United States
Environmental Protection
Agency
Environmental Research
Laboratory
Corvallis, OR 97333
EPA/600/9-91/041
October 1991
Research and Development
PA Plant Tier Testina:
A Workshop to Evaluate
Nontarget Plant Testing
in Subdivision J
pesticide Guidelines
-------
PLANT TIER TESTING: A WORKSHOP TO EVALUATE
NONTARGET PLANT TESTING IN SUBDIVISION J
PESTICIDE GUIDELINES
29 November - 1 December, 1990
Corvallis, Oregon, USA
Sponsored by the
US ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Environmental Research Laboratory
Corvallis, Oregon
Workshop Coordination and Report
Prepared by
John Fletcher and Hilman Ratsch
US EPA Environmental Research Laboratory
200 S.W. 35th Street
Corvallis, Oregon 97333
October 1, 1991
-------
TABLE OF CONTENTS
EXECUTIVE SUMMARY iv
VALUE OF WORKSHOP PROCEEDINGS vi i
DISCLAIMER ix
ACKNOWLEDGEMENTS ix
PURPOSE OF THE WORKSHOP x
ORGANIZATION AND OPERATION OF THE WORKSHOP xi
PRESENTATIONS 1
Session I: Inception and Implementation of Subdivision J i
Pesticide Phytotoxicity Testing: A Historical Perspective 2
Robert W. Hoist, Environmental Fate and Effects,
OPP, EPA, Washington, D.C.
Plant Data Analysis by Ecological Effects Branch in
the Office of Pesticides Program 6
Charles Lewis and Richard Petrie, Ecological Effects, OPP,
EPA, Washington, D.C.
Session II: Ecological and Taxonomic Considerations 16
Role of Biotic and Abiotic Factors in Plant Growth and
Biodiversity 17
George E. Taylor, Jr., Desert Research Institute, Reno, NV
Assessment of Published Literature Concerning Pesticide
Inf1uence on Nontarget PI ants 28
John S. Fletcher, University of Oklahoma, Norman, OK
GIS-Based Risk Assessment: Applications from an Approach
to Ozone Risk Assessment to Assessing Risk of Pesticides
to Nontarget Organisms and Ecosystems 37
Bill Hogsett, ERL-C, EPA, Corvallis, OR
Session III: Laboratory and Greenhouse Testing (Tiers I and II,
Subdivision J) 47
Difficulties in Performing Existing Tier I and II Tests
in Subdivision J Guidelines 48
Joseph W. Gorsuch, Environmental Sciences Section, Eastman
Kodak Company, Rochester, NY.
Development of Nontarget Plant Test Methods at ICI
Agrochemicals 58
-------
Richard A. Brown, Deborah Farmer and Lorraine Canning
ICI Agrochemicals, Bracknell, Berkshire, England
Tissue Culture Tests for Studying Phytotoxicity and
Metabolic Fate of Pesticides and Xenobiotics in Plants 70
Hans Harms and Elke Kottutz, Institute of Plant Nutrition
and Soil Science, Braunschweig, Germany
Plant Reproduction and/or Life Cycle Testing 80
Hilman Ratsch and John Fletcher, ERL-C, EPA, Corvallis, OR
and University of Oklahoma, Norman, OK
Session IV: Field Testing (Tier III, Subdivision J) 90
Investigating Herbicide Sensitivity Thresholds 91
Robert Callihan, University of Idaho, Moscow, ID
Research Report on 1988 Potato-Herbicide Injury
Research 98
Philip Westra, Gary Franc, Brian Cranmer and Tim D'Amato
Colorado State University, Ft. Collins, CO
Symptom Expression with Selected Herbicides on Four
Perennial Plant Species 105
Robert Parker, Washington State University, Prosser, VIA
Impact of Airborne Pesticides on Natural Plant
Communities 108
Thomas Pfleeger, ERL-C, EPA, Corvallis, OR
DISCUSSION 124
Friday Afternoon - Comments on Summary Presentations
Saturday Morning - Comments on Preliminary Workshop Recommendations
FINAL RECOMMENDATIONS 129
APPENDIX A 1982 Subdivision J
APPENDIX B 1986 Standard Evaluation Procedure Nontarget Plants .
APPENDIX C Good Laboratory Practice Standards
APPENDIX D Participants
APPENDIX E Schedule Followed during Workshop
APPENDIX F Summary of Key Subdivision J Issues
APPENDIX G Discussion on Summary Presentation (Friday
Afternoon and Saturday Morning)
-------
EXECUTIVE SUMMARY
The U.S. Environmental Protection Agency is required by law (The Federal
Insecticide, Fungicide, Rodenticide Act [FIFRA]) to determine the potential
hazard posed by pesticides to nontarget vegetation. This is accomplished by
examining phytotoxicity data collected and submitted by registrants according
to procedures described in Subdivision J of the Pesticide Assessment
Guidelines (Appendix A).
Subdivision J was published in 1982 and Standard Evaluation Procedures
(Appendix B) were provided in 1986. Although the guidelines have been in use
for several years, their performance has never been evaluated by a working
group representing different scientific expertise and economic interests.
Therefore, the purpose of this workshop was to assemble a group of persons to
critically evaluate the tier test system described in Subdivision J, and make
recommendations as to how the guidelines may be improved.
Following the presentations and discussions, the workshop culminated with the
adoption by the group of a set of recommendations. The intent of these
recommendations is to encourage modification of Subdivision J so that:
1. the tier system is streamlined and brought in harmony with test
requirements of other regulatory bodies in an effort to reduce cost
without jeopardizing the accuracy or usefulness of the tier-test data;
2. the document is easier to understand, and thereby, registrants and test
laboratories are in a better position to design experiments, conduct
tests, and report data in a manner acceptable to EPA without unnecessary
delay or cost to either the agency or the registrant; and
3. tier III (field testing) is described in sufficient detail so that the
objective, performance, and interpretation of this level of testing can
be incorporated into the tier testing scheme without undue expense and
confusion.
RECOMMENDATIONS
I. Harmonize differences in test procedures between different regulatory
authorities or governing bodies (OECD, EEC, FIFRA, TSCA, FDA, CERCLA)
and work toward adopting universal standard tests for use throughout the
international community. Because of these inconsistencies, testing
costs for laboratories maintaining two or more programs are increased.
1) Establish what inconsistencies exist between agency test
guidelines, e.g.,
a) EC25 for effect under FIFRA, compared to 1-5% tolerance for
FDA.
IV
-------
b) Number of species required, number of plants per test, and
number of replicates per test.
c) Nutrient addition (FDA) compared to no nutrient addition
required by FIFRA.
d) Photoperiod requirements under FDA, but not specified in
some.
e) Watering regiments that should be optimized.
f) Endpoints that are required: FIFRA does not require shoot
heights, root length, and shoot and root weights, whereas
FDA does.
2) Call for joint efforts to arrive at a consensus on testing
procedures.
II. Revisions in tier I and tier II testing are needed to expedite the
procedure and to obtain the most meaningful data. The overall goal is
to reduce the cost, yet maintain the sensitivity of the screening tests.
A priority listing of the suggested revisions includes:
1) Drop tier I seed germination tests except for those cases where
there is reason to believe germination is a more sensitive
indicator of effects.
2) Develop evaluation criteria for seed germination and emergence
response in a defined soil type (see recommendation 4 in this
section). Specific definitions are needed for what constitutes a
germinated seed, an emerged seed, and the length of test-time
needed to conclude a negative result.
3) Simplify and reduce the cost of the tier I screening test by
eliminating the analytical determination of chemical test
solutions (a GLP requirement, Appendix C). The exception will be
when a negative result (no plant response) occurs, then analysis
should be conducted to prove that the chemical was administered at
the stated concentration. No recommendation is made to change the
current tier II requirements for chemical analysis.
4) Identify and characterize the nature of the soil required for
testing procedures (perhaps start with OECD guidelines). This
should include a consideration of organic content and soil
pasteurization.
5) Provide better statistical guidelines addressing: 1) experimental
design (number of replicates, etc.), 2) statistical procedures,
and 3) interpretation of statistical results.
6) Evaluate and expand the current recommended list of test species
with the objective of enhancing the use of more diversity. The
intent would not be to require more species to be tested, but to
include representative genera and families that might be
extrapolated to woody species and/or endangered species, where
appropriate.
-------
7) Review the current guideline requirements regarding the nature of
the chemical test product. Address the issue of how closely the
test-product must resemble the end-use formulated product which
usually includes surfactant, stickers, etc.
8 Provide guidelines for minimum test conditions in tier I and II,
i.e., temperature, photoperiod, light, and humidity. The
guidelines must be sufficiently flexible to accommodate the
different physiological needs of various test species.
III. Design and implementation of field experiments should be clarified and
developed in parallel with accompanying research (section IV-3).
1) Current tier III requirements appear to be more efficiently
accomplished if divided into two phases, or considered as two
separate tiers.
a) Develop protocols or consensus methodology for small plot
tests with regard to critical species, soil type, and other
field variables.
b) Preliminary field tests are carried out to identify
sensitive variables noted above.
2) The nature of more extensive tests, conceived as tier IV, can only
be determined after doing part (1, a and b). This includes the
regional conditions needed for the studies, species and species
assemblages to be included in the tests, and range of treatment
levels expected.
a) Establish the minimum information necessary for conducting a
valid risk assessment.
b) Research needs to be conducted to determine under what
conditions test data are adequate without tier IV
information (Section IV).
IV. Research is needed to improve the efficiency, and in some cases the
validity of testing protocols. Special case needs include: (in no
priorital order)
1) Establish the feasibility of using tissue culture methods as
options for tier I and II testing. Tier I might include several
different exposure concentrations, comparable to range-finding
tests in tier II, but without GLP/analytical determinations. A
special focus should be to use tissue cultures to test slow
growing woody perennials and endangered species.
2) Develop efficient life-cycle bioassays, both for representative
dicot and monocot species. Methods for the application of
chemicals should be included in these bioassays.
vi
-------
3) Research the possibility and procedures for using mesocosms and
field studies to evaluate chemical effects on plant communities.
a) Understanding agroecosystem models versus natural community
models.
b) What parameters should be evaluated to determine the extent
of effects?
c) When should such studies be required?
d) Evaluate the feasibility of using soil-core and terrestrial
microcosm chambers and other "off-the-shelf" technologies.
4) Study the possibilities of using new technologies, for example,
thermal sensing procedures, to monitor chemical effects in field
tests to predict and identify possible effects on nontarget plants
and plant communities.
5) Research is needed to provide optimal culture techniques for plant
species identified for testing. Species include forest species
(both canopy and understory) and wetland species.
6) Research is needed on the validity and accuracy of intraspecies
and interspecies extrapolation of toxicological data from
surrogate test species to potential nontarget plants.
V. National Agricultural Chemists Association workshop mini-workshop.
There appears to be a clear need for setting up a subsequent workshop
group or dialogue.
In summary, neither the government nor private sector alone has the resources
or expertise to accomplish the objectives set forth in these recommendations
and goals. A group effort will be required and this must include mechanisms
for the sharing of data and improved communications between all involved.
VALUE OF WORKSHOP PROCEEDINGS
The workshop succeeded in providing an open forum for discussion of key plant-
testing issues. Two of the workshop presentations identified Subdivision J
issues requiring attention. Seven issues of concern to EPA (page 13 of this
report) were listed in the paper by Drs. Charles Lewis and Rick Petrie, and
six additional issues (page 53) were provided as recommendations by Joe
Gorsuch based on his experiences as the director of a plant-testing
laboratory. The comprehensiveness of the workshop is illustrated by noting
that 10 of the 13 issues raised by these authors were discussed in at least
one of the formal presentations, and specific recommendations were made to
address 10 of the issues (see Appendix F).
Analysis of the workshop proceedings clearly shows areas of general agreement
among attendees, but it also reveals the depth of disagreement held by
opposing factions on certain suggested changes in Subdivision J or its
vii
-------
implementation. It is the opinion of the workshop organizers that the primary
value of this workshop is the documentation of key issues of concern, and
acknowledgement of the opposing views held on some issues. This documentation
provides the basis for systematic and efficient resolution of key differences.
In numerous cases, as the recommendations indicated, it appears that the
differences will not be resolved smoothly without conducting research
necessary to answer many fundamental questions about how plant tests should be
conducted and interpreted, especially field tests.
-------
DISCLAIMER
The information in this document has been funded wholly by the United States
Environmental Protection Agency. It has been subjected to the Agency peer and
administrative review, and it has been approved for publication as an EPA
document. Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
ACKNOWLEDGEMENTS
We acknowledge the participation and contributions of everyone who took part in
the workshop. We are grateful to Bob Hoist, Charles Lewis, and Rick Petrie of
EPA's Office of Pesticides Program (OPP) in Washington, D.C. for their
suggestions in planning the workshop. We thank Bill Hogsett and Dave Tingey for
their administrative support. A special thanks to those persons who made
presentations, and those participants who responded with constructive comments
and thought-provoking concerns.
John Fletcher and Hilman Ratsch
Organizers and Editors
-------
PURPOSE OF WORKSHOP
As a part of EPA's mission under The Federal Insecticide, Fungicide,
Rodenticide Act (FIFRA), the agency is required to determine whether or not a
pesticide will cause unreasonable adverse effects on the environment.
Evaluating the potential hazard of each pesticide to the environment is the
responsibility of the Environmental Fate and Effects Division within the
Office of Pesticide Programs. One component of the evaluation process is to
determine the potential hazard posed by pesticides to nontarget vegetation.
This is accomplished by examining phytotoxicity data submitted by registrants.
The data are generated from toxicity tests conducted and analyzed according to
procedures described in Subdivision J of the Pesticide Assessment Guidelines
(Appendix A). Subdivision J, prepared by Robert W. Hoist and Thomas C.
Ellwanger, describes three tiers of tests to be performed in sequence
depending on the test results of each proceeding tier.
Subdivision J was published in 1982 and based upon experience with its use.
Supplemental information was provided in 1986 in Standard Evaluation
Procedures for Nontarget Plants (Appendix B). For the most part, EPA believes
the guidelines are adequate to help characterize the risk to nontarget plants.
However, as with any guideline, there is a need to periodically evaluate its
performance and to look for possible improvements based upon experience gained
from its use. As such, the primary purpose of this workshop was to provide a
forum to discuss the scientific aspects of nontarget plant testing as
prescribed in Subdivision J, to identify perceived limitations in EPA's tiered
testing approach, and to identify associated research and development needs.
Furthermore, because registrants have sought additional guidance on the
design, conduct, and interpretation of tier III tests for risk assessment, a
special need was identified to examine tier III guidance (field
testing/validation) and to discuss the conditions under which tiered III
testing may be required.
In addition to the plant tests which OPP requires of the pesticide industry,
there are other US and international regulatory bodies which require
comparable plant test data for registration purposes. Since the purpose of
these test data is essentially the same for each agency, it is not surprising
that the test procedures and data reporting are similar. However, to the
chagrin of industry, datasets collected and processed according to one
agency's protocol are not always accepted by another agency because of
differences in test procedures or statistical analyses. These inconsistencies
among agency protocols creates confusion and places additional financial
burden on industry if multiple test facilities and/or conditions must be
maintained to satisfy the requirements of different regulators. Thus, there
are several issues stemming from the use of Subdivision J which make it timely
to review the effectiveness of this document. This will be the first open
review of Subdivision J since it was published in 1982.
-------
ORGANIZATION AND OPERATION OF WORKSHOP
A three-day workshop was conducted to evaluate the tier test system described
in Subdivision J,and to make recommendations to improve guidelines. Twenty-
nine people participated in the workshop (Appendix D). Approximately equal
numbers of persons were selected from regulatory agencies, industry, plant
test laboratories, agriculture, and academia. The combined expertise of these
individuals covered a broad spectrum of knowledge and skills. Various persons
in attendance had written Subdivision J, participated in preliminary round-
robin testing of Subdivision J test protocols, supervised personnel in
commercial laboratories performing Subdivision J tests, submitted test data to
EPA, evaluated test data received by EPA, conducted basic research to evaluate
new or revised plant-test systems, tested newly registered herbicides on
nontarget plants under field conditions, and studied natural plant community
and population changes.
The input from participants came in three forms: formal talks were given,
indepth discussions held, and recommendations were made (Refer to Appendix E
[page 141-143] for the schedule.). Participation by attendees in all facets
of the workshop was facilitated by holding 4 sessions each devoted to a
separate topic. The four sessions dealt with: 1) Inception and Implementation
of Subdivision J, 2) Ecological and Taxonomic Considerations, 3) Laboratory
and Greenhouse Testing (Tiers I and II), 4) Field Testing (Tier III). After
each presentation there was a question-answer period. After the fourth
session, an indepth discussion was held to review all aspects of Subdivision
J. Based on this discussion, a committee drafted recommendations for
revisions of Subdivision J. These tentative recommendations were discussed by
attendees on the subsequent morning, and from this discussion, emerged the
final list of recommendations provided in this report (page 141-143).
-------
SESSION I: INCEPTION AND IMPLEMENTATION OF SUBDIVISION J
In this session, the need and use of data from Subdivision J, tier tests were
presented from the perspective of the U.S. Environmental Protection Agency. The
first paper focused on the preparation and introduction of the Subdivsion J
document whereas, the second paper described how plant data received from
registrants are currently used by EPA.
-------
PESTICIDE PHYTOTOXICITY TESTING
A HISTORICAL PERSPECTIVE
by
Robert W. Hoist, Ph.D.*
Office of Pesticides and Toxic Substances
U.S. Environmental Protection Agency
Content: The history of the Federal Insecticide, Fungicide and Rodenticide
Act was presented with special emphasis on the implementation of
the 1972 amended act by the Environmental Protection Agency.
INTRODUCTION
The early forms of the Federal Insecticide, Fungicide and Rodenticide Act
(FIFRA), originally enacted in 1947, were designed as consumer protection acts
under which efficacy of products was tested. A part of efficacy testing
included target-area phytotoxicity.
With the 1972 amendments to FIFRA, the Act was changed to an environmental
protection law. The amended Act required the publishing of guidelines
specifying the kinds of information required to support registration of a
pesticide. In part, the required information allows the EPA to reach a
conclusion that a pesticide will perform its intended function without
unreasonable adverse effects on the environment. The 1972 amendments to FIFRA
also called for periodic updates of the guidelines.
