These are:
    (1) Substances to be tested: While a case-by-case determination is 
appropriate, it is important that the substance be tested in an 
ecologically relevant form under an ecologically relevant protocol. 
Unlike conventional pesticides for which direct contact with the 
pesticide is a primary means of exposure, exposure to plant-pesticides 
will be primarily through ingestion of, or contact with, pesticide-
containing plant tissues. Experience with bioassays of plant defensive 
chemicals has shown the plant milieu, and the method of exposure, may 
dramatically modify (enhance or ameliorate) the resultant effects of 
the chemical. Consequently, test substances and test protocols that 
lack ecological relevance may provide results that are inappropriate as 
a basis for regulatory decisions. The joint Subpanel recommended that 
EPA develop a series of guidelines for identifying ecologically 
relevant test substances and develop ecologically relevant test 
protocols.
    (2) The joint Subpanel noted that EPA identifies expression of the 
plant-pesticide by the plant throughout the entire plant life-cycle as 
a consideration in determining the potential for chronic exposure on 
nontarget species to the plant-pesticide. Because plant life-cycles can 
vary in duration, and because exposure of nontarget species to a plant-
pesticide for periods of shorter duration than a plant's life-cycle 
could have potentially significant ecological effects, expression of 
the pesticidal substance throughout the plant's life cycle seems to be 
a criterion that is not necessarily relevant in an ecological context. 
Although recognizing that it would lack the regulatory consistency of 
the plant life-cycle criterion, the joint Subpanel concluded that a 
criterion based on expression of the plant- pesticide for ``prolonged 
periods'' would seem to have greater ecological relevance. The joint 
Subpanel also stated that the meaning of ``prolonged period'' would 
have to be determined on a case-by-case basis and would depend on the 
types of nontarget organisms that are likely to be affected. The joint 
Subpanel recommended that EPA develop guidelines that define the 
potential for chronic exposure in terms of the duration of expression 
of the pesticide substance by a plant, relative to the life cycles of 
nontarget organisms likely to be exposed.
    (3) The joint Subpanel also commented on triggers for requiring 
EUPs for plant-pesticides during the testing stage, noting that the 
potential for nontarget exposure through gene capture by wild relatives 
will increase as the number of test sites increases, in areas where 
wild relatives may occur. Accordingly, the joint Subpanel recommended 
that EPA give careful consideration to addressing this potential for 
increased ecological risk in those situations in which there is 
potential for gene capture by wild plants when establishing an EUP 
trigger.
    3. Human health effects. In terms of human health effects, the 
joint Subpanel noted that nontarget populations at potential risk from 
exposure to plant-pesticides may include humans and data requirements 
should take into account the diverse nontarget subpopulations that 
could be potentially exposed or at risk. Examples of specific 
subpopulations at potential risk may include individuals who are high 
consumers (e.g., vegetarians; children with high intake to body weight 
ratios; over eaters), individuals with compromised digestive systems 
(e.g., elderly individuals; persons receiving treatment for digestive 
disorders or diseases).
    4. EPA response. EPA agrees with the joint Subpanel's suggestions 
and has considered them in the points to consider and will consider 
them as the Agency develops its data requirements for plant-pesticides.

IX. Request for Comment

A. Scope of Coverage for the Regulatory Scheme for Plant-Pesticides 
under FIFRA and FFDCA

    1. FIFRA exemptions: Requests for comment on the proposed 
exemptions under FIFRA and the options for these exemptions can be 
found in the Federal Register document entitled: ``Plant-pesticide 
Subject to the Federal Insecticide, Fungicide, and Rodenticide Act; 
Proposed Rule.'' All comments relating to the proposed exemptions under 
FIFRA should be submitted in the context of this proposed regulation 
and should be identified by the docket control number OPP-300369.
    2. FFDCA exemptions: Requests for comment on EPA's proposed 
exemption from the requirement of a tolerance for categories of plant-
pesticides that will not result in significantly different dietary 
exposure can be found in the Federal Register document entitled, 
``Plant-pesticides; Proposed Exemption from the Requirement of a 
Tolerance Under the Federal Food, Drug, and Cosmetic Act'' and should 
be identified by the docket control number OPP-300368. All comments 
relating to this proposed exemption from the requirement of a tolerance 
should be submitted in the context of that proposed regulation. 
Similarly, all comments relating to the other two proposed exemptions 
from the requirement of a tolerance should be submitted in the context 
of these proposed regulations and should be submitted under the docket 
control numbers OPP-300367 and OPP-300371 (see the proposals published 
elsewhere in todays issue of the Federal Register entitled ``Plant-
pesticides; Proposed Exemption from the Requirement of a Tolerance for 
Viral Coat Proteins in Plants'' and ``Plant-pesticides; Proposed 
Exemption from the Requirement of a Tolerance For Nucleic Acids 
Produced in Plants.''