The guidelines for evaluating phytotoxicity of pesticides were developed in
1977 through 1980. The first efforts were based on my experience that I had
working with pesticide, salt and air pollution phytotoxicities gained while
working on my'advanced degrees at Southern Illinois University - Carbondale
and at Boyce Thompson Institute for Plant Research, Yonkers, New York. The
first emphasis with respect to phytotoxicity testing was the evaluation of
metabolic changes that occurred because of pesticides.
In 1978, I began working with Dr. Frank Benenati of OTS and together developed
a basic set of phytotoxicity testing based on whole plant responses. This
testing regime is based on the fact that plant growth and development, as may
be affected by pesticides and other toxic substances, is best demonstrated by
the whole plant rather than individual metabolic processes. Testing of
certain metabolic processes were still required because I felt that some
* presently affiliated with The Naval Research Laboratory
-------
processes could not be evaluated by a whole-plant test. Nitrogen fixation was
one such test. This test was later dropped.
Tier Testing
The basic scheme of testing using a tiered approach was arrived at through
discussions with various researchers in academia, industry and government at
scientific meetings and a visit to the DuPont Laboratories in Wilmington, DE
in 1979. Three phases of terrestrial plant growth and development, seed
germination, seedling emergence, and vegetative vigor would be tested, first
under controlled conditions and then in the field. Neither reproduction
potential nor a life-cycle test was included.
Also considered were three other terrestrially related tests. They were
mutagenicity using Tradescantia, soil microorganism viability taken from the
environmental fate testing series, and spray drift. The mutagenicity test was
dropped because its results could be manifested in the whole plant test
scheme. The soil microorganism test was set aside because of the difficulty
in identifying the soil microflora and how they may be affected by the
pesticide. The effects on soil microflora can be seen to some extent in the
soil metabolism studies performed as part of the environmental fate studies.
The spray drift studies were initially connected with phytotoxicity because of
the numerous problems associated with off-target movement of herbicides.
However, it became readily apparent that off-target movement of pesticides in
general is important in an overall exposure assessment. This set of studies
was finally given its own guideline section (Subdivision R).
Species Selection
Species selection was the most difficult portion of the work. The original
list of ten specific species was based on testing experiences by various
researchers and the ease of acquiring and growing the plants. However, after
some evaluation by Spencer Duffy and myself at the OPP laboratory in
Beltsville, Maryland, and further discussions with researchers, it was
determined that the species list should be liberalized. Two specific species
were identified: soybean and corn. A root crop such as carrot or radish was
also required. The other seven species were to be divided between monocots
and dicots with a good family representation of these orders. Other problems
with species selection also came to light. Hybrid variations with respect to
responsiveness to pesticides and the choice of species that may represent
endangered or threatened species are two such problems.
The first set of phytotoxicity "guidelines" were published as a proposed
regulation in the Federal Register on November 3, 1980 (USEPA, 1980). Shortly
thereafter, it was decided by OPP management that the guidelines should be
non-regulatory to allow more freedom of growth with advances in the various
scientific fields they represented. All of the existing guidelines were
revised and published in October 1982 including the one on phytotoxicity -
Subdivision J (Appendix A). All of these guidelines are available through the
National Technology Information Service in Springfield VA.
-------
In 1980, the Agency also decided that there was no need for target area
phytotoxicity testing allowing that the manufacturers would generally test for
this as a matter of course and place the appropriate phytotoxicity warnings on
the label. Non-target area phytotoxicity testing was also waived because of
the same conceptions.
However, phytotoxicity problems continued and a more pressing matter made the
Agency reconsider its policy with respect to non-target area phytotoxicity
testing. Endangered and threatened plant species were being identified in
increasing numbers and in locations close to agricultural, forestry and
industrial pesticide application sites. Since 1982, increasing amounts of
information have been requested by the Agency with respect to phytotoxicity of
various pesticides because of application methods, formulations, increased
pesticidal activity, and the escalating numbers of endangered and threatened
plant species.
The Agency is undertaking a program of testing harmonization between the
testing required by the Office of Pesticide Programs and the Office of Toxic
Substances. The basic set of phytotoxicity tests for these two offices were
developed along the same lines. There may be some divergence due to
differences in the two Acts (FIFRA [Hoist and Ellwanger, 1982] versus TSCA
[USEPA, 1985]) but the scientific principles remain the same. Also, the
Organization of Economic Cooperation and Development (OECD, 1984) tests for
phytotoxicity were developed by Dr. Benenati, Dr. Clive Price of Imperial
College, England and myself follow these principles of testing the whole plant
during various phases of its life cycle. The species selection issue for the
OECD tests has been a matter of contention due to the desire to test species
common to a country's economic or natural situation.
The Agency is undertaking a general updating of some of the pesticide testing
guidelines. This workshop is very timely with respect to looking at various
issues that need resolution in order to test for phytotoxicty more effectively
and provide the EPA with better information on which it can base its risk
assessments for pesticides and other toxic substances.
REFERENCES
Hoist, R.W. and T.C. Ellwanger. 1982. Pesticide Assessment Guidelines,
Subdivision J, Hazard Evaluation: Nontarget Plants. Office of Pesticides and
Toxic Substances, U.S. Environmental Protection Agency, Washington, D.C.
Organization for Economic Cooperation and Development. 1984. Terrestrial
Plants, Growth Test. OECD Guideline for Testing of Chemicals, Guideline 208.
OECD, Paris, France.
-------
U.S. Environmental Protection Agency. 1980. Proposed guidelines for
registering pesticides in the United States: Hazard evaluation of use of
pesticides for nontarget plants and microorganisms. Federal Register Vol. 45,
No. 214: 72948-72978.
U.S. Environmental Protection Agency. 1985. Environmental Effects Test
Guidelines. Federal Register 50: 39321-39397.
-------
PLANT DATA ANALYSIS BY ECOLOGICAL EFFECTS
BRANCH IN THE OFFICE OF PESTICIDES PROGRAM
by
Charles Lewis and Richard Petrie
U.S. Environmental Protection Agency
Content: The basis for requiring plant-test data from industry was
described. The nature of the data required and how they are used
by EPA was discussed with the use of an example. Several issues
of concern were identified regarding current testing and
evaluation procedures.
INTRODUCTION
Under the Federal Insecticide, Fungicide, Rodenticide Act (FIFRA), the EPA
(Figure 1) is responsible for registering and re-registering pesticide
products to ensure that they "will not generally cause unreasonable adverse
effects on the environment when used in accordance with widespread and
commonly recognized practice."
Although not contained in the title, FIFRA regulates any product that claims
the control or mitigation of a pest including: herbicides, algicides,
dessiccants, defoliants, plant growth regulators, sanitizers, disinfectants,
and biological control agents.
Widespread concern for off-target effects of primarily herbicides on plants
first occurred in the 1950's following the introduction and widespread use of
phenoxy herbicides. Sensitive crops growing in close proximity to treated
fields were damaged; resulting in enactment of state laws and tighter Federal
labeling limiting conditions of use (such as maximum wind speeds). The
phenoxy manufacturers developed formulations that were non-volatile or of very
low volatility. Very little attention was given to off-target effects of
herbicide drift or volatility on near-by terrestrial and aquatic plants of
lesser economic importance that are used by fish and wildlife for food and
cover.
As more and more classes of herbicides were introduced and utilized on more
and more acres across the U.S., complaints of off-target effects on plants
increased. The development and broad use of soil applied herbicides in the
60's, 70's, and 80's greatly reduced aerial drift concerns but unfortunately
increased surface runoff and ground water concerns. Coming full circle,
pesticide manufacturers are once again focusing their research and development
-------
OFFICE OF THE ADMINISTRATOR
Administrator
William K. Reilly
Deputy
F. Henry Habicht II
ASSISTANT ADMINISTRATOR FOR PESTICIDE AND TOXIC SUBSTANCES
Assistant Administrator
Linda J. Fisher
Deputy
Victor J. Kimm
OFFICE OF PESTICIDE PROGRAMS
Director
Douglas D. Campt
Deputy
Susan H. Wayland
ENVIRONMENTAL FATE AND EFFECTS DIVISION
Director
Anne L. Barton
Deputy
Paul F. Schuda
OTHER DIVISIONS
Registration
Special Review and Re-Registration
Program Management Support
Hazard Evaluation
Field Operations
Biological and Economic Analysis
ECOLOGICAL EFFECTS BRANCH
Acting Chief
Douglas J. Urban
Acting Deputy
Norman J. Cook
ENVIRONMENTAL FATE AND GROUNDWATER BRANCH
Chief
Henry M. Jacoby
Deputy
Elizabeth M. Leovey
Figure 1 - Organizational structure of the Environmental Protection Agency as it pertains to
the Office of Pesticide Programs (OPP) and the Ecological Effects Branch (EEB)
-------
activities on foliar applied herbicides that are active at very low use rates
(some in the ppt range).
In October of 1982, Subdivision 0 Nontarget Plant Pesticide Assessment
Guidelines were published (US EPA, 1982). In these guidelines, the Agency
describes test protocols and reporting procedures for the conduct and
submission of nontarget phytotoxicity data. These data are required of
registrants on a case-by-case basis depending on the activity of the pesticide
being registered, and the proposed use pattern.
TEST DATA REQUIRED FROM REGISTRANTS
Purpose
All data requirements are to provide data which determines the need for
precautionary label statements.
Source
The data required to assess the hazards of herbicides to nontarget terrestrial
plants are derived from short-term laboratory tests and simulated field
studies. Results from each tier are evaluated to determine the need for
further testing.
Documents
Test Guidelines : Subdivision J, Hazard Evaluation: Nontarget Plants, October
1982, and
Standard Evaluation Procedures (SEP's) - 1986:
A. Nontarget Area Plants
B. Seed Germination/Seedling Emergence, Tier I & II
C. Vegetative Vigor, Tier I & II
D. Growth and Reproduction of Aquatic Plants, Tier I & II
E. Terrestrial Field Testing. Tier III
F. Aquatic Field Testing, Tier III
G. Pesticide Spray Drift Evaluation:
1. Droplet Size Spectrum Test
2. Drift Field Evaluation Test
8
-------
CURRENT USE AND POLICY CONCERNING TIER TESTS
Using a Tier approach to data development, registrants who are asked to submit
data proceed as follows:
1. Determine if the chemical is toxic to plants. If phytotoxic, proceed to
tier I. If an herbicide, proceed to tier II.
No herbicide phytotoxicity data are required if applied solely to
food/feed crops; and, if applied with ground equipment only; and, if the
herbicide volatility is less than 1.0 x 10"5 mm Hg and if the herbicide
is less than 10 ppm water solubility. Exceptions to these rules
include: known cases of documented adverse effects in the field,
potential for adverse effects to endangered species, or if the pesticide
is in Special Review at EPA.
2. Determine if greater than 25% adverse effects (EC25)a are occurring to
terrestrial species in Tier I and Tier II tests, and that the use rate
will result in excessive off-target movement. If so, Tier III tests are
required.
DESCRIPTION OF TIER TESTS
Tier I
122-lb Seed germination/seedling emergence/vegetative vigor (Table 1)
122-2 Aquatic plant growth
Conducted with technical grade active ingredient (TGAI) in the laboratory at
the maximum label rate, or at least three times the estimated environmental
concentration (EEC).
Tier II
123-1 Seed germination/seedling emergence/vegetative vigor
123-2 Aquatic plant growth
Conducted with TGAI in the laboratory. At least 5 concentrations should be
tested to establish an EC25a for terrestrial species or EC50 for aquatic
species.
a ECjjg - External pesticide concentration required to cause a 25% detrimental
change or alteration in plant growth and/or development.
b Series number as listed in Pesticides Assessment Guidelines Subdivision J
(see Appendix A).
-------
Tier III
124-1 Terrestrial field
124-2 Aquatic field
Conducted with typical end-use product in the field.
Table 1 - The plant species recommended by EPA for tier tests.
Dicotyledons (6 species from 4 families)
Tomato Lvcopersicon esculentum
Cucumber Cucumis sativus
Lettuce Lactuca sativa
Soybean* Glvcine max
Cabbage Brassica oleracea
Carrot* Daucus carota
* soybean and a root crop are required
Monocotyledons (4 species from 2 families)
Oat Avena sativa
Ryegrass Lolium perenne
Corn* Zea mays
Onion Alii urn cepa
* corn is required
10
-------
THE USE OF TIER TEST DATA (EC25) BY EPA
The Environmental Effects Branch (EEB) in OPP (Figure 1) makes decisions and
recommendations pertaining to individual pesticides by comparing EC25 values
resulting from registrant's tier testing with EEC values (estimated
environmental concentrations) calculated by EPA. Estimated environmental
concentration values are based on the amount of chemical assumed to move off-
target (drift and/or runoff) and the theoretical distribution of the chemical
in the matrix (soil or water) at the nontarget site. The EEC values are based
on the maximum label rate. In making the estimates, the influence of a host
of different use factors are considered such as sites of application, method
of application (aerial versus ground), number of applications per year, and
the maximum amount of pesticide applied per acre per year.
Fundamental to making the EEC estimates are assumptions which must be made
pertaining to the amount of chemical which moves off-site and how it becomes
distributed off-site. A major distinction is made between aerial application
by plane or air blast and sprinkler irrigation versus terrestrial application
by ground equipment. The general guidelines used for these methods of
application are shown in Tables 2 and 3.
The use of the guidelines in Tables 2 and 3 are illustrated by the sample
calculations shown in Table 4. In this example, a chemical with a water
solubility of 300 ppm has been applied by aircraft at a rate of one pound
active ingredient per acre. The concentration of chemical which would be
present in the top one inch of soil in the nontarget acre adjacent to the acre
where the chemical was applied was estimated by calculating the combined
amount arising from both runoff and drift. In the sample calculation, the EEC
is estimated to be 0.176 ppm. In evaluating test data submitted by a
registrant, these values are compared to EC25 values obtained from tier II
testing. If the EC25 value for any of the species tested is greater than the
calculated EEC, then tier III testing is required of the registrant.
Comparison of EEC and EC25 Values
After comparing the EEC value with the EC25 values, we then determine if :
1. label rates, methods of application and other conditions of use such as
limitations and precautions (as specified on the label) are adequate;
2. decide if all proposed use sites are adequately supported by the data
in-hand;
3. determine if any threatened/endangered plant species are at risk; if so
limit use accordingly by restricting the pesticide from being applied in
specific counties or locations;
4. determine if restricted use classification (application by certified
professional applicators only) is necessary;
11
-------
Table 2 - Guidelines for estimating movement of chemical from target to nontarget
areas.
Application Drift9 Runoff
Method Amount Distance Amount Distance
Aerial
Ground
5% of amount
applied to
target area
none
adjacent
acre
1-5%" of
60% of
amount6
applied to
target area
1-5%
adjacent
acre
adjacent
acre
a Drift estimates are based on data reported in references 2 and 3.
b Estimation of the precent runoff is based on solubility:
5% if water solubility is > 100 ppm
2% if water solubility is 10 to 100 ppm
1% if water solubility is < 10 ppm
Refer to references 1 and 4 for details.
0 When herbicides are applied by air, the Ecological Effects Branch of EPA
assumes a 60% application efficiency.
Table 3 - Guidelines for estimating the concentration of drift and/or runoff chemical in
nontarget water or soil matrix.
Amount and Area Exposure
of Nontarget Chemical
Concentration in Nontarget Martix
Water Soil
1 Ib. active ingredient/acre
734 ppb (6 in. deep)8
61 ppb (6 ft. deep)
2.2 ppm (1 in. deep)1
8 Refer to reference 1 for details.
6 Based on a specific gravity of soil in the range of 1.8 to 2.6. Estimated
soil weight of 125 Ib/cu ft.
12
-------
Table 4 - Sample calculation for determining the "estimated environmental
concentration" of a chemical applied aerially at one Ib ai/A whose water
solubility is 300 ppm.
Step Purpose and Description
1 Drift determination
Runoff determination
a. Amount available
for runoff
b. Amount which
runs off
Sum of drift and
runoff deposition to
nontarget area
Determination of
chemical in top inch
of soil in nontarget
acre adjacent to
target area
Calculation Result
Applied chemical
x % drift factor
0.05 Ib ai/A
0.6 Ib ai/A
0.03 Ib ai/A
1 Ib ai/A x 0.05a
Applied chemical
x application
efficiency
1 Ib ai/A x 0.6b
Amount available x
% runoff factor
0.6 Ib ai/A x 0.05C
Amount of drift
chemical + amount
of runoff chemical
0.05 Ib ai/A +
0.03 Ib ai/A
Amount of chemical
deposited on
nontarget x standard
concentration
0.08 Ib ai/A x 0.176 ppm
2.2 ppm/ai/Ad
0.08 Ib ai/A
8 Standard % drift factor (Table 2) based on references 2 and 3.
b Standard application efficiency used by the Ecological Effects Branch of OPP
(Table 2).
c Standard % runoff factor (Table 2) based on references 1 and 4.
" Standard concentration used by the Ecological Effects Branch when estimating
the distribution of 1 Ib of chemical in the top 1 inch of 1 acre of soil
(Table 3).
13
-------
5. determine if additional studies to quantify pesticide levels and effects
are needed; if so issue 3C2b data request;
6. determine if suspension/cancellation action is warranted to ensure
environmental safety. This action called a Special Review begins with a
Grassley-Allen letter notifying the registrant of our concerns and
allowing the registrant a response period of 90 days after receipt. A
Position Document 1 (PD-1) is then issued by EPA. This document
publicly states the Agency concerns. After receipt of public comments
and registrant rebuttals, a PD 2/3 is then issued which includes the
benefit/risk assessment. After another comment period, a PD-4 final
decision document is issued by EPA. This may address cancellation of
all, or some uses, rate changes, changes in methods of application,
formulation changes, and other restrictions on use. The registrant can
challenge the PD-4 decision resulting in an administrative hearing
before an EPA administrative law judge (ALJ). The ALJ decision can be
appealed to a higher court by the registrant.
CONCLUSIONS
Having reviewed a number of tier I and II non-target plant studies submitted
by registrants, we raise the following issues regarding test adequacy and data
interpretation:
1. Are we testing the correct species in the tier I and II tests?
2. Should only one test method be required for each test?
3. Should we have a life-cycle test using Arabidopsis and/or a Brassica
species at the tier II level?
4. Should we require that these tests be conducted on the technical end
product?
5. Are our estimated environmental concentration scenarios valid in the
interpretation of tier II test data?
6. Should a safety factor be required for endangered species?
7. How important is the development of test methods for tier III?
REFERENCES
U.S. Environmental Protection Agency. 1986. Ecological Risk Assessment.