B. EUP Triggers for Field Testing

    EPA has described in this Federal Register document one preferred 
approach and four alternative approaches for identifying when producers 
would be required to obtain an EUP. Commenters are advised that the 
Agency may choose to implement any of these five approaches or some 
combination thereof. EPA recognizes that each of these five approaches 
has advantages and disadvantages. EPA is soliciting comment on these 
approaches. Commenters in stating a preference for an approach are 
asked to describe the factors weighing most heavily in forming their 
preference.

X. Economic Analysis

    The regulatory impact analysis (RIA) evaluates the costs and 
benefits of amending the Code of Federal Regulations to allow for the 
regulation or exemption of specific types of plant-pesticides under 
FIFRA (40 CFR 152.20 and 40 CFR part 174). This report is intended to 
meet the requirements for a RIA as established by Executive Order No. 
12866, the Regulatory Flexibility Act, and section 25 of FIFRA.
    The RIA presents the alternative regulatory options and the costs 
that were considered by the Agency under FIFRA, including two options 
that were considered by the Agency but not included in the proposed 
rule. Four possible approaches to the regulation of plant-pesticides 
under FIFRA were evaluated in the RIA that allowed for varying degrees 
of regulatory coverage. RIA Option 1 is the most limited alternative in 
regulatory scope. RIA Option 2 represents EPA's proposed, and 
preferred, regulatory scope and is broader in coverage than RIA Option 
1. In addition to those plant-pesticides regulated under RIA Option 2, 
RIA Option 3's scope would include viral coat proteins used as plant-
pesticides. Finally, under RIA Option 4, all plant-pesticides, 
including those that result from traditional plant breeding, would be 
subject to the requirements of FIFRA. The costs for RIA Option 2 (the 
Agency's preferred option) are substantially lower than RIA Option 4. 
Refer to the proposal published elsewhere in todays issue of the 
Federal Register entitled ``Plant-Pesticides Subject to the Federal 
Insecticide, Fungicide, and Rodenticide Act; Proposed Rule'' for a 
summary of the RIA. The entire text of the RIA can be found in the 
public docket for this proposed rule (OPP-300370).

XI. References

    (1) EPA issue paper. FIFRA: Benefit and environmental risk 
considerations for inherent plant-pesticides.
    (2) EPA issue paper. Issues associated with the regulation of viral 
coat proteins under FIFRA and FFDCA.
    (3) Food and Drug Administration. 1992. Foods derived from new 
plant varieties. (57 FR 22984).
    (4) Klaasen, C.D., M.O. Amdur, and J.D. Doull. 1986. Casarett and 
Doull's Toxicology: The Basic Science of Poisons. Third Edition. 
Macmillan Publishing Company, New York.
    (5) International Food Biotechnology Council, 1990. Biotechnologies 
and food: Assuring the safety of foods produced by genetic 
modification. In: Regulatory Toxicology and Pharmacology. Vol. 12. 
Academic Press, New York.
    (6) Lamb, C.J., J.A. Ryals, E.R. Ward, and R.A. Dixon. 1992. 
Emerging strategies for enhancing crop resistance to microbial 
pathogens. Bio/Technology. 10:1436-1445.
    (7) Tabashnick, B.E., M.L. Cushing, N. Finson and M.W. Johnson. 
1990. Field development of resistance to Bacillus thuringiensis in 
diamondback moth (Lepidoptera: Plutellidae). Journal of Economic 
Entomology. 83:1671-1676.
    (8) Tolin, S.A. 1991. Persistence, establishment, and mitigation of 
phytopathogenic viruses. In: Risk Assessment in Genetic Engineering. 
Edited by M.A. Levin and H.S. Strauss. McGraw-Hill, Inc., New York. pp 
114-139.