EPA-640/9-85-001. U.S. Environmental Protection Agency. EEB/EFED/OPP/EPA.
U.S. Environmental Protection Agency. 1990. Preliminary estimation of EEC
from surface water runoff. Ecological Effects Branch Internal Document. U.S.
Environmental Protection Agency. EEB/EFED/OPP/EPA.
14
-------
Akesson, N.B. and Yates, W.E. 1984. Physical parameters affecting aircraft
spray application. Chemical and Biological Controls in Forestry. 95-115 pp.
U.S.D.A. Forest Service. 1984. Herbicide Spray Drift Predictions Using the
Forest Service FSCBG Forest Spray Model. A report by H. E. Cramer Company,
Salt Lake City, Utah. FPM 84-1.
Hoist, R.W. and Ellwanger T.C. 1982. Pesticide assessment guidelines,
Subdivision J, hazard evaluations: nontarget plants. EPA 540/9-82-020. U.S.
Environmental Protection Agency, Office of Pesticide and Toxic Substances,
Washington, D.C. 55 pp.
15
-------
SESSION II: ECOLOGICAL AND TAXONOMIC CONSIDERATIONS
In risk assessment analyses, including the potential impact of pesticides on both
natural- and agro- ecosystems, it is worthwhile to occasionally reexamine the
overall goal of regulatory measures in view of the complexity of natural and
agricultural habitats. The first two papers in this session called attention
to the ecological and taxonomic complexity of nontarget vegetation. The third
paper provided insight into how computer technology may be useful in the future
to predict regional impact of a broad spectrum of air toxicants, including
pesticides.
16
-------
ROLE OF BIOTIC AND ABIOTIC FACTORS
IN PLANT GROWTH AND BIODIVERSITY
by
G. E. Taylor, Jr.
Desert Research Institute
Content: The role of environmental stresses in plant sciences was discussed
with a specific focus on anthropogenic factors and plant
biodiversity. Specific topics covered included: nature and scope
of environmental stresses affecting plant growth, general response
of plant systems to stress, aspects of biodiversity and the role
of stress, and relative significance of natural and anthropogenic
stresses in governing biodiversity.
INTRODUCTION
A fundamental theme in the life sciences is that the environment exerts a
pervasive influence on the ability of plants and animals to grow, survive and
reproduce, which collectively are the determinants of Darwinian fitness. This
tenet is clearly embodied in the well-documented features of population
biology in which a population's rate of growth is always well below that which
can be sustained under optimal conditions (i.e., biotic potential). Thus,
genetically determined, intrinsic rates of growth are dampened by the local
environment. The significance of the environmental constraint on biotic
potential affects a variety of ecological issues including the species numbers
and abundance, microevolution of populations, biome structure and function,
and community dynamics.
This concept is a rudimentary underpinning in the disciplines of plant biology
and ecology and heavily dictates research activities in both the basic and
applied sciences. Its significance is experienced at levels of organization
ranging from cellular biochemistry and molecular biology to regional and
global issues in conservation biology. From the more practical standpoint,
the concern is that anthropogenic stresses in some areas significantly impact
the physiology, growth, and reproductive success of biota. With respect to a
population biology, the issue can be rephrased to one in which anthropogenic
stresses may be quantitatively important in constraining the biotic potential
of a population or species.
The objective of this paper is to explore in general terms the role of
environmental stresses in the plant sciences, with a specific focus on
anthropogenic factors and plant biodiversity. This objective is met by
17
-------
discussing in sequence the following topics: (i) nature and scope of
environmental stresses affecting plant growth and development; (ii) general
response of plant systems to stresses; (iii) general aspects of biodiversity
and the role of stress; and (iv) relative significance of natural and
anthropogenic stresses in governing biodiversity.
NATURE AND SCOPE OF ENVIRONMENTAL STRESSES
In an ecological context, stress can be defined as any environmental factor
capable of inducing a potentially injurious response, which in turn is defined
as any stress-induced change in the organism's biochemistry, physiology and/or
growth (Levitt, 1972). The relationship between the stress and biological
response can be either direct (i.e., easily established cause-effect
relationship with a defined suite of symptoms) or indirect (i.e., subtle
changes in physiology, without any stress-specific symptomology). The
exposure dynamics can be chronic (occurring at low levels over a sustained
period of time) or acute (short-term exposure to highly intense level of the
stress).
Whereas a number of ways exist to categorize environmental stresses, one of
the most convenient is as biotic versus physiochemical (abiotic) (Levitt,
1972). The former is defined as any stress that originates specifically from
species' interactions either at the interspecific (between) or intraspecific
(within) level; these are typically studied in the disciplines of ecology and
pathology. The most notable examples are competition for limited resources
and pest/pathogen interactions. While competition exists at the intraspecific
level, in most natural situations the more significant form of competition
operates among species as organisms compete for limited resources (e.g.,
light, soil water, nutrients, microsites for seedling establishment, etc.).
Competition among conspecifics (individuals of the same species) dominates
community dynamics in intensively-managed ecosystems comprised of monocultures
(e.g., agriculture), whereas interspecific competition is far more significant
in natural ecosystems comprised of mixed age classes of multiple species.
Pest/pathogen interactions are extremely varied and include such diverse
issues as herbivory, viral and fungal infestations, and predation among animal
species. In the vast majority of natural ecosystems and many that are
managed, biotic interactions are the dominant stresses governing flow of
energy, cycling of nutrients, and dynamics of community structure. This
feature is particularly relevant to those landscapes in which species
diversity is high (e.g., mixed deciduous forest, tropical rain forests, coral
reefs) since the plant and animal communities have co-evolved an intricate web
of biotic interactions. In this context, it is important to recognize that
physiochemical stress effects at the community level may manifest themselves
as biotic interactions whereby competitive relationships are altered.
The physiochemical category of environmental stresses includes all factors of
abiotic origin that are physical or chemical in nature. One convenient scheme
for further classifying these stresses (Figure 1) details classes of
temperature (high or low), water (drought or flooding), radiation
(ultraviolet, infrared, visible, ionizing), chemical ions, and physical
18
-------
ENVIRONMENTAL STRESS
BIOTIC
PHYSIOCHEMICAL
TEMPERATURE
WATER
RADIATION
COLD HEAT DROUGHT FLOODING
INFRARED VISIBLE ULTRAVIOLET IONIZING
SALTS
AIRBORNE
CHEMICALS
CHEMICAL
MISC
PHYSICAL
WIND 80IL EMF
COMPACTION
HERBICIDES INSECTICIDES
Figure 1 - Components of environmental stress that influence the physiology and
growth of plant species (Levitt, 1982). EMF refers to electro magnetic
field.
19
-------
disturbance (EMF, soil compaction, wind). Each of these classes is further
subdivided to identify stresses that are related in origin but quite different
in their effects on plant processes.
The class of stresses most relevant to this workshop are those listed as
chemical in Figure 1. This class is inclusive of stresses whose effects are
commonly associated with the biochemical and physiological consequences of
high concentrations of chemicals in the soil, atmosphere or water media.
Notable examples include salts, heavy metals, carbon dioxide, organics,
particles, herbicides, and toxic gases. It is important to recognize that the
exposure pathway for these stresses may be highly circuitous, involving a
convoluted pathway through the terrestrial or aquatic ecosystem (i.e., food
chain transfer). This exposure pathway allows for progressive loading of
incrementally small stress levels (Johnson and Taylor, 1989).
The sources of environmental stress provide another means of classification,
and one of the most convenient is natural versus anthropogenic in origin.
This dichotomy is not absolute because many stresses commonly assumed to be
solely anthropogenic also exist naturally although the stress' distribution
(spatial and temporal) and intensity are typically greater than that which
would occur under more pristine conditions. The most notable examples are the
airborne chemicals (sulfur oxides, tropospheric ozone, nitrogen oxides),
temperature, water relations, and multiple aspects of solar radiation (e.g.,
UV-B).
These schemes for classifying environmental stresses are solely of
organizational value because in an ecological context most populations and
species are continuously challenged by an array of stresses that vary
spatially and temporally in their relative importance. Consequently, it is
extremely rare that a species' population or even an individual's physiology,
growth, and reproductive success is solely controlled by a single stress. The
more common situation is for the existing biological state to reflect a mix of
interacting stresses (natural and anthropogenic in origin).
The response of biological systems to environmental stress typically is
initiated at a given level of hierarchy (e.g., leaf-level physiology,
community dynamics) but thereafter is propagated to another level. For
example, some of the most immediate effects of elevated levels of carbon
dioxide are on the plant's carbon economy, but the organismal effect may be
propagated to community level as the competitive relationships between species
are altered. Equally relevant are those situations in which stress responses
are confined to a given level simply because the system possesses a
homeostatic capacity to repair or compensate for the injury. One of the
underlying aspects of this responsiveness is the issue of time scale since
biochemical or physiological responses commonly occur on time scales of
seconds-to-minutes whereas ecosystem and community-level responses require
years-to-decades to materialize.
The key issue is to identify the homeostatic factors that control the
propagation of stresses between levels of biological organization. For the
issue of biodiversity and this workshop, this aspect is particularly important
since the initial site of action for a pesticide is likely to be at a
20
-------
biochemical or physiological level of organization and yet the regulatory
concern lies at a higher level. This aspect is quite distinct from that of
the more commonly discussed aspects of biodiversity (e.g., deforestation) in
which the underlying cause is habitat destruction rather than selective
species removal.
RESPONSE OF BIOLOGICAL SYSTEMS TO STRESS
The response of biological systems to environmental stresses is a function of
both exogenous and endogenous factors. Endogenous control is largely
genetically determined, and control is regulated at one of three sites: (i)
uptake or assimilation of the stress; (ii) intrinsic sensitivity of the
biochemical target site to sustain injury (Tingey and Taylor, 1982); and (iii)
capacity of homeostasis to repair or compensate for the injury. This analysis
is relevant at all levels of biological organization ranging from the response
of individuals through that of communities, although the specific mechanisms
governing uptake, sensitivity, and homeostasis vary as a function of the level
of biological organization.
Exogenous factors also control the responsiveness of biotic systems to
environmental stresses. The most important is the dynamics of the chemical
potential exposure regime of the stress. This simply reflects the
concentration of the chemical in the air, soil, or water medium and the
temporal variability in the exposure dynamics. Also important are the
non-chemical aspects of the exogenous environment including edaphic,
atmospheric and climatic variables. These factors play a role by modulating
the concentration or exposure dynamics of the stress or the array of
endogenous factors that control uptake, intrinsic sensitivity, and
homeostasis. A notable example of environmental modulation of seemingly
unrelated stresses is the role of UV-B radiation in the
activation/deactivation mechanisms of organic pesticides, in which specific
wavelengths of UV-B light can biochemically alter the pesticide's specific
toxic site.
The speed and magnitude of stress effects are governed by the interplay of the
exogenous and endogenous factors. Since the endogenous factors are largely
genetically encoded, responsiveness will differ significantly from species to
species, and this is one of the most generic and characteristic aspects of
plant response to all environmental stresses. Consequently, the effect of any
stress, independent of its mode of action or origin as a natural or
anthropogenic, will differentially affect species with some organisms
exhibiting no effects while others experience significant changes in
physiology, growth, and reproductive success.
At the whole-plant level of organization, a stress places an organism at a
disadvantaged state, requiring that energy be expended to challenge the
stress. This expenditure of energy is commonly observed in changes in
respiration (dark and light) or a diversion of energy reserves from growth and
21
-------
BIOLOGICAL RESPONSES TO ENVIRONMENTAL STRESS
ASSIMILATION
INTRINSIC SENSITIVITY HOMEOSTASIS
RESPONSE
Food Chain Transfer
Foliar Sorptlon
Root Uptake
Bloconcentratlon
Litter Accumulation
-Biochemical
Target SUe
P/R Ratio
Species Sensitivity
• Biochemical Repalr\
• Whole-Plant
Compensation
• Species Invasion'
• Species Competitive
Balance
• Opportunistic
Species
• Species
Distribution
• Growth
• Reproductive
Success
• Biodiversity
• Fitness
• Productivity
• Mlcroevolutlon
Figure 2 - The processes that control plant response to any environmental stress
(Tingey and Taylor, 1982).
22
-------
reproduction to maintenance (Mooney, 1972). This paradigm suggests that
differences in response among populations or species reflect in a general
sense the ratio of production (P) to respiration (R), and this agrees well
with the observed chronic level stress effects in such notable areas as those
impacted by ionizing radiation, heavy metal smelters, and gaseous air
pollutants (Woodwell, 1970). Moreover, it explains why populations at the
margins of a species' distribution are particularly responsive to new
environmental stresses since these individuals are operating on a P:R ratio
approaching one.
Equally important in the context of assessment is to recognize that species
are comprised of pockets or clines of populations that are genetically related
but quite different in the array of traits that control fitness. This
plant-to-plant or intraspecific variability is a consequence of both
stochastic events (e.g., founder effects) as well the nonrandom "molding" of
the population's variation over time to adjust to the vagaries of the
environment. The consequence is that the responsiveness within a single
species to environmental stress can vary dramatically in space and time as a
function of the genetic structure. This understanding of the role of genotype
in governing stress responses also explains why effects within a individual
may differ over time since an organism's developmental state is driven by a
progression of differentially expressed, genetic configurations.
At the level of plant populations and species, an environmental stress can
have five consequences that are not necessarily mutually exclusive. The first
is simply one in which the level of stress does not exceed the intrinsic
sensitivity of the biota, and as a consequence there is no effect. The second
is one in which organisms are affected but the ecological amplitude of the
population or the individual's phenotypic plasticity is sufficiently robust to
accommodate or compensate for the stress with a minimum effect. The third is
a change in geographical distribution; this response occurs most commonly on
the margins of a species' distribution (rather than in the insular areas). If
the stress is prolonged in time and clinal in intensity, the population or
species may migrate such that its geographical distribution shifts from one
region to another. Notable examples are the pronounced changes in species
distribution that occurred as a function of glaciation and the rather ominous
changes in natural and managed species projected as a function of global
climate change. The third consequence is a change in abundance, which simply
reflects a decline in reproductive success at the level of populations or
species. The forth is extinction whereby the gene pool is eliminated either
locally (population) or throughout the species distribution. The fifth and
final response is one in which the stress results in selection within a
population because the fitness of resistance genotypes is enhanced at the
expense of more sensitive counterparts. The net consequence is that the
genetic structure of the population changes as the frequency of alleles that
confer resistance increases.
It is important to recognize that stress effects may actually cause changes in
species abundance and distribution that constitute increases rather than
decreases. This direction of change simply reflects the previously discussed
aspect of homeostasis. At the level of populations and plant communities,
competition for limited resources may change due to environmental stress, and
23
-------
the effect for some species may be to open up niches that were previously
occupied by other species. This type of response is commonly observed for
more opportunistic or generalist species (i.e., weedy species).
In the context of this workshop, the most commonly discussed consequence of
biodiversity is a decline in abundance due to an environmental stress (Holt,
1990). In terms of genetic diversity however, the far more common response is
likely to be a change in population genetic structure on a very local scale.
It is anticipated that most applications of pesticides that have a negative
impact on plant growth and development of nontraget species will result in the
selective removal of sensitive genotypes and that long-term application will
maintain the selective pressure. However, while this response may be the most
common, its consequences are conjectural and experimental detection is labor
intensive.
While it is commonplace to pursue stress-specific mechanisms of action in the
environmental sciences, it is important to recognize that many chronic-level
stresses may share a common mechanism of action at levels of organization
ranging from the individual cell to that of the entire species (Parsons,
1990). At the biochemical level, this hypothesis states that plants possess a
common means of (i) perceiving a stressful environment and (ii) translating
the stress into a biological response. This hypothesis is analogous to the
mode of multiple stress interactions in mammalian systems in which the
response to disparate chronic-level stresses at the organismal level is
orchestrated by the balance of hormones. In analogous manner in plants, it is
proposed that a comparable system operates, based on the role of phytohormones
including abscissic acid, stress ethylene and cytokinens. The significance to
the individual, population and species is that it allows for individually
small environmental stresses to cumulatively and collectively influence
fitness in a magnitude that mimics effects of a single dominant stress factor.
This fosters recognition of the concept in ecology that stress interactions
are very important in governing the physiology, growth and reproductive
success of terrestrial vegetation.
BIODIVERSITY
Biodiversity is a general term for describing the variability that exists at
the level of ecosystems, communities, species, and populations. In its basic
meaning, it can be simplified to the issue of genetic diversity, focusing on
variability in the gene pool. In this context a distinction is not drawn on
individually unique species (e.g., California condor, spotted owl) or
communities (e.g., coral reefs) but rather on the collection of genes that
exists at a given level of organization. Consequently, the specific
configuration or "stoichiochemistry" of the genes are of less importance than
the degree of diversity.
The issue of biodiversity has evolved from one that focused exclusively on
individual species to one that now is synonymous with entire communities or
ecosystems, in which case the term conservation biology is more descriptive.
This progression reflects the concept that while endangered species are
24
-------
important, a more pressing issue is the genetic diversity of the biosphere
including "umbrella species", keystone species, and the array of co-dependent
organisms. This broader context embraces a varied class of less obvious
organisms including decomposers (microbes, litter invertebrates), dispersal
agents (ants, bees, birds and moths), less dominant producers in the overstory
and understory, and array of consumer species.
The issue of biodiversity is inextricably linked to environmental stresses
since it is well documented that the most prominent declines in biodiversity
either locally, regionally or globally, are or were driven by exogenous rather
than endogenous factors (Cronin and Schneider, 1990). Notable examples in a
geological context are the marked reductions in biodiversity that characterize
transitions in time in the Ordivician, Devonian, Permian, Triassic and
Cretaceous periods. In some cases (e.g., Permian) the change in environment
resulted in reductions in marine animal species between 77 and 96 percent.
More recent examples indicate that even local changes in environment can have
effects on the species richness (number of species) and equitability (evenness
in abundance) within a region. Notable examples include the Mt. St. Helens
eruption in the Pacific Northwest and glaciation in the northern hemisphere.
Biodiversity may also be responsive to anthropogenic stresses as evidenced on
a local or regional scale by predictable changes in species' richness or
equitability due to large scale activities of urbanization, forestry, and
agriculture. These are all driven by radical changes in land use or
fragmentation of habitats available for colonization. More subtle and
inadvertent effects are also well documented from such activities as the
introduction of exotic species (e.g., chestnut blight), emission of noxious
gases (e.g., Copper Hill, Tennessee), toxic byproducts from mining activities
(e.g., heavy metals), application of pesticides, and misapplication of toxic
wastes.
However, the more societally prominent issues are those that are global in
scale and commonly discussed in terms of deforestation in the tropics and the
anticipated impact of global climate change (temperature, rainfall,
ultraviolet B radiation on high latitude ecosystems). The effects of
deforestation are driven by the rate at which tropical landscapes are modified
through the deliberate harvesting of dominant species and the more pervasive
habitat destruction for the myriad of other producer, consumer, and decomposer
species in the ecosystem. While the rate of habitat destruction is alarming
in and of itself, the more salient aspect is that these landscapes are areas
of immense and uncharted biological diversity at all levels of organization.
The case of endangered species is only one component of biodiversity and is
really a relic in the broader context of conservation biology. However, in
spite of its relic nature, the concept of endangered species is particularly
relevant to this workshop because of its importance in a regulatory context.
This reflects the fact that endangered species exhibit several key attributes.
First is a biological underpinning that states that an endangered species'
finite numbers reflect a species-wide P:R ratio that approaches one such that
any new stress will further erode the individual's ability to survive and
reproduce (i.e., Darwinian fitness). Second is the tenet that unlike more
widely distributed species, the current habitat of an endangered species is
25
-------
the only area in which survival is possible, thus limiting the species'
options. These endogenous factors predispose a species or population to be
impacted by a new stress. In conjunction with their limited distribution,
endangered species embody a number of easily identified aspects of the issue
of conservation biology and can be addressed in a specific regulatory manner
on a local or regional scale.
In the context of this workshop, the most likely and prominent impact of
pesticide use on biodiversity will be twofold. The first is an effect on
endangered species, and this will continue to be a significant regulatory
issue in the near term. From an ecological perspective and recognizing the
broader definition of biodiversity, the greater impact will occur or is
occurring at the population level as inadvertent exposure results in local
populations being eliminated or sufficiently challenged by the stress to
result in shifts in the genetic structure. These effects are most likely to
occur in areas of repeated application. The net consequence at the community
level will be elimination of species' populations or a progressive shift in
gene pool variability. While an a priori assessment of effects is subject to
large uncertainty, the probability that a population would be eliminated would
be a function of intrinsic sensitivity and the P/R ratio. In the case of
microevolution, the probability of a response is largely dependent on
intrinsic genetic variation within the population upon which natural selection
can operate.
RELATIVE SIGNIFICANCE OF NATURAL AND ANTHROPOGENIC
STRESSES
At a global scale the effects of anthropogenic stress on biodiversity are well
documented, and the impact is accelerating. In some respects the patterns of
response are similar to those periods in geological time in which biodiversity
was markedly impacted by changes in the environment. These similarities
include (i) preferential loss of endemics and species in more northern and
southern latitudes, (ii) vulnerability of the tropics, and (iii) role of
"indirect" extinction whereby habitats are fragmented, thus limiting a
species' ability to colonize. There are, however, a number of unique aspects
to the current impact of anthropogenic stress on biodiversity at the global
scale. These include (i) anthropogenic origin of the stress as compared to
the more natural processes of vulcanisms, asteriod impact, etc., (ii)
accelerated speed of biodiversity changes in space and time as compared with
previous episodes, (iii) number of species being eliminated, (iv) greater
quantitative decline reflecting the combination of species and rate of
extinction, (v) anticipated, muted rate of recovery as compared with that
observed in geological time (5-10 million years), and (vi) preferential
destruction of the tropics, which is likely to dampen the rate of recovery
since tropic ecosystems have served as a "powerhouses" or refugia for species
colonization and macroevolution. Consequently, in comparison to more natural
stresses that influence biodiversity, the current sweep of anthropogenic
stresses tend to operate in far shorter time scales and substantially larger
geographical areas. Two particularly unique aspects of anthropogenic
intervention is the fragmentation of habitats and the unnatural pattern and
26
-------
intensity with which anthropogenic stresses are applied. The consequences for
biodiversity are not resolved. Moreover, an analogous situation due to a
natural stresses is not likely to exists, which would provide a setting with
which to predict the consequences for biodiversity.
REFERENCES
Cronin, T.M. and Schneider, C.E. 1990. Climatic influences on species:
evidence from the fossil records. Trees. 5: 275-279.
Holt, T.D. 1990. The microevolutionary consequences of climate change.
Trees. 5: 311-315.
Johnson, D.W. and Taylor, G.E. 1989. Role of air pollution in forest decline
in eastern North America. Water, Air & Soil Pollut. 48: 21-43.
Levitt, J.A. 1972. Response of Plants to Environmental Stresses. Academic
Press, NY.
Mooney, H.A. 1972. Carbon balance in plants. Annual Rev. Ecol &
Systematics. 3: 315-346.
Parsons, P.A. 1990. The metabolic cost of multiple environmental stresses:
implications for climate change and conservation. Trees. 5: 315-317.
Tingey, D.T. and Taylor, G.E., Jr. 1982. Variation in plant response to
ozone: a conceptual model of physiological events. J_n: M.H. Unsworth and
D.P. Ormrod (eds.), Effects of Gaseous Air Pollution in Agriculture and
Horticulture. Butterworth Scientific, London. 113-138 pp.
Woodwell, G.M. 1970. Effects of pollution on the structure and physiology of
ecosystems. Science. 168: 429-433.
27
-------
ASSESSMENT OF PUBLISHED LITERATURE CONCERNING
PESTICIDE INFLUENCE ON NONTARGET PLANTS
by
John S. Fletcher
University of Oklahoma
Content: Summary data from the PHYTOTOX database were presented for the
most often used plant genera and species in phytotoxicity testing.
These lists of plants were compared to lists of plants making up
the dominant species present in major U.S. ecosystems and those
plants on the U.S. list of endangered species. How frequently
different test methods are used and the comparative sensitivity of
various methods were also discussed.
INTRODUCTION
There has been a substantial amount of data published during the past 60 years
regarding the response of vascular plants to treatment with xenobiotic
chemicals. The PHYTOTOX Database developed at the Univ. of Oklahoma (1)
provides a convenient means of assessing the general makeup of published
phytotoxicity studies and data resulting therefrom. The analyses reported in
this paper considers specific questions dealing with the use of either test-
species or test-procedures. Questions addressed include:
I. Plant Species
1. What plant species have been used most often in plant testing?
2. What do we know about the sensitivity of endangered species to
pesticides?
3. How reliable is it to extrapolate test results from one taxa to
another?
4. How well do EPA's surrogate species represent potential nontarget
vegetation?
II. Tests
5. How frequently have different tests been used?
6. What are the most frequently used endpoints?
7. Where are tests most often conducted?
8. How do laboratory results relate to field results?
28
-------
MOST OFTEN USED SPECIES
Examination of the species found most frequently in PHYTOTOX (Table 1) shows
that data in the literature predominantly relate to plants of agronomic
importance as either food, forage or noxious weeds. In the database, 42% of
the records deal with only 20 plant species, of which 19 are agronomic plants,
pigweed being the exception. Thus, a taxonomic evaluation of the data in
PHYTOTOX shows that the database is heavily biased toward north-temperature
agricultural species. The two most important families used in North American
and European agriculture (Gramineae and Leguminoseae) account for about 40% of
the total number of records. Important temperate nonagricultural plants (such
as timber trees) are not well represented.
The information on natural vegetation pertains primarily to only a few genera
and species (2). Twenty genera account for 47.8% of the records pertaining to
plants growing in the wild, and 10 genera for 61.9% of the old-field records.
Furthermore, 33.6% of the wild and 59.7% of the old-field records deal,
respectively, with 20 and 10 individual species. The limited number of
records in PHYTOTOX that deal with plants growing in natural habitats and the
confinement of this information to only a few genera and species make it clear
that inadequate research attention has been given to ascertaining the
influence of chemical insult on natural plant populations. To further
emphasize this point, PHYTOTOX possesses data pertaining to 2,057 different
plant species, which represents only 12% of the approximately 17,000 native
and cultivated species growing in the United States.
ENDANGERED SPECIES
It is distressing that of the 211 plant species listed on the Federal
Endangered Species List only one of these species, Eriognum ovalifolium. is
represented in PHYTOTOX, and only a single record is present for this plant.
The 211 Federally Endangered Species are distributed among 163 genera. When
PHYTOTOX was searched for records pertaining to these genera only 48 of them
were represented.
EXTRAPOLATION OF DATA BETWEEN SPECIES
The influence of taxonomic differences on plant response was examined by
making EC^ comparisons between taxa at different taxonomic levels (3) as
previously done by Suter et al. (4) for fish toxicity data. This type of
analysis showed that the sensitivity of plant species to chemical exposure is
strongly correlated with their taxonomic classification. When the EC^, values
of 2 taxa at the same taxonomic level were compared, such as species in a
genus, genera in a family, families in an order, and orders in a class, the
respective coefficients of determination were 0.868, 0.559, 0.134, and 0.081
(Table 2). These results are quite different from those reported by Suter et
al. (4) for fish toxicity data, where it was shown that at all taxonomic
levels from species to order a comparison of paired taxa yielded high
29
-------
coefficients of determination. These analyses indicate that great care must
be exercised in extrapolating test results from one species to another unless
they belong to the same genus, and even this practice is questionable because
of the limited research specifically addressing this issue.
SURROGATE SPECIES
The Pesticide Assessment Guidelines Subsection J from the U.S. EPA requires
testing of 6 species in at least 4 families, and the OECD requirement is 3
species from 3 families. Although the analyses presented in this paper (Table
3) indicate that multifamily testing of plants as required by EPA and OECD is
a commendable policy, the analyses also cast some suspicion on the suitability
of the current lists of recommended plants (Table 3). Of the 300 families in
the plant kingdom, 8 are represented on EPA's list and 4 on OECD's list.
Review of these lists show that native species have not been included, and of
greatest concern is that some major families of native and cultivated plants
in the U.S. have been ignored. Most noticeable is the absence of any
representatives of the Fagaceae (oaks, beeches, chestnuts) Pinaceae (pines,
spruces, fir) and Roseacea (apple, pear, peach) families. Regarding these
families there are regions in the United States where agroecosystems may be
described as a patchwork of cultivated crops and native wood lots, which in
some cases are in close proximity to industrial centers. Under circumstances
where agriculture, industry, and native vegetation co-exist, the question
arises as to whether or not the currently recommended surrogate species
provide an adequate safeguard for environmental protection against potential
organic pollutants. Because of the inadequate understanding of chemical
toxicity to most native plants, such a reevaluation would have to include
chemical toxicity studies on numerous families of native plants for which we
currently have virtually no toxicology data.
ENDPOINTS
Phytotoxicity tests may be conducted at different stages during the life cycle
of a plant and many different endpoints may be measured. Various combinations
of these two test features give rise to a multitude of distinctively different
test protocols. A rough estimate of the variability in phytotoxicity data
pertaining to these features was determined by tabulating the number of
records in PHYTOTOX dealing with each of these test features. These analyses
indicated that 34% of the reported phytotoxicity data are collected on seeds,
31% with seedlings, 8% with mature vegetation, and only 5% with reproductive
plants. The 10 most frequently measured endpoints listed in descending order
are: size change, plant number, kill, seed germination, fresh weight change,
dry weight change, harvest yield, leaf abscission, respiration, and deformed
organs (5).
30
-------
CULTURE CONDITIONS
Another important variable in phytotoxicity testing is culture conditions. A
survey of the data in PHYTOTOX shows that 26% of the tests are greenhouse
studies as compared to 23% conducted in cultivated fields (2). Data collected
under natural (wild) and old-field conditions are quite infrequent, 4.7 and
1.5% of the time, respectively. Growth chambers are used for 9.3% of the
reported studies.
LAB VERSUS FIELD RESULTS
The large proportion of data collected in either greenhouse or growth chamber
studies always raises the question of "How well laboratory results reflect the
actual field toxicity of chemicals?" This question was addressed by
evaluating chemical response data held in PHYTOTOX. A comparison was made
between greenhouse vs. field data for 13 different plant species tested with 1
or more of 17 chemicals from 11 different classes of herbicides (3).
Comparative analysis was based on EC^ values. To facilitate this analysis,
response ratios (greenhouse EC^ /field EC^,) were calculated for each plant-
chemical combination. Response ratios <1 indicate greater plant sensitivity
in the greenhouse; whereas, ratios >1 reflect lower sensitivity in the
greenhouse than in the field.
Analysis of the response ratios (Table 4) showed that in 6 of the 20
combinations which were considered, plants treated in the greenhouse are more
sensitive than those treated in the field. In 3 cases sensitivities are
approximately equal, and in the remaining 11 cases the plants in the field are
more sensitive. The lowest response ratio is 0.26 for pigweed treated with
linuron and the highest is 3.26 for red pine treated with simazine.
The magnitude of the variability between sensitivities observed in the
greenhouse versus field were examined without regard for the direction of
differences. For this purpose, a response variability was calculated by
dividing the larger EC^ value by the smaller for each plant-chemical
combination (Table 4). The mean of the individual variabilities was 1.8 with
a confidence interval of ± 0.4 at the 95% level. In general terms, this
indicates a 95% possibility that there will be less than a 2-fold difference
between greenhouse and field sensitivities.
31
-------
TABLE 1 - The most frequently listed species in PHYTOTOX.
Species
Common name
No. of
records
Percentage8
Triticum aestivum
Pisum sativum
Lvcopersicon esculentum
Avena sativa
Phaseolus vulqaris
Mai us sp.
Glvcine max
Zea mays
Hordeum vulgare
Linum usitatissimum
Cucumis sativus
Nicotiana tabacum
Amaranthus retroflexus
Oryza sativa
'Solanum tuberosum
Gossy&ium hirsutum
Lactuca sativa
Raphanus sativus
Echinochioa crusgalli
Beta vulqaris
Setaria yiridis
Prunus perslea
Digitaria sanquinalis
Cynodon dactyl on
Brassica kaber
Poa pratensis
Sinapsis alba
Festuca arundinacea
Ipomoea purpurea
Sorghum halepense
Lolium perenne
Trifolium repens
Medicago sativa
Cyperus rotundas
Punicum miliaceum
Pinus taeda
Totals
Wheat
Pea
Tomato
Oats
Bean
Apple
Soybean
Corn
Barley
Flax
Cucumber
Tobacco
Pigweed
Rice
Potato
Cotton
Lettuce
Radish
Barnyard grass
Sugar beet
Green foxtail
Peach
Crabgrass
Bermuda grass
Mustard
Kentucky bluegrass
White mustard
Tall fescue
Morning glory
Johnsongrass
Ryegrass
Clover
Alfalfa
Purple nutsedge
Millet
Loblolly pine
4,810
3,757
2,349
2,148
1,767
1,619
531
1,465
1,462
1,401
1,353
1,324
1
1,184
1,193
1,001
1,044
982
960
907
677
642
621
555
523
522
506
463
442
438
420
398
350
306
284
269
157
39,830
1
6.2
5.8
3.0
2.8
2.3
2.1
2.0
1.9
9
1.8
1.7
1.7
1.5
.5
1.3
1.3
1.3
1.2
1.2
1
0.9
0.8
0.8
0.
0.
0,
0.
0.6
0.6
0.6
0.5
0.5
0.4
0.4
0.4
0.3
0.2
51.2
"Percentage of the 77,825 records in the database. (Taken from reference 2.)
32
-------
TABLE 2 - Comparison of ECM values at four taxonomic levels.
Taxon 1
Ipomoea lacunosa
Ipomoea lacunosa
Ipomoea quamoclit
Ipomoea quamoclit
Ipomoea purpurea
Ipomoea purpurea
Ipomoea lacunosa
Ipomoea quamoclit
Ipomoea purpurea
Ipomoea wrightii
Amaranthus retroflexus
Amaranthus retroflexus
Amaranthus retroflexus
Setaria viridis
Setaria viridis
mean
Digitaria
Festuca
Panicum
Echinochloa
Phalaris
mean
Cyperaceae
Amaranthaceae
Caryophyllaceae
Rosaceae
mean
Asterales
Asterales
Liliales
Papaverales
Rosales
Resales
mean
Taxon 2
Species in genus
Ipomoea quamoclit
Ipomoea wrightii
Ipomoea wrightii
Ipomoea purpurea
Ipomoea lacunosa
Ipomoea hederacea
Ipomoea hederacea
Ipomoea hederacea
Ipomoea wrightii
Ipomoea hederacea
Amaranthus hvbridus
Amaranthus palmeri
Amaranthus spinosus
Setaria faberi
Setaria glauca
Genera in family
Echinochloa
Echinochloa
Setaria
Eleusine
Sorghum
Families in an order
Poaceae
Chenopodiaceae
Portulacaceae
Leguminoseae
Orders in a class
Chenopodiales
Pol emoni ales
Poales
Polemoniales
Euphorbiales
Scrophulariales
na
4
3
3
3
4
5
4
3
3
3
5
5
8
3
3
3
4
4
3
3
35
11
7
10
24
26
9
13
16
9
r2b
0.984
0.988
0.998
0.998
0.919
0.997
0.931
0.996
0.994
0.997
0.971
0.484
0.841
0.752
0.236
0.868
0.645
0.146
0.481
0.886
0.841
0.559
0.183
0.132
0.065
0.013
0.134
0.024
0.001
0.003
0.361
0.126
0.060
0.081
"The number of different chemicals tested on each pair of taxa.
"The coefficient of determination log transformed EC^ values.
(Taken from reference 3.)
33
-------
TABLE 3 - Surrogate Test Species Recommended by the EPA and the OECD.
Common name
Latin name
Family
U.S. EPA/FIFRA
lettuce
cabbage
cucumber
soybean
onion
corn
oat
ryegrass
tomato
carrot
OECD
lettuce
Chinese cabbage
cress
mustard
radish
rape
turnip
fenugreek
mungbean
red clover
vetch
oat
rice
ryegrass
sorghum
wheat
Lactuca sativa
Brassica oleracea
Cucumis sativus
Glvcine max
Alii urn cepa
Zea mays
Avena sativa
Lolium perenne
Lvcopersicon esculentum
Daucus carota
Lactuca sativa
Brassica campestris
Lepfdium sativum
Brassica alba
Raphanus sativus
Brassica napus
Brassica rapa
Trifolium ornithopodioides
Phaseolus aureus
Trifolium pratense
Vicia sativa
Avena sativa
Orvza sativa
Lolium perenne
Sorghum bicolor
Triticum aestivum
Asteraceae
Brassicaceae
Cucurbitaceae
Leguminoseae
Liliaceae
Poaceae
Poaceae
Poaceae
Solanaceae
Umbelliferae
Asteraceae
Brassicaceae
Brassicaceae
Brassicaceae
Brassicaceae
Brassicaceae
Brassicaceae
Leguminoseae
Leguminoseae
Leguminoseae
Leguminoseae
Poaceae
Poaceae
Poaceae
Poaceae
Poaceae
(Taken from reference 3.)