XII. Regulatory Assessment Requirements

A. Executive Order 12866

    Under Executive Order 12866 (58 FR 51735, October 4, 1993), the 
Agency must determine whether the regulatory action is ``significant'' 
and therefore subject to review by the Office of Management and Budget 
(OMB) and the requirements of the Executive Order. Under section 3(f), 
the order defines a ``significant regulatory action'' as an action that 
is likely to result in a rule (1) Having an annual effect on the 
economy of $100 million or more, or adversely and materially affecting 
a sector of the economy, productivity, competition, jobs, the 
environment, public health or safety, or State, local, or tribal 
governments or communities (also referred to as ``economically 
significant''); (2) creating serious inconsistency or otherwise 
interfering with an action taken or planned by another agency; (3) 
materially altering the budgetary impacts of entitlement, grants, user 
fees, or loan programs or the rights and obligations of recipients 
thereof; or (4) raising novel legal or policy issues arising out of 
legal mandates, the President's priorities, or the principles set forth 
in this Executive Order. Pursuant to the terms of this Executive Order, 
it has been determined that this proposed policy statement is a 
``significant regulatory action'' because it raises novel policy issues 
arising out of FIFRA legal mandates. As such, this action has been 
submitted to OMB for review, and any comments or changes made in 
response to OMB suggestions or recommendations, will be documented in 
the public record.

B. Regulatory Flexibility Act

    This proposed policy statement was reviewed under the provisions of 
section 3(a) Regulatory Flexibility Act (RFA) [5 U.S.C. 605(b)]. The 
RFA requires that agencies take special note of the impact of proposed 
regulations on small entities. Analysis requirements under the RFA can 
and should be combined with the analysis required under Executive Order 
12866.
    The regulatory flexibility analysis of this proposed policy for 
plant-pesticides on small entities is demonstrated within the 
structuring of the four regulatory options proposed. These options were 
considered after extensive evaluations of the benefit/risk tradeoffs 
between option cost and risk reduction provided. The Agency has 
structured the resulting options from a narrow regulatory scope (RIA 
Option 1) to a broad regulatory scope (RIA Option 4) and as such, has 
conducted an ``inherent'' sensitivity analysis for small firms likely 
to be affected by this regulation. The Agency has determined that the 
tradeoffs between the benefits and risks of the proposed regulations 
are optimized under RIA Option 2, EPA's proposed scope.

C. Paperwork Reduction Act

    The information collection requirements in this proposed policy 
statement have been submitted for approval to the Office of Management 
and Budget (OMB) under the Paperwork Reduction Act, 44 U.S.C. 3501 et 
seq. An Information Collection Request document has been prepared by 
EPA (ICR No. 1693.01) and a copy may be obtained from Sandy Farmer, 
Information Policy Branch, (Mail Code 2136), Environmental Protection 
Agency, 401 M St., SW., Washington, DC 20460, or by calling (202) 260-
2740.
    This collection of information has an estimated reporting burden 
averaging 1,143 hours per response and an estimated annual 
recordkeeping burden averaging 74 hours per respondent. These estimates 
include time for reviewing instructions, searching existing data 
sources, gathering and maintaining the data needed, and completing and 
reviewing the collection of information.
    Send comments regarding the burden estimate or any other aspect of 
this collection of information, including suggestions for reducing this 
burden to Chief, Information Policy Branch, (Mail Code 2136), 
Environmental Protection Agency, 401 M St., SW., Washington, DC 20460; 
and to the Office of Information and Regulatory Affairs, Office of 
Management and Budget, Washington, DC 20503, marked ``Attention: Desk 
Officer for EPA.'' The final rule will respond to any OMB or public 
comments on the information collection requirements contained in this 
proposed policy statement.

List of Subjects

    Environmental protection, Biotechnology, Labeling, Plant-
pesticides, Plants.

    Dated: November 15, 1994.
Carol M. Browner,
Administrator.
[FR Doc. 94-28821 Filed 11-22-94; 8:45 am]
BILLING CODE 6560-50-F