34
-------
TABLE 4 - Comparison of ECM Values of Greenhouse and Field Treated Plants.
Plant Name
Chemical Name
EC,,, (Kg/ha) Response Response
ratio difference
Herbicide Class* Greenhouse Field
Aqropyron repens
Amaranthus retroflexus
Amaranthus retroflexus
Amaranthus retroflexus
Amaranthus retroflexus
Amaranthus retroflexus
Apocynum cannablnum
Avena fatua
Avena fatua
Diqitaria sanqulnal Is
Eleusine Indica
Festuca arundlnacea
Glycine max
Pinus resinosa
Pinus resinosa
Poa annua
Sorghum halepense
Triticum aestlvum
Zea mays
Zea mays
Quackgrass
Pigweed
Pigweed
Pigweed
Pigweed
Pigweed
Dogbane
Wild oats
Wild oats
Crabgrass
Goosegrass
Tall fescue
Soybean
Red pine
Red pine
Bluegrass
Johnson gr.
Wheat
Corn
Corn
Glyphosate
Linuron
Prometryn
Dacthal
Atrazine
Chloramben
2,4-D
Dicl of op-methyl
Triallate
Prometryn
Triflural in
Alachlor
Chloroxuron
Atrazine
Simazine
Ethofumesate
Dalapon
Barban
Cl of op-methyl
Cl of op-methyl
Organophosphates
Ureas
Triazines
Aromatic acids
Triazines
Aromatic acids
Phenoxyal kanates
Phenoxyal kanates
Thiocarbamates
Triazines
Nitroanil ines
Anil ides
Ureas
Triazines
Triazines
Heterocycl ic
Haloal kanates
Carbamates
Phenoxyal kanates
Phenoxyal kanates
0.
0.
1.
7.
1.
1.
0
0.
1.
0,
1.
0,
4.
2.
6.
0.
4.
6.
0.
0.
.80
.60
.40
.54
.18
.70
.15
.55
.96
,76
,14
,56
,39
99
,95
55
,18
,92
61
,56
1
2
2
4
1
1
0
0,
2,
2,
0,
0,
2,
0,
2,
0.
2.
3,
0,
0,
.55
.30
.40
.98
.10
.36
.12
,63
.24
.12
.66
,52
,58
,93
,13
.55
.70
,36
,51
,44
0
0
0
1
1
1
1
0
0
.52"
.26
.58
.51
.07
.25
.25
.87
.88
0.36
1
1
1,
3,
3,
1.
1.
2,
1.
1,
.72
.08
.70
.22
.26
.00
.55
.06
,19
,27
mean of response difference
1.
3,
1.
1,
1
1
1
1.
1.
2,
1.
1.
1.
3.
3.
1.
1.
2.
1.
1.
1.
.94e
.83
.71
.51
.07
.25
.25
,15
,14
,79
,72
,08
,70
22
26
00
55
06
19
27
,78
"Herbicide classes as assigned by Fletcher and Kirkwood (6).
bGreenhouse EC,,, divided by the field ECM.
'Larger ECM value divided by the smaller ECW value for each plant-chemical combination.
(Taken from reference 3.)
35
-------
REFERENCES
Royce, C.L., Fletcher, J.S., Risser, P.R., McFarlane, J.C. and Benenati, F.E.
1984. PHYTOTOX: A database dealing with the effect of organic chemicals on
terrestrial vascular plants. 0. Chem. Inf. Comput. Sci. 24:7-10.
Fletcher, J.S., Johnson, F.L. and McFarlane, J.C. 1988. Database assessment
of phytotoxicity data published on terrestrial vascular plants. Environ.
Toxicol. Chem. 7:615-622.
Fletcher, J.S., Johnson, F.L. and McFarlane, J.C. 1990. Influence of
greenhouse versus field testing and taxonomic differences on plant sensitivity
to chemical treatment. Environ. Toxicol. Chem. 9:769-776.
Suter, G.W. II, Vaughan, D.S. and Gardner, R.H. 1983. Risk assessment by
analysis of extrapolation error: A demonstration for effects of pollutants on
fish. Environ. Toxicol. Chem. 2:369-378.
Fletcher, J.S. 1990. Use of algae versus vascular plants to test for
chemical toxicity. In: W. Wang, J.W. Gorsuch, and W.R. Lower (ed.), Plants
for Toxicity Assessment. ASTM, STP 1091. 33-39 pp.
Fletcher, W.W. and Kirkwood, R.C. 1982. Herbicides and Plant Growth
Regulators. Granada Publishing, London, U.K. 15-68 pp.
36
-------
CIS-BASED RISK ASSESSMENT: APPLICATIONS FROM AN
APPROACH TO OZONE RISK ASSESSMENT TO ASSESSING
RISK OF PESTICIDES TO NONTARGET ORGANISMS AND
ECOSYSTEMS
by
W.E. Hogsett* and Andy Herstrom
U.S. EPA and METI, Inc.
INTRODUCTION
The risk assessment of a particular stress, either anthropogenic or natural,
to a species, community, habitat or ecosystem involves; (1) a description and
estimation of the magnitude and spatial (geographical) extent of the stress;
(2) the geographical distribution of the resource under assessment, i.e.,
species, communities, habitats, or ecosystems; (3) an estimation or
quantification of the sensitivity of this resource to the stress in question;
and, (4) a description of the influential environmental/climatic factors that
modify the sensitivity of the resource and an estimation of the altered
sensitivity. The interaction of these components (multiplicative or additive)
provides a measure or estimation of risk, i.e perceived risk, to the resource
at various spatial scales, including county, region or national. We are in
the process of developing such a spatial-based risk assessment using the
Geographical Information System (GIS) for estimating the extent and magnitude
of a regional criteria air pollutant, ozone, and its potential impact on
forests in the United States. The GIS-based assessment is an interactive tool
capable of considering various assessment scenarios, because the model
includes interactive factors both in the formation and dispersal of ozone as
well as those factors influencing the response of a species to ozone.
Predictions will be possible for vulnerability of particular tree species to
current ozone levels in particular regions, as well as predicted future levels
or ozone levels expected if various control strategies are applied.
Perceived risks can also be made across different site conditions and climate
scenarios. Also interaction with other stresses in which the response to one
is exacerbated or mitagated as a result, e.g., interaction of ozone and bark
beetle. These interactive stresses can be layered in the GIS to enhance the
stress or sensitivity.
This same approach would be useful in assessing the potential adverse effect
of pesticides and their usage on various resources in the U.S.; including
nontarget species, community structure, habitats, and ecosystem function.
*Presenter
37
-------
Such assessments are needed to describe the field behavior of these chemicals
under various application protocols and under various meterological factors
potentially influencing their dispersal such as seasonal wind speed.
The first step in understanding and assessing the potential impact of an
anthropogenic stressor on a plant or animal species or on an ecosystem and its
integrety as expressed in its function and structure is knowledge of the
spatial extent and magnitude of the exposure. In the case of a regional
pollutant like ozone that extent can be quite large, but there is a paucity
of air quality monitoring data in the areas under consideration, i.e.,
forested areas of the United States. Yet for a risk assessment you ultimately
need an estimation of the exposure in those nonmonitored areas. Thus there is
the need to estimate or interpolate ozone concentration in nonmonitored areas.
Two approaches to make this estimation are available. The one currently in
use is various interpolation techniques such as distance weighting or kriging
(Lefohn et al., 1987). These techniques are based on distance between
monitoring sites and the geometric placement of the sites relative to each
other. The accuracy of these interpolation techniques depend on the density
of monitored sites, the representativeness of the site of its surrounding
area, and the behavior of ozone with time. Alternatively, an estimation of
the exposure can be made by modeling the spatial extent of those factors
involved in the formation and transport of the pollutant. In the case of
ozone that includes emission source strength, both volatile organic carbons
(VOCs) and NOX, sunlight, wind speed and decay rate. This technique does not
estimate ozone concentration in a nonmonitored area simply based on a
triagulation of concurrent sites, but rather uses those chemical and
meterological factors known to contribute the formation and transport of
ozone. This is especially critical in the case of ecosystems or habitats
located outside of intensely monitored areas of the United States. A
Geographical Information System (GlS)-based model is being used for the
development of spatial analysis procedures as an alternative to traditional
interpolation techniques, as well as compiling all risk factors for a
perceived risk of the resource to ozone.
This model-based approach is directly relevant to the case of pesticide
application and risk to nontarget organisms. An estimation of exposure would
include some of these same factors, i.e. extent of application, manner of
application, extent of drift (dependent on meterological and climatological
data including wind speed, direction), and degradation rate of the pesticide.
This estimation of exposure combined with the geographical extent of the non-
target organisms of interest or the habitat of interest and the sensitivity of
the resource would yield a perceived risk, as well as provide an interactive
tool for estimating risks under various meterological conditions or various
application protocols.
GIS MODEL
The objective of the model is to qualitatively categorize areas according to
ozone exposure potential at a national spatial scale and a monthly temporal
scale based on the assumption that locations in close downwind proximity to
38
-------
areas characterized by: 1) large sources of ozone precursosrs, VOC and NO,;
2) great amounts of solar radiation; and 3) large numbers of days with calm
or weak surface winds, will have a greater potential for experiencing elevated
ozone exposure than locations not situated in such high potential downwind
areas. The spatial scale could also be regional, state, or even county
depending on the needs of the assessment. This is accomplished by modeling
the major factors influencing ozone formation and transport. These factors
include emissions of ozone precursors, VOCs and NOX, temperature, stagnating
air masses, wind speed, distance and wind direction from emissions sources,
decay rate of ozone and changes in elevation.
The model considers counties as sources of anthropogenic VOC which is
transformed into ozone as a function of temperature (surrogate for solar
radiation) and wind speed (air stagnation masses). The ozone is then
dispersed via 16 plumes (a wind rose) radiating out from each source and
decayed as a function of distance and wind speed. The formation and dispersal
from Fulton County Georgia is shown in Figure 1 as an example. Ozone plumes
from all sources are summed to produce an estimate of ozone exposure potential
over the study region.
RELATIVE OZONE COK1RIBU1ION
r~jHt Codrikilisn
QJlmnl CtntfiktlUn
Figure 1 - Formation and dispersal of ozone from Fulton County, Georgia.
39
-------
Conceptually, the model for estimating ozone exposure potential consists of
computerized map layers of each of the factors considered in the formation ,
transport, and decay of ozone. Each map layer is precisely registered to each
other, and is treated as a matrix of real numbers with each number describing
a cell within a grid superimposed over the earth s surface. These numbers
describe some attribute or characteristic of that cell such as land use type,
termperature, precipitation, wind speed, air quality, etc. In the case of the
ozone example, the study area of the eastern U.S. consists of 18,900 cells and
each cell is 20 Km square.
GIS technology uses the digital nature of computerized maps in a process
called "cartographic modelling." With this process, maps are no longer viewed
merely as pictures describing the location of features, but as matrixes of
numbers with spatial and analytical significance. By manipulating these
matrixes in a series of ordered mathematical expressions, a map algebra is
created that generates new information from existing data (Tomlin, 1990;
Berry, 1987).
The cartographic model (Figure 2) consists of three parts which: 1) form ozone
as a function of VOC emissions, temperature and wind speed; 2) disperse this
newly formed ozone as a function of wind direction and decay and dilute the
ozone as function of time; 3) sum the ozone contribution each source makes on
every cell in the study area grid to create a final data layer representing
the relative ozone exposure potential of each cell in the study area. The
computations are graphically depicted in Figure 2. The model results for the
Eastern U.S. are shown in Figure 3. This map illustrates the predicted ozone
exposure potential using 417 counties and their VOC emission for 1985, air
temperature daily maximums for July 1985, 10-year July average wind direction
and wind speed values. The exposure intensity increases with darkerr color
and represent the numerical calculations from the map algebra depicted in the
model in Figure 2.
An advantage modelling has over interpolation is that it allows an estimation
of exposure based on the factors making up that exposure, i.e., addresses the
question of "why" an area is at greater exposure risk than another. The GIS
model also allows an assessment of the effect changes in influential
environmental factors will have on ozone exposure as well as to predict ozone
exposure potential under a given set of "what if" questions.
At the present time we consider the model to be "in development." Tasks to be
accomplished during this development phase are to: 1) identify important
spatial factors needed to model ozone formation, transport, and exposure; 2)
identify the spatial and temporal scales needed; 3) identify data bases that
meet the criteria in steps 1 and 2; 4) identify the analytical processes to be
performed on the data; 5) identify gaps in knowledge for relevant assessment
of forests; and 6) identify computer software and hardware to perform these
analytical processes at a national level.
40
-------
15 VOC Source (15 Units)
,5 X
Average Temperature Factor
I 75 *
Average Stagnating Air Mass Factor
LJ i i i i r Cm I I > r i i i i i r i n n-E.n 11 ; i i. in i i i i i i i i FTi i i i r i i i i i i iTI
22.3
Miiiiiniiiiiiiiim Ozone Formed Within 6 Hour Zone
x
62? 751 87* 1001 871 751 62* .. _
Minija^uin mi ••••BBIIIIB Decoy Factor
'3.8 15.7 19.4 22.3 19.4 16.7 13.8 *
j-i ••*«i i in1 ••••••• 11 mini lima Ozone Values In All Zones
20X IQX
percen(age
2.6 3.2 3,8 4.4 2.2 1.9 1.8 1.3 ... . . ,,./«« i
Ozone Value (One Source)
,ii.iMj,nini.in!Ui.i-.
Value
' ...... '"""•'•" ....... ii"1 ..... •<•!•.. i. .-..-..... ..... .r,. pina| Ozone value (Source fN)
8
••••^ millim^' i^— i^— «i^— «— »i"— • Ozone Exposure Potential
Figure 2 - Graphic presentation of ozone exposure potential predicted by the CIS
model.
41
-------
ESTIUATED
EXPOSURE POTENTIAL
1 .000
1 ,001-3,000
3.001-6,000
6,001-10,000
>10,000
Figure 3 - Exposure potential for ozone in eastern United States as estimated with
the CIS model.
42
-------
The model for ozone exposure potential will be part of a larger GIS-based risk
assessment considering tree species, geographic and habitat distribution,
species sensitivity to ozone, and growing conditions affecting the sensitivity
of forests to ozone. The model for perceived risk of species is shown in
Figure 4. An overlay of the base case and the species distribution of
Liriodendron tulipfera. an ecologically important tree species in the Eastern
Hardwood and Southeastern Mixed Hardwood forests, illustrating potential areas
of high ozone exposure and a sensitive tree species. The model for perceived
risk would incorporate these spatial data, and would have the additional
layers of species sensitivity based on growth response functions to ozone, and
an altered sensitivity based on the influence of various environmental factors
to yield a spatial representation of ozone risk to Liriodendron tulipfera.
=EŁED Species Sensitivity
333 Environmental Factors
"'-" ....... " ........ ' .............. ...
^
Ozone Exposure Potential
233333 Perceived Risk
Figure 4 - Model for perceived risk of species.
PESTICIDE RISK
A similar GIS-based assessment model could be envisioned for assessing
perceived risk of nontarget plant or animal species, or even habitat or
ecosystem function to pesticide application in nearby agricultural lands.
Such assessment would be beneficial for pesticide field evaluation studies,
pesticide registration processes, land-use and land management evaluations,
and the recent concern for habitat modification. A similar set of model
components or data layers for the GIS could be suggested. These would
include:
* Crop distributions
* Pesticide usage
* Nontarget organisms and their geographical distributions
43
-------
* Estimation of exposure using those factors that contribute
to non-target exposures
- application methods
- volatility
- air temperature
- wind speed
- wind direction
An example of a possible approach might involve assessment of pesticide
application and potential drift exposure of nontarget species, e.g. threatened
and endangered plants. Such a data set is shown in Figure 5, and illustrates
the potential areas of drift exposures in the areas having wind speeds greater
than 3 knots, and the spatial distribution of endangered and threatened plant
species across the United States. This map does not yet have incorporation of
crop and pesticide usage, and sensitivity of nontarget species to the
pesticide. All of this information could be acquired or developed. Again, as
with the ozone exposure potential and ozone perceived risk, the GIS-based
assessment is foremost an interactive tool to examine various usage and
climate scenarios that may help guide research in areas where there is lack of
information on environmental interaction, and help policy decisions based on
usage, land management, and other factors important in the evaluation of
pesticide registration in the United States.
44
-------
PERCENT TIME
WINDSPEEDS ARE
CALM TO 3 KNOTS
>30
Federally Listed
Threatened and
Endangered Plants
Figure 5 - Federally listed threatened and endangered plants and % time wlndspeeds are from calm to three knots.
-------
REFERENCES
Berry, O.K. 1987. A mathematical structure for analyzing maps.
Environmental Management. 11: 317-325.
Lefohn, A.S., Knudsen, H.P., Logan, J.A., Simpson, J. and Bhumralker, C.
1987. An evaluation of the kriging method to predict 7-h seasonal mean ozone
concentrations for estimating crop losses. J. Air Pollut. Control Assoc. 37:
595-602.
Tomlin, C. Dana. 1990. Geographic Information Systems and Cartographic
Modeling. Englewood Cliffs, N.J.: Prentice Hall.
46
-------
SESSION III: LABORATORY AND GREENHOUSE TESTING (TIERS I AND II,
SUBDIVISION J)
The first paper in this session analyzed the tier tests as they are currently
used. The remaining papers described hew test procedures which may be used to
improve certain aspects of the current tests.
47
-------
DIFFICULTIES IN PERFORMING EXISTING TIER I
AND II TESTS IN SUBDIVISION J GUIDELINES
by
Joseph Gorsuch
Eastman Kodak Company
Content: Difficulties encountered in conducting the tier I and II tests in
Subdivision J were discussed. An overview of the current tests
was presented with comments pertaining to the need for
clarification and standardization of the existing tests. Specific
items discussed included: test specifications, choice of test
species, environmental considerations, test facilities, chemical
application, soil media, visual observations, and statistics.
INTRODUCTION
This document was prepared to identify current areas of difficulty in
performing the Tier I and Tier II Terrestrial Plant Tests as described in the
original "Pesticide Assessment Guidelines, Subdivision J, Hazard Evaluation:
Nontarget Plants"(l) and updates (2,3). Although there are some definite
"difficulties" in performing various aspects of these terrestrial plant tests,
my focus has been rather on, "needed considerations" in performing these
tests. Although not perfect, these test guidelines have provided a means of
evaluating potential adverse effects on nontarget plants. To my knowledge
these tests (seed germination/seedling emergence and vegetative vigor) have
never been standardized outside the Agency. The issue of standardization will
be further addressed in the Recommendations Section of this document.
Currently, an industry wishing to register a chemical or pesticide that may
come in contact with nontarget plant species, may be faced with performing
plant bioassays following one or more test guidelines (e.g., EPA/FIFRA,
EPA/TSCA, OECD, or FDA). Although there are some similarities in each of
these test guidelines, for the most part, they are quite dissimilar. For
example, the FDA guidelines for evaluating germination and seedling growth are
both quite rigorous (e.g., requiring many more replicates, more frequent
monitoring, etc.). In contrast, the FIFRA guidelines, the focus of this
document, are more flexible and perhaps more reasonable in scope.
48
-------
Overview of Tier I and II Terrestrial Plant Tests
A brief review of the Tier I and Tier II Terrestrial Plant Tests for Seed
Germination/Seedling Emergence and Vegetative Vigor follow.
It is the intent that these tests be used to determine phytotoxicity, not
efficacy, of a chemical on plants. These tests are used to evaluate the
phytotoxic effects of pesticides on nontarget plants. Tiers I and II are
designed to "screen" technical chemicals; however, Tier II does require that
at least five concentrations be tested (considered a "definitive" test). In
Tier I, a single dose which is equal to or greater than the maximum label rate
of the technical chemical is evaluated in each of these tests. In Tier II,
multiple doses of at least five concentrations are required, with one of the
concentrations equal to or greater than the maximum label rate. Exposure
concentrations in the Tier II tests are established in geometric series (less
than or equal to two-fold), with one concentration less than or equal to the
EC50 value, and another concentration at the No Observed Effect Concentration
(NOEC). Tier II testing is triggered if a detrimental effect greater than or
equal to 25% (for these tests a 25% effect is generally considered within
biological variability) is observed in any of the plant species tested in Tier
I. Tier II guidelines require only those plant species affected be retested.
Tier III testing, involving the evaluation of the pesticide under field
conditions, is triggered if detrimental effects greater than or equal to 25%
are confirmed in Tier II tests.
TEST SPECIFICATIONS
The following test specifications for conductance of the Tier I and Tier II
Germination/Emergence and Vegetative Vigor Tests are presented in Table 1
below.
Table 1 - Test specifications.
REQUIREMENTS
No. Plant Species
Dicot Sp/Family
Monocot Sp/Family
Seeds/Replicate
No. Replicates
Length (Days)
Observations
Endpoint Measurements
Germination
10
6/4
4/2
10
3
5
Day 5
Emergence Vegetative Vigor
10 10
6/4 6/4
4/2 4/2
10 5
3 3
14 >14
(14-Day Post Germination)
Day 10 & 14 Day 7, 14, Weekly
[can be
extended to
28 days for
soil vigor]
49
-------
CHOICE OF TEST SPECIES
The Standard Evaluation Procedures (SEPs)(2,3) provide a list of plant species
suitable for these tests. Both the guideline and the SEPs state that corn,
soybean, and a dicot root crop, e.g., carrot, are required species. The
guideline requires that at least ten species be tested, the exact number
listed in the June 1986 SEP. It should be noted that the optimum germination
temperature is not the same for these ten species. For example, ryegrass
germinates best at 20°C while corn and oats germinate best at 30°C.
Therefore, if a germination/emergence test was being conducted in a growth
chamber, it might be necessary to pick an intermediate temperature, e.g.,
25°C, to accommodate these two extremes. Otherwise, it would be necessary to
carry out the germination/emergence test over several test periods with the
optimum germination temperatures being maintained for those plant species
being exposed. The germination response, even at the optimal temperatures, is
generally low for carrots (about 75%) and onions (about 80%). Species with
greater germination response are generally desired; however, alternative
choices of species required to meet the specifications of these guidelines may
not be commercially available. The time for germination response also varies
from two-to-three days for species such as cabbage and lettuce to as long as
six days for tomatoes. Since the germination test is to last at least five
days, where "germinated" is defined as "a radical growth of at least 5 mm,"
this may be too short of a time period for tomatoes. Seeds should be selected
that are "uniform in size" to minimize variability in heights, root length,
etc.
ENVIRONMENTAL CONSIDERATIONS
Environmental considerations are briefly mentioned in the test guidelines (1)
and SEPs (2,3). There were no requirements identified for photoperiod,
temperature control, humidity, and C02 levels. Since FDA, EPA/TSCA, and OECD
all address these parameters, either in detail or in generalities, they should
be addressed in the FIFRA guidelines.
Photoperiod
It is usually not necessary to control the photoperiod for a germination test.
Germination tests can be conducted either in the dark or in light. If a test
chemical is known to photodegrade in the presence of sunlight or light, then
it may be best to perform the test in the dark (or both dark and light). If a
growth chamber is used for the seedling emergence and seedling vigor studies,
then photoperiod should be controlled (length and intensity). Currently FDA
requires a 16-hour light period at greater than or equal to 2,000 fc. Since
much of the U.S. rarely sees 16-hour photoperiods on a continuing basis, a
14-hour photoperiod seems more reasonable. Greenhouses might need
supplemental lighting providing 2,000 fc. In geographical areas that have
limited sunlight, EPA (OPP) has required supplemental lighting be used in
greenhouses. If plants are germinated/emerged indoors then moved to a
greenhouse, they should be shielded from direct sunlight or supplemental
50
-------
lighting during the first 24 hours. A canopy made from cheesecloth placed
above the plants is generally adequate.
Temperature Control
It is not unreasonable to expect temperature fluctuations in a greenhouse
(even with climatic controls). Daytime temperatures often reach 30°C, while
nighttime temperatures dip to 20°C. Trying to maintain a 25 + 5°C temperature
range in a greenhouse works best. Although air temperatures can be and should
be monitored continuously, consideration should be given to periodic
monitoring of soil temperatures at opposite ends of exposure plots. If soil
temperature is likely to drop below 20°C, consideration should be given to
using propagating mats.
Humidity
Humidity is often difficult to control in a greenhouse. Plants should be kept
out of direct drafts in order to control transpiration rates. Humidity is
also difficult to control in growth chambers when trying to maintain
temperatures in a cycling day-night temperature regime, e.g., 30°C day and
20°C night temperatures. Relative humidity values near 50-60% are generally
adequate.
C02
Generally ambient C02 (350 ± 50 ppm) is acceptable. If plant studies are
being conducted near industrial sites in a greenhouse, FDA has expressed
concern that C02 levels might be elevated. Therefore, it may be necessary to
monitor CO, under such circumstances.
TEST FACILITY CONSIDERATIONS
The test guidelines state that germination/emergence tests should be conducted
in a "controlled" environment, in either a growth chamber or a greenhouse. As
discussed in the previous section (Environmental Considerations), the control
within a greenhouse is limited. The guideline goes on to state that
vegetative vigor can be conducted in a growth chamber, a greenhouse, or a plot
of land. Therefore, it appears that less "controlled" environments are
acceptable for vegetative vigor studies. It is doubtful that EPA really meant
for this to be implied. The use of growth chambers provides greater control
of the temperature and humidity; however, their space is limited in comparison
to a greenhouse. They are also expensive ($100,000 for a 125 ft. growth
chamber). Greenhouses on the other hand, provide adequate space, frequently
what is necessary for a large study such as a Tier II test with many plant
species. Greenhouses, however, are more difficult to control the temperature
and humidity, thus excursions are generally greater. In comparison to the
growth chambers, greenhouses are less expensive, for example, a 400 ft.
greenhouse that is fully insulated runs about $80,000. Since supplemental
51
-------
lights most likely are necessary in greenhouses, then the cost would be
increased by $20,000 - $30,000. A major concern in conducting these studies
is crowding the plants, which can lead to chemical contamination of the
control plants, shading of smaller plants, and difficulty in watering. Block
randomization (randomizing within a block or concentration) is recommended
over the total randomization of plants, if space is not adequate to prevent
treated plants from touching each other, or the controls.
UNIFORM APPLICATION OF TEST CHEMICAL
A major area of concern is the application of the test chemical to the surface
of soil or plants. Minimal guidance is provided in the test guidelines on the
application of pesticides to soil and plants.
Foliar Application
Foliar application requires the application of a uniform spray. Handheld
sprayers and commercial applicators can be used if the spray can be uniformly
applied. It may be necessary to affix the wand or boom above the plants and
soil which pass through the spray on a conveyor belt. An alternative, would
be to place the plants being treated out of doors and apply the spray using a
boom mounted on a tractor. Canopy height must also be considered when
applying spray. Taller plants, such as corn, may require the raising of the
wand, while lettuce may require its lowering. Regardless, plants should be
within the cone of application for maximum coverage if the chemical is to be
applied uniformly.
Chemical Mix in Soil
A small portion of a test chemical with low aqueous solubility sometimes must
be mixed with a large volume of soil in order to achieve the application rate.
Coating of sand or glass beads with a test chemical that has been dissolved in
a solvent, e.g., acetone, has also been used to introduce chemicals with low
water solubility. To ensure the thorough mixing of the chemical throughout a
soil matrix, samples can be collected, extracted with the coating solvent, and
analyzed to determine if the desired concentration is present. (See GLP
Considerations Section.)
Subirrigation
The use of subirrigation, placing the test chemical solution and/or water in a
trough or tray in which the pots containing soil are emerged provides for
another means of applying a test chemical. Chemicals applied in this manner
may result in soils exhibiting chromatographic effects and/or salting at the
surface. Plants that are watered by subirrigation may further result in
salting at the surface, as well as encouragement of root growth out the bottom
of holding pots (important if root length or weight is being evaluated).
52
-------
Direct Addition of Solutions
Chemicals that are water soluble can be added directly to soil medium after
dissolving in water. A concern with this form of application is the movement
of the chemical solution through the soil profile (especially where sand is
used). Also, leaching of the chemical during subsequent watering plants from
above is a concern.
Repeated Applications
Although not specifically mentioned in the test guidelines, there may be times
that a pesticide would be re-applied. Consideration should be given to when
it is warranted to re-apply the test chemical in these tests.
SOIL MEDIA CONSIDERATIONS
Germination Test
The guideline allows the use of paper moistened with the chemical solution in
the germination test. The method described by Gorsuch et al.(4) is
appropriate for this test. During the use of this test, it has been observed
that adsorption and chromatographic effects of the test chemical can occur if
the test solution is not thoroughly mixed over the paper as it is being
moistened. It may be desirable to conduct a germination test in the same soil
that is being used for the emergence test. When using the same soil, twice as
many plants can be set up, randomly selecting one-half for the germination
test and designating the second half for the emergence test. An advantage to
using the same soil is that only one chemical analysis to verify test
concentrations needs to be done (see GLP Considerations Section).
Emergence Test
This test allows the use of acid-washed sand or "standard" soil. Sand may not
be representative of soils receiving a pesticide, plus some chemicals have
been shown to bind to it. Leaching and salting may also be a problem in sand.
It is important that the particle sizes be carefully chosen for sand.
A "standard" soil has not been identified. Jiffy Mix* and Promix* are
"standard" soil media, but EPA considers them unacceptable for the germination
and emergence tests due to their high peat content. Jiffy Mix* and Promix*
contain all the necessary nutrients for growth of most plants for four to five
weeks, and based on my experience of testing more than 250 compounds in
seedling studies, these soil media generally do not interfere with the
activity of a chemical. The peat moss present in Jiffy Mix* and Promix* often
absorbs water soluble chemicals, keeping them in contact with the seed or
roots of plants, thus available to potentially influence growth. If natural
soils, e.g., sandy loam with less than 3% organic matter as specified by OECD,
are used, they should have no previous history of pesticide use. Synthesized
53
-------
soils, such as those used for earthworm tests (sand, clay, peat moss), have
been standardized; however, for the most part, they are lacking soil
structure, nutrients, and microbial populations. Microbial populations are
important for the legumes, and provide the opportunity for biodegradation of a
chemical. The article "Homemade potting soil can help plants, pocketbook;
published in the Democrat and Chronicle. Rochester, N.Y., on Saturday,
November 24, 1990, addresses some of the concerns with using natural soils in
pots.(5) For the studies lasting longer than two weeks, specifically
vegetative vigor, nutrient additions are required to maintain healthy plants.
This issue has not been addressed in the EPA guidelines. The standard
Hoagland Solution at one-half strength applied three times weekly is generally
adequate for plants between one week and five weeks of age.
INTERPRETATION OF VISUAL OBSERVATIONS
Test endpoints include the determination of adverse effects by visual
observations using a uniform scoring procedure (e.g., rating systems of 0-100,
or 0-10, or 0-4), which requires defining. Such visual observations are
subjective, and may not be standardized among personnel collecting data, and
certainly not among laboratories. Visual observation ratings cannot be
"averaged" as though equal, when they actually represent a range.
Example: 0 = 0% effect
1 •« 1-25% effect
2 = 26-50% effect
3 = 51-75% effect
4 = 75-100% effect
If observed ratings were 1,1,2,1,2 for a
Vegetative Vigor replicate of five plants,
it would be incorrect to express an
average of 1.4., but that effects ranged
between 1 and 2.
The use of a scoring system with less than ten increments cannot be analyzed
statistically.
The test guideline indicates that direct measurements of height and weight may
also be made. Since direct measurements are generally more reliable than
visual observations and are conducive to statistical evaluations,
consideration should be given to requiring them in order to determine the
effect of a chemical on plants. This would also be consistent with EPA/TSCA,
FDA, and OECD endpoints.
STATISTICAL CONSIDERATIONS
The test guidelines suggest statistical analyses be performed on data.
Specific statistical analyses, and their advantages, were not included in
either the test guideline or the SEPs. Consideration should be given as to
54
-------
which statistical analyses are appropriate in analyzing the data. Tier I
requires that an EC^ be determined (which is quite reasonable), provided
there are effects equal to or greater than 25%. In addition to the EC25, Tier
II also requires that an EC^, value and a No Observed Effect Concentration
(NOEC) be determined. Statistical analyses such as the mean, standard
deviation, and 95% confidence intervals should be included with both Tiers I
and II. The use of statistical analysis to determine the confidence intervals
is based on the assumption that the data are normally distributed.
Statistical tests that are required by the FDA include the Power of the Test,
or Power Calculations, to determine the number of plants and replicates needed
to demonstrate the percent level of statistical differences desired. The
FIFRA guidelines state that the sample size should be adequate to provide
significance at the 90 to 95% confidence interval. No guidance was given as
to how this was to be determined. Linear Trend Analysis allows the comparison
of the means of treated and control groups, rather than the actual data
points, in order to identify the maximum concentration not associated with
significant Linear Trends. Analysis of Variances (ANOVA) are frequently used
to determine the homogeneity of variance and to analyze ranks rather than data
points. The ANOVA and Linear Trend Analysis are generally used to determine
the No Observed Effect Concentration (NOEC). The inclusion of an appendix on
statistical tests that are appropriate for analysis of plant data, including
the handling of outliers, is encouraged. Ms. Merrilee Ritter, a
biostatistician of Eastman Kodak Company, presented two papers at the ASTM
Symposia on "Use of Plants for Toxicity Assessment." The paper titled "An
Overview of Experimental Design" was accepted by ASTM in July 1990 for
publication in the ASTM Special Technical Publication Plants for Toxicity
Assessment (STP 1115). A second, more detailed document, was not published,
however, it does contain materials that would be helpful in preparing an
appendix on biostatistics for plant analysis. It is critical that an
experiment be reviewed by a biostatistician prior to its initiation to ensure
that the experimental design will meet the expectations desired.
GLP CONSIDERATIONS
Good Laboratory Practice (GLP) standards (40 CFR Part 160, effective October
16, 1989) require all FIFRA studies to conform to these regulations. GLP
requires: a protocol describing the study; SOPs for the test, the equipment,
the test article distribution and its use, solution and soil preparations; a
Summary of Training and Experience (STE) for all personnel; training records
certifying personnel capability; the archive of raw data, supporting data, and
a sample of the test article, all which must be maintained for the life of the
registration (which averages 17 years for pesticides); dose solutions must be
analyzed to confirm concentrations, as should soil media to demonstrate
homogeneous mixing; analysis of dilution waters for priority pollutants;
characterization of the soil that is used in the study; and, documentation of
all deviations from SOPs and protocols. The GLP requirements do not
necessarily maintain good quality tests; however, through the audit trail,
they will identify a study that is not of good quality.
55
-------
RECOMMENDATIONS
Although there may be many areas for consideration in improving the EPA/FIFRA
Plant Testing Guidelines, the current guidelines have, and do, provide useful
information. Contrary to popular opinion, however, these studies are not
"cheap." The combined Tier I and Tier II tests can easily cost $75,000 or
more to conduct. This cost is due in large part to the GLP compliance
standards. The following are several recommendations to improve the current
testing:
1. Greater standardization among agencies (EPA/TSCA, EPA/FIFRA, FDA, and
OECD) on key issues is encouraged. This could include such items as the
number of replicates to use, the number of plant species to use, which
plant species to use, the number of plants per replicate, and the
identification of a "standard" soil. To bring about standardization,
the EPA and other regulatory agencies are encouraged to take an active
role in developing standards within the ASTM Subcommittee E47.ll on
Plant Toxicity.
2. Since statistical analyses are required on the data, it would be helpful
to include an appendix that references the appropriate statistical
analysis for plant data. In addition, guidance on experimental design,
using a statistical approach, should be included in the appendix.
3. The guidelines should require direct measurements such as shoot height
and weight rather than making it optional. These quantifiable measures
can be analyzed statistically and are less subjective. A description of
what to measure and how to make such measurements should also be
addressed.
4. A set of environmental conditions such as the optimal temperatures,
photoperiod, light intensity, and relative humidity are recommended. If
vegetative vigor studies are conducted in plots out-of-doors, then the
parameters expected to be followed and maintained should also be
described.
5. The Agency should consider identifying a "standard" soil and the source
of such a soil. Consideration in identifying those times that sand and
other soil media are not appropriate for use in these tests should also
be addressed. The use of nutrients for tests lasting longer than two
weeks should be addressed.
6. Where possible/practical, the choice of test species, environmental
considerations, and soil type, should be applicable to the geographical
area that a herbicide might reach. Herbicides that are used in isolated
regions of the U.S. might provide more useful information by considering
studies designed specifically for those regions.
56
-------
REFERENCES
Gorsuch, J.W., Kringle, R.O., and Robillard K.A. 1990. Chemical Effects on
the Germination and Early Growth of Terrestrial Plants, Plants for Toxicity
Assessment. In: W. Wang, J. W. Gorsuch, and W.R. Lower (eds)., American
Society for Testing and Materials, Philadelphia, PA. ASTM STP 1091. 49-58
pp.
Hellenman, D. 1990. Homemade potting soil can help plants, pocketbook,
Democrat and Chronicle, Rochester, N.Y., Saturday, November 24, 1990.
Hoist, R.W. and Ellwanger, T.C. 1982. Pesticide assessment guidelines,
Subdivision J, hazard evaluation: nontarget plants. EPA 540/9-82-020, U.S.
Environmental Protection Agency, Office of Pesticide and Toxic Substances,
Washington, D.C. 55 pp.
Hoist, R.W. 1986. Hazard evaluation division standard evaluation procedure
nontarget plants: seed germination/seedling emergence - tiers 1 and 2. EPA
540/9-86-132, U.S. Environmental Protection Agency, Office of Pesticide
Programs, Washington, D.C. 13 pp.
Hoist, R.W. 1986. Hazard evaluation division standard evaluation procedure
nontarget plants: vegetative vigor - tiers 1 and 2. EPA 540/9-86-133, U.S.
Environmental Protection Agency, Office of Pesticide Programs, Washington,
D.C. 13 pp.
57
-------
DEVELOPMENT OF NON-TARGET PLANT TEST
METHODS AT ICI AGROCHEMICALS
by
Richard Brown*, Deborah Farmer & Lorraine Canning
ICI Agrochemicals
Content: A glasshouse bioassay designed at ICI Agrochemicals to test for
effects of pesticide spray-drift on terrestrial nontarget plants
was described. The use of glasshouse data alongside environmental
exposure models to predict field effects was discussed.
INTRODUCTION
In recent years, there has been increased concern about the risks posed to
non-target plants by pesticides. Initially these concerns were directed
towards crops growing in the target area, either at the time of spraying or
in following seasons. These concerns continue, but more recently they have
been added to by concerns for crops and other non-target flora outside the
target area that may be exposed to spray-drift of pesticide or volatilisation
from leaves and subsequent deposition in rainfall.
Currently in Europe, there is also considerable interest in the production
of "Conservation Headlands" at the margin of cereal fields. In these areas
(half a spray-boom width from the edge, usually 6m), the use of agrochemicals
and other management techniques are organized such that the growth of certain
plants is encouraged (e.g., Pheasant Eye [Adonis annual, Shepherd's Needle
[Scandix pecten-virensl and Knotgrass [Polygonum avicularel). but that the
more pernicious weeds of low conservation value (e.g., Cleavers FGalium
aparinel. Couch grass [Agropyron repensl and Sterile brome [Bromus sterilisl)
are controlled.
In the USA, Subdivision J (Non-target Plants) requires that tier I tests
are carried out at maximum labelled rate or maximum environmental exposure
concentration (MEEC) and three times this rate on six dicots (including
soybeans and a root crop) from four families and four monocot species
(including corn). Detrimental effects of greater than 25% compared to control
trigger a tier II test. ICI Agrochemicals progresses all herbicides to the
tier II test immediately. The tier II test requires that no observable effect
levels (NOELs), EC25s and EC50s are obtained for a similar range of species
to the tier I test. The top dose is the MEEC and the lower doses arranged to
* Presenter
58
-------
allow for the estimation of the NOEL. The guidelines state that a 25%
detrimental effect compared to control should trigger a tier III or field
test.
Obviously, where the MEEC = the maximum application rate, all herbicides
would trigger a tier III test, for which no protocols or regulatory precedents
exist. The maximum labelled rate must therefore be discounted to allow for
drift or volatilisation deposition outside the target area and this exposure
estimate compared with the effect levels from the tier II test. What is
unresolved at present is (a) the nature of the drift-deposition model and (b)
what constitutes an unacceptable effect on a non-target plant.
In line with EPA requirements, at ICI Agrochemicals we conduct two tier I
and tier II tests on each product; one on seedling emergence when seeds are
treated before germination and one on vegetative vigour when emerged seedlings
are treated. This paper briefly summarises our approach.
MATERIALS AND METHODS
Plant Species
In any one test four monocots and six dicots are tested. The species are
chosen to be as taxanomically diverse as possible whilst giving good
germination, uniform growth in the glasshouse and sensitivity to herbicidal
materials. The current choice of species is given in Table 1.
Test Chemical
The test chemicals are applied as a typical end-use, single active ingredient
(ai) formulated product. The % ai will be checked by chemical analysis before
application and, where possible, the formulated product will be made from
fully characterised technical material. We consider it essential to use
formulated product rather than technical material as it would be necessary to
make a simple formulation to apply as a spray anyway, which is unlikely to be
as phytotoxic as an optimised formulation.
Application is made using an hydraulic track-sprayer fitted with a single,
even stainless steel jet checked for output and eveness of spray-pattern
using a LURMARK "patternator." Jet travelling-speeds are calculated to give
the required output at a set height. Before each spray session, the
track-sprayer is calibrated by spraying and re-weighing petri dishes
containing filter paper of known weight. Output is constrained to within 10%
of nominal.
For tier II tests, the top rate is the highest labelled rate and the lowest
<1% of this.
59
-------
TABLE 1 - Plant species used in non-target terrestrial plant bioassay.
Latin name
DICOTYLEDONS
Glvcine max
Beta vulqaris
Brassica napus
Abutilon theophrasti
Sida spinosa
Xanthium strumarium
Ipomoea hederacea
MONOCOTYLEDONS
Zea mays
Triticum aestivum
Avena fatua
Cvoerus rotundus
Code
GLXMA
BEAVA
BRSNN
ABUTH
SIDSP
XANST
IPOHE
ZEAMX
TRZAW
AVEFA
CYPRO
Common name
Soybean
Sugar beet
Oilseed rape
Velvet leaf
Spiny teaweed
Italian
cocklebur
Purple morning
glory
Maize
Winter wheat
Wild oat
Purple nutsedge
Family
Leguminosae
Chenopodiaceae
Cruci ferae
Malvaceae
Malvaceae
Compositae
Convolvulaceae
Gramineae
Gramineae
Gramineae
Cyperaceae
Climate1
Warm
Cool
Cool
Warm
Warm
Cool
Cool
Warm
Cool
Cool
Warm
'Warm: Temperature day/night = 24/19°C
Humidity day/night = 70/40%
Photoperiod = 14 hours
Cool: Temperature day/night = 18/12°C
Humidity day/night = 70/40%
Photoperiod = 14 hours
60
-------
Seedling Emergence Test
Three replicates of ten seeds of each species are tested at each rate. The
species are divided between those for warm and cool climate regimes
(Table 1) which are planted in separate trays. Seeds are sown into fully
characterised compost of <5% organic matter; crop species are sown at 2cm
and weeds at 1cm depth. The spray is then applied to the soil surface;
controls are unsprayed. The seed trays are then moved to the appropriate warm
or cool regime and laid out in a randomised block design to account for
position in the glasshouse. The position of each treatment within each block
is randomised. Seed trays are placed on individual containers to prevent
cross-contamination following watering. The trays are top-watered twice a
day.
ASSESSMENTS
Percentage emergence is assessed by daily counts, full emergence is taken as
when there is no increase for three days, and the days to emergence as the
first of those three days.
Damage of the emerged plants is assessed at weekly intervals according to
the scale in Table 2. Assessments of the treated plants are made in
comparison with the controls; therefore no scores are made for the control
plants. Additional notes on symptomolgy are made to support these
assessments.
Growth stage is assessed at four weeks as in Table 3.
Dry weight is assessed at four weeks. Plants are cut at soil level and dried
to constant weight and results expressed as weight per plant.
Vegetative Vigour Test
Three replicates of five plants of each species are tested. Plants are grown
in individual pots in the same compost as for the seedling emergence test.
Healthy, vigorously-growing plants only are selected and are sprayed at the
3-4 leaf stage. Apart from this, the plants are handled as for the seedling
emergence test.
ASSESSMENTS
These are as for the seedling emergence test, except for the seedling
emergence assessment.
Statistical Analysis
For the seedling emergence test, data are normalised for parametric analysis
by conversion to angles using an arcsine transformation:
Y = sin'1 7 % seeds emerged/100
61
-------
TABLE 2 - Damage assessment scale for glasshouse bioassays.
% Description
0 Vigorous plant, indistinguishable from control
5 Vigorous plant but slight detectable differences
10 Vigorous plant but readily distinguishable differences
I r ii M ii ii ii ii
20 Less vigorous with pronounced differences
40 Poor vigor with increasing severity of effects
rn ii ii ii ii M ii ii
cr\ ii ii ii ii M M ii
70 Very poor vigor but still growing, recovery possible
7 C ii M M n II II II M
80 Very poor vigor, still growing but recovery unlikely
85 Very poor vigor, ceased growing, recovery very unlikely
90 Not all tissue dead but further growth unlikely
95 Moribund
100 Dead
62
-------
Table 3 - Growth stage key.
MONOCOTYLEDONS
Definition Code
DICOTYLEDONS
Definition Code
Seedling emergence
Coleoptile emerged 1.0
Leaf just at 1.1
coleoptile tip
Leaf Production on
Main Shoot
1st leaf through 2.0
coleoptile
1st leaf unfolded 2.1
2 leaves unfolded 2.2
3 leaves unfolded 2.3
etc.
Tillering
Main shoot only 3.0
Main shoot & 1 tiller 3.1
Main shoot & 2 tillers 3.2
etc.
Stem elongation 4.0
Booting 5.0
Inflorescence Emerged 6.0
Cotyledon Production
Seedling emergence 1.0
Cotyledons expanding 1.05
Cotyledons expanded 1.1
Leaf/whorl Production on Main
Stem
1st leaf expanding 2.05
1st leaf expanded 2.1
2nd leaf expanding 2.15
2nd leaf expanded 2.2
etc.
Branch/shoot Production
on Main Stem
Main stem only 3.0
One branch/shoot 3.1
(>0.5cm)
Two branches/shoots 3.2
etc.
Flower buds present 4.0
Flowering 5.0
Leaf senescence 6.0
63
-------
Final dry-weight data is normalised for parametric analysis with a square-root
transformation:
Y = y dry weight (g)
These data are then analysed with a 2-way ANOVA (treatment and block) and
the estimate of within-plot error used to calculate least significant
difference (LSD) values. Treatment-rate means for each species are then
compared to untreated controls at the 5% significance level. The NOELs are
taken as the highest treatment rate below which there are no significant
differences from the untreated controls.
For the damage dose-response data, % damage is transformed to logits:
logit (% damage) = % damage + 0.5
100.5 - % damage
and plotted against loge treatment rate. NOELs are equated to EC10 on this
scale. This value has been found to be typical of control variation in these
studies.
RESULTS AND DISCUSSION
Figure 1 shows the contrasting effects of acetochlor on Sida spinosa and corn
when applied pre-emergence. Figure 2 shows the differing effects of the TVC
herbicide glyphosate-trimesium and the grass-killer tralkoxydim when applied
post-emergence to sugar beet. Figure 3 shows the effects of glyphosate-
trimesium on soybean when applied pre and post-emergence.
To be interpreted, the amount of herbicidal material deposited outside the
target area has to be assessed and thi's can then be related to the toxicity
levels noted in the glasshouse study. In general, herbicides are more toxic
in the glasshouse than the field due, amongst a variety of factors, to thinner
plant cuticles caused by low light levels and high humidity. This "transfer"
factor is generally about x2, but can be unreliable.
Currently, we model the decreasing percentage of herbicide deposition with
downwind distance using a quadratic model. Matching this data with toxicity
values allows us to predict plant damage downwind of herbicide spraying at
the maximum labelled rate. Figure 4 shows the results of this for a
post-emergence application of glyphosate-trimesium and Italian cocklebur
(Compositae) and Figure 5 for pre-emergence exposure of sugar beet
(Chenopodiacae) to acetochlor. These results show effects becoming negligible
within a few metres of the target area and correspond well with work conducted
to assess "safety zones" for nature reserves in the UK and with protection of
sensitive crops in the USA.
64
-------
fD
OJ
c.
O)
QJ
(O
CO
QJ
4-*
(O
O
O.
OJ
n
O)
CT
(0
(O
C
(D
O)
31
100 -
90 -
60 -
70 -
60 -
50 -
40 -
30 -
20 -
10 -
0 -
D Si da spinosa
m Zea mays
Si da splnosa ECM- 0.26 kg ha*1
SJfta splnosa EC_- 0.17 kg ha"1
SJtfa splnosa ECJO- 0.1 kg ha"1
TT
0.007 0.02 0.05 0.1 0.4
Application rate, kg ha"1 (loge scale)
Max. field rate
2.52 kg ha'1
Figure 1 -Comparing the effects (percentage damage) of a non-selective herbicide,
acetochlor, on two plant species included in the seedling emergence test.
65
-------
QJ
4-»
(O
OJ
c_
(0
4-*
c
ro
t— <
O.
in
x
to
0)
4->
0)
o
ex
0)
Ł_
ro
0)
CT
(TJ
E
c
(D
0)
100 -
90 -i
BO -
70 -
BO -
50 -
40 -
30 -
20 -
10 -
0 -
I
Glyphosate-
trimesium
Tralkoxydim
i
Max. field rate
4.48 kg ha-»
Max. field rate
0.35 kg ha-i
l
0.00004 0.0003 0.002 0.02
-1
I
0.1
I
7
Application rate, kg ha (loge scale)
Figure 2 - Comparing effects (percentage damage) on sugar beet of a non-selective
herbicide, sulfosate, and a graminicide, tralkoxydim, in the vegetative vigour
test.
66
-------
O)
CT
CO
e
(O
•o
0)
CT
ID
-M
C
O)
u
C.
O>
O.
C
(O
O)
to
4J
C
ID
r—»
Q.
in
c
o
CO
T3
0)
O)
in
10
O)
4->
(O
u
Q.
0)
100 -
90 -
60 -
70 ->
60 -
50 -
40 -
30 -
20 -
10 -
Ł2 OH
o Post emergence
• Pre emergence
i
0.0009
I
0.007
0.05
I
0.4
Application rate, kg ha"1 (loge scale)
Max. field rate
4.4B kg ha'*
Figure 3 - Comparing pre- and post-emergence effects (percentage damage) on
soybean, of a non-selective, post-emergence herbicide, glyphosate-trimesium.
67
-------
XJ
O)
O>
C_
•o
Q)
c_
ro
Q.
E
O
(J
C
(U
-------
•o
0)
-t-J
07
0)
c
ID
T3
0)
t_
ro
a.
E
O
u
4J
C
(U
u
c_
0)
a.
Dry weight
Damage
2468
Distance downwind from swath edge (m)
10
Figure 5 - Predicted pre-emergence hazard of the broad-spectrum herbicide, acetochlor,
to a broad-leaved non-target plant (Chenopodiaceae).
69
-------
TISSUE CULTURE TESTS FOR STUDYING PHYTOTOXICITY AND
METABOLIC FATE OF PESTICIDES AND XENOBIOTICS
IN PLANTS
by
Hans Harms* and Elke Kottutz
Federal Agricultural Research Center, Germany
Content: The use of plant cell cultures for ecotoxicological evaluation of
pesticides and xenobiotics was compared with intact plants grown
under aseptic conditions.
INTRODUCTION
One of the potential applications of plant cell culture technology is as a
test system for investigating the phytotoxicity and metabolic fate of chemi-
cals. Callus and cell suspension cultures of various plant species have been
most commonly used for an ecotoxicological evaluation of the fate of pesti-
cides and xenobiotics (Zilkah and Gressel, 1977; Mumma and Davidonis, 1983;
Sandermann et al., 1977; Schuphan et al., 1984; Harms and Langebartels, 1986).
In the investigations reported here, we have compared xenobiotic metabolism in
cell cultures with that of whole plants.
MATERIALS AND METHODS
Cell Suspension Cultures
Various cell suspension cultures of different monocots such as Triticum
aestivum L.. Hordeum vulqare L., Pennisetum americanum L.. and dicots such as
Glvcine max L.. Daucus carota L.. Lycopersicum esculentum L. were cultured as
described previously (Harms, 1973; Langebartels and Harms, 1984) and were used
for metabolism tests during the last 48 hours of the late logarithmic growth
phase.
* Presenter
70
-------
Aseptically Grown Plants
Wheat cv. Heines Koga. tomato cv. Money maker and Atriplex hortensis plants
were hydroponically grown under aseptic conditions (Langebartels and Harms,
1986). After two to three weeks of growth, the test compounds were added to
the nutrient solution. After five days treatment, the plants were harvested
and analyzed.
Phytotoxicity and Metabolism Test
The phytotoxicity, uptake, and metabolic fate of xenobiotics were studied by
standardized methods (Harms and Langebartelsi 1986). The bound or non-
extractable residues were analysed by a sequential fractionation procedure
(Langebartels and Harms, 1985).
RESULTS AND DISCUSSIONS
Determination of Phytotoxic Effects of Xenobiotics
In order to compare the phytotoxic effect of xenobiotics on intact plants and
cell cultures, pentachlorophenol was added in 10"* to 10"8 molar concentrations
to the nutrient solutions of both systems (Figure 1). Plant growth in both
systems was remarkably reduced by pentachlorophenol concentrations higher than
10"6 molar. In intact plants, the shoots are much more sensitive than the
roots, but in both organs growth was reduced to nearly 25% at 5 • 10"6 molar
0.2
'
*»
far
I
\ Intict plant
.
• •— .«
\
-
I . i
0 »-«
\ihoot
O
\root
I i
10-5 10-4
I
1
o?
cell culture
KT6 10~s
tr>
2*
mot -I-1
Figure 1 - Phytotoxicity of pentachlorophenol in wheat.
71
-------
concentrations. The wheat cell cultures responded very similarly. It is
obvious that the conductivity of the medium followed the same but reversed
pattern. Thus cell cultures seem to be a good system for testing
phytotoxicity of xenobiotics.
In order to investigate phytotoxic effects of xenobiotics on different plant
species, cell cultures of barley, carrot and tomato were incubated with
different concentrations of phenanthrene and 4-nonylphenol (Fiqure 2).
dry weight (g)
0.7-
0,6
0.01 mM 0.06 mM 0.1 fflM 0.6 mM 10 mM control
concentration
I carrot HUD barley l§§ tomato
dry weight (g)
0.01 mM 0.06 mM 0.1 mM 0.6 mM 1 mM
concentration
ESS carrot GHH3 barley S3§ tomato
control
Figure 2 - Phytotoxic effect of (A) phenanthrene and (B) 4-nonylphenol.
72
-------
Carrots are known to enhance the uptake of nonpolar compounds due to their
lipid content, whereas tomatoes are very sensitive to organic chemicals, thus
providing a good comparison for these tests.
Carrot growth was hardly influenced at any of the tested concentrations of
phenantherene, tomato showed a drastic decrease in growth at concentrations
higher than 0.01 mM. Barley growth was decreased by about 35% at concentra-
tions higher than 0.05 mM.
As with phenanthrene, carrot growth did not seem to be influenced by 4-
nonylphenol at any concentration, whereas in tomato, concentrations higher
than 0.5 mM inhibited growth completely. In barley, at concentrations from
0.01 mM to 0.1 mM, growth was barely affected whereas at concentrations higher
than 0.1 mM, cell mass was reduced by 66%.
The phytotoxicity studies indicate that plant cell cultures are a good system
for testing toxic effects of chemicals. The cell cultures responded with
different sensitivity towards the chemicals, which reveals that plant cell
cultures maintain their species specific peculiarities.
Comparative Studies of Xenobiotic Metabolism by Cell Cultures and
Intact Plants
METABOLISM OF PENTACHLOROPHENOL AND 4-CHLOROANILINE by WHEAT CELL
SUSPENSION CULTURES AND WHOLE PLANTS
The validity of extrapolating data obtained with cell culture techniques to
those of intact plants is still a matter of debate. In order to compare these
two systems, cell suspension cultures and wheat seedlings of the same cultivar
were incubated with pentachlorophenol and 4-chloroaniline. The metabolic
rates of these compounds in the two differently differentiated plant systems
are shown in the Figure 3.
The compounds were taken up and metabolized by both plant systems. The 14C-
label of both compounds was transported from the roots into the shoots of the
intact wheat plants. Cell cultures adsorbed pentachlorophenol very rapidly
and formed high amounts of polar metabolites which were mainly associated with
the cells. Forty-one percent of the radiolabel was converted (via the
conjugate fraction) into the non-extractable residue fraction. The 14C-label
was bound mainly to lignin and to a high molecular weight hemicellulose
fraction. Polar conjugates could also be extracted from roots and shoots, and
PCP glycosides were also predominant as found in the cultured cells. More
than 16% of the total radioactivity from shoots and roots were found as bound
residues. These were fractionated into several wall components and yielded a
pattern similar to that in cell cultures.
73
-------
of applied radioactivity
100-
80-
60-
40-
20-
pentachlorophenol
4-chloroanlilne
T"' "I"' rrV™ r"i'" 1 1 '"f" "I" I f
eell* medium root* sheet* nalr. eeL eel It medlym roow •beet* Mir. eel.
evil culture Intact plants cell culture Intact plant!
^^ parent compound l;:::| metabolites I\\\\1 bound residues
Figure 3 - Metabolism of pentachlorophenol and 4-chloroaniline by wheat eel
suspension cultures and intact plants.
In cell cultures, 72% of 4-chloroaniline was detected in the bound residue
fraction. Further studies showed that this high proportion of 1
-------
METABOLISM OF 4-NONYLPHENOL AND PHENANTHRENE BY TOMATO CELL SUSPENSION
CULTURES AND INTACT PLANTS
Both 4-nonylphenol, an isomer of alkylphenol polyethoxylate surfactants, and
phenanthrene build up In wastes and sludges and subsequently enter other
environmental spheres. The metabolism of these compounds has been studied
with respect to their phytotoxic effects.
Although both chemicals are non-polar compounds they were almost entirely
assimilated by the cell cultures (Figure 4). Whereas 4-nonylphenol was
completely converted to polar metabolites (glucoside conjugates), phenanthrene
was predominantly detectable as parent compound and only 7% as polar
metabolites. The intact tomato plants took up only small amounts of both
compounds. Most of the radioactivity was located in the nutrient solutions,
but obviously not only in the form of the parent compound but also as
metabolites. 4-nonylphenol was not translocated within the plants (no
radioactivity in the shoots), whereas up to 7% of the applied phenanthrene was
identified in the shoots.
% of applied radioactivity
100
60-
60-
40-
20-
0
4-nonylph«nol
phenanthrana
-JJM4&
ssss.
BBSS
56SS
mttfkM i»ou •to*t« Mtr. Ml.
oultur* IntMt plant*
! parent compound
«•!!• mtdlim ro«t« •to*t« Mtr. Ml.
oail culture Intaot plants
HOI metabolltea ES3 bound reelduae
Figure 4 - Metabolism of 4-nonylphenol and phenanthrene by tomato cell suspension
cultures and intact plants.
75
-------
METABOLISM OF 2,2',5,5'-TETRACHLOROBIPHENYL AND PHENANTHRENE BY WHEAT CELL
SUSPENSION CULTURES AND WHOLE PLANTS
The metabolism of a PCB-cogener, 2,2',5,5'-tetrachlorobiphenyl, was compared
with that of phenanthrene (Figure 5).
ft of applied radioactivity
100-
80-
60-
40-
20-
2,2*,e.e'-tetracnloro-
blphenyl
phenanthrene
Mil* m«lym root* »hoot» mm. »oL Mil* «i«dlum r«et» •*»•<• Mir. ••!.
o«ll oultura Intact plants oHI oultura Intact plant*
g^5 parent compound !"••! metabolite* E3 bound realdue*
Figure 5 - Metabolism of tetrachlorobiphenyl and phenanthrene by cell suspension
cultures and intact plants in wheat.
The fate of phenanthrene, when provided to wheat, was similar to what was
observed for tomato cell cultures and plants (Figure 4). In contrast to this
and all other compounds which we have studied up to now, 2,2',5,5'-
tetrachlorobiphenyl was not metabolized at all, making it the single
exception. Although wheat cell cultures adsorbed all of the applied PCB-
cogener, and the wheat plants took up more than 25% of this compound, all of
the 2,2'5,5'-tetrachlorobiphenyl remained unchanged. The distribution of the
radioactivity was about 18 times higher in the roots than in the shoots,
indicating little translocation within the plants. From the studies of
Fletcher et al. (1987), we know that another tetrachlorinated cogener
76
-------
2,2',4,4'-PCB was metabolized to polar and insoluble residue products by rose
cultures. In a screening assay to characterize microorganisms for their
ability to degrade PCPs, Bedard et al. (1986) proved that 2,3 and 2,4 cogeners
with open 2,3 and 2,4 sites are degraded much more easily by numerous
microorganisms than the 2,2',5,5' and 2,2',4,4',5,5'-cogeners. The metabolism
of the PCBs seems to be dependent on differences in cogener specificity.
METABOLISM OF BENZO(A)PYRENE AND DIBENZ(A,H)ANTHRACENE BY CELL SUSPENSION
CULTURES AND INTACT PLANTS OF ATRIPLEX HORTENSIS:
Comparative studies with two different plant systems, namely cell suspension
cultures and aseptically grown seedlings of Atriplex hortensis, with the 5-
ring-systems benzo(a)pyrene and dibenz(a,h)anthracene showed that uptake and
metabolism of polycyclic aromatic hydrocarbons depended not only on the size
of the molecule, but also on the structure of the compound (Figure 6).
% of applied radioactivity
100-f
80-
60-
40-
20-
benzo(a)pyrene
888
888
888
dlbenz(a.h)anthracent
e«ll* mtdium reoti •hoot* mtir. Ml. <••!)• radium root* shoot* nutr. »oL
o*ll oultur* Intaot plants ocll culture Intsot plant*
B8SS2 parent compound l;::;l metabolltea ^VJ bound realduea
Figure 6 - Metabolism of benzo(a)pyrene and dibenz(a,h)anthracene by cell cultures
and aseptically grown plants of Atriplex hortensis.
77
-------
Benzo(a)pyrene was assimilated to the largest extent (63% of applied amount)
by the cell suspension cultures. About 54% could be detected as the parent
compound, 2% as metabolites and 7% as bound residue, whereas in intact plants
only small amounts of radioactivity were found in the roots. There was hardly
any transport into the shoots. Most of the applied chemical remained unal-
tered in the nutrient solution.
Dibenz(a,h)anthracene was only taken up to a small extent by the cell cultures
as well as by the intact plants. The magnitude of the parent compound
remained in the nutrient solution.
CONCLUSIONS
Both plant cell cultures and intact plants were shown to metabolize the
compounds examined by common metabolic pathways. Qualitatively, the
metabolites were the same in both systems. However, using cell suspension
cultures, the results may be obtained much quicker with less analytical
expense. They are therefore a useful system to obtain rapid evidence of the
ecotoxicological behaviour of chemicals in plants.
REFERENCES
Bedard, D.L., Uterman R., Bopp, L.H., Brennan, M.J., Haberl, M.L. and Johnson,
C. 1986. Rapid assay for screening and characterizing microorganisms for the
ability to degrade polychlorinated biphenyls. Appl. Environ. Microbiol. 51,
761-768.
Fletcher, J., Groeger, A., McCrady, J. and McFarlane, J. 1987.
Polychlorobiphenyl (PCB) metabolism by plant cells. Biotechnol. Lett. 9, 817-
820.
Harms, H. (1973): Pflanzliche Zellsuspensionskulturen- Ihr Leistungsvermb'gen
fiir Stoffwechselunterschungen. Landbauforsch. Volkenrode 25, 83-90.
Harms, H. and Langebartels, C. (1986): Standardized plant cell suspension test
systems for an ecotoxicologic evaluation of the metabolic fate of xenobiotics.
Plant Science 45, 157-165.
Langebartels, C. and Harms, H. (1984): Metabolism of pentachlorophenol in
cell suspension cultures of soybean and wheat: pentachlorophenol glucoside
formation. Z. fur Pflanzenphysiol. 113, 201-211
Langebartels, C. and Harms, C. (1985): Analysis for nonextractable (bound)
residues of pentachlorophenol in plant cells using a cell wall fractionation
procedure. Ecotox. Environ. Safety 10, 269-279.
78
-------
Langebartels, C. and Harms, H. (1986): Plant cell suspension cultures as test
systems for an ecotoxicological evaluation of chemicals. Growth inhibition
effects and comparison with the metabolic fate in intact plants. Angew.
Botanik 60, 113-123.
Mumma, R. 0. and Davidonsis, G. H. (1983): Plant tissue culture and pesticide
metabolism. In: D. H. Hutson and T. R. Roberts (eds.), Progress in Pesticide
Biochemistry. Volume 3, pp. 225-278, Wiley, Chichester.
Sandermann, H., Diesberger, H. and Scheel, D. (1977): Metabolism of xeno-
biotics by plant cell cultures. In: W. Barz, E. Reinhard and M. H. Zenk
(eds.), Plant tissue culture and its bio-technological application, pp. 178-
196, Springer-Verlag, Berlin.
Schuphan, I., Hague, A. and Ebing, W. (1984): Ecochemical assessment of
environmental chemicals. Part 1: Standard screening procedure to evaluate
chemicals in plant cell cultures. Chemosphere 13, 301-313.
Zilkah, S. and Gressel, J. (1977): Cell cultures vs. whole plants for
measuring phytotoxicity. I. The establishment and growth of callus and
suspension cultures; definition of factors affecting toxicity on calli.
Cell Physiol. 18, 641-655.
Plant
79
-------
PLANT REPRODUCTION AND/OR LIFE CYCLE TESTING
by
Hilman Ratsch* and John S. Fletcher
U.S. EPA and University of Oklahoma
Content: The rationale for using a reproduction/life cycle test in
pesticide registration procedures was presented. Evaluation of
the test systems used to date was critically evaluated relative to
exposure scenarios, test species, and objective of adding a
reproduction/life cycle test to nontarget plant testing
(Subdivision J).
INTRODUCTION
Currently, more than 50,000 pesticides are registered for use and an estimated
500 million kgs of pesticides are applied annually in the United States.
Herbicides, which make up 60% of the total, are applied on 100 million ha of
the nations agricultural and forest lands (Pimentel and Levitan, 1986).
Depending on environmental conditions, the spray apparatus, and the form of
the pesticide applied, up to 25 and 50 percent of the pesticides applied by
ground and aerial techniques, respectively, leaves the targeted land and
impacts nontarget areas. Protection of nontarget vegetation from herbicide
damage is dependent in part on accurate evaluation of each new herbicide
during the registration process. At the heart of this evaluation and
registration process are the tier tests conducted in accordance with the
Subdivision J Guidelines. At present, this three-tiered system does not
include a reproduction/life cycle test, the ramifications of which are
discussed in this paper.
IMPORTANCE OF UNDERSTANDING PESTICIDE EFFECTS
AT DIFFERENT GROWTH STAGES
Most herbicide investigations study the effect of a single application at one
particular stage in the life cycle of a plant, usually during early vegetative
growth. As a result of this narrow focus of herbicide research, there are
relatively few reported investigations in which studies have established how
* Presenter
80
-------
plants respond to a herbicide at different stages of development, and how a
sublethal exposure during one stage influences a later stage. These are
important issues since there are studies such as that of Weaver et al. (Table
1) which clearly showed that, although the growth of young soybean plants was
influenced by 2,4-D application, their yield was not reduced as severely as
when the chemical was applied to plants during early pod fill.
Table 1 - Soybean dry weight at harvest after application of 2,4-D at a rate of 0.5 Ib/A
Stage when treated Mean seed wt./plot
3 inches
6-8 inches
early flowering
early pod
Control
135.5 g
145.5 g
72.0 g
22.8 g
161.7 g
Data from: Weaver, et al., 1946. Bot. Gaz. 107, 563-568.
The importance of knowing how herbicides influence plants at different stages
of their development is further dramatized by Atkinson's contention (1985)
that the two most important factors in determining damage to plants from drift
are rate or concentration of chemical, and timing of applications. Timing is
critical because when a pesticide is applied to the target area, nontarget
plants are in various growth stages, some of which may be very sensitive to
the herbicide. There are numerous examples (Table 2) to illustrate that in
agroecosystems, adjacent crops are often impacted by herbicide drift at
various stages of crop development.
Table 2 - Reports of pesticide damage to nontarget crops at various stages of
development.
Chemical Target Nontarget Stages Reference
MH tobacco soybean reproductive stages Helsel. et al., 1987
2,4-D wheat fieldbeans flower and pod Lyon and Wilson, 1986
sethoxydim soybean corn later leaf stages Chernicky and Slife, 1986
propanil rice cotton early stages Hurst, 1986
dicamba corn . soybean pre and postbloom Weidhamer et al., 1989
glyphosate weeds tomatoes flowering stage Romanowski, 1980
2,4-D wheat sugar beet early growth Schroeder, et al. 1983
81
-------
REVIEW OF TEST GUIDELINES
Under the authority of the Federal Insecticide, Fungicide and Rodenticide Act
(FIFRA), EPA can require toxicity test data to support the registration of a
pesticide. Subdivision J prescribes the data required to protect nontarget
plants and describes the test guidelines used to gather the data. A tier
system (Table 3) made up of a screening test, definitive test and a field test
is described.
Table 3 - Key features of Subdivision J tier tests.
Tier I -- Screening test - 1 concentration of chemical
(max. label rate a.i.)
a. Seed germination/emergence
1. Ten seeds per dish incubated for at least 5 days
2. Filter paper, sand, or standard soil
3. Report number of seeds germinated
b. Emergence
1. Ten seeds in sand or standard soil
, 2. At 10 -14 days report number emerged
c. Vegetative vigor - foliar spray
1. Laboratory, greenhouse, field plants
2. Plants 1 -4 week post emergent
3. Record phytotoxicity, morphology, height, weight
Tier II -- Definitive test - 5 concentrations
a. Required if there is an adverse effect of greater than 25%
to seeds/plants in one or more species in tier I
b. Same test procedures are used as in tier I except, 5 dosages
are required and a 25 to 50% detrimental effect level and
confidence limits must be determined
Tier III -- Field test - 1 concentration - multiple applications
a. Required when recommended field application rate is greater
than EC25 for one or more species in tier II
b. Seed germination, vegetative vigor, and reproductive
potential under field conditions
c. Species are expanded to include Vascular Cryptogamae,
Bryophyta or Hepatophyta, and Gymnospermae
d. Field-use conditions similar to natural habitat of test-
plants with a test duration of at least two weeks and up to
a maximum of 4 weeks following last application
82
-------
Comparison of the tier test guidelines with the stages of a plant's life cycle
(Figure 1) clearly shows that the guidelines are focused on only the early
stages. In both tiers I and II, both the exposure and analysis of endpoints
are restricted to vegetative growth with absolutely no attention given to
reproduction. Therefore, if a chemical is inhibitory to any reproductive
event such as flowering, pod fill, or seed maturation; neither tier I or II
testing will detect this toxic influence. The omission of reproduction
testing from tiers I and II is surprising in the sense that normal
reproduction culminating in maximum seed set is precisely what must be
maintained in a healthy agroecosystem to ensure that maximum crop yields are
obtained and that native annuals growing in buffer zones survive from viable
seed banks. Tier III, as described in Subdivision J, has the potential for
filling the void for reproduction testing, but unfortunately test protocols
have never been written to clarify how reproduction potential should be
assessed.
It is important to know how the application of pesticides during different
phases of growth, i.e., vegetative, flowering and pod filling, affect plant
survival and yield (biomass and/or seed-mass). If we are concerned about long
term effects like yield, total biomass, seed production, then the results from
tier I and II may not be good estimates. It is evident that seed germination
and vegetative vigor as evaluation procedures at tiers I and II are not
entirely adequate methods for assessing pesticide effects. A slight or
negative response early in the cycle does not mean that later vegetative or
reproductive stages of the plant will respond in the same manner.
LIFE CYCLE TESTS DEVELOPED BY EPA
Research has been conducted by EPA-Corvallis Laboratory to develop life cycle
tests for toxicity assessment. The plants used in the tests are commonly
referred to as the Arabidopsis and Brassica life cycle tests after Arabidopsis
thaliana (Ratsch, et al., 1986) and Brassica rapa (Shimabuku, et al., 1990).
The Arabidopsis test was developed to examine the influence of continuous
chronic exposure of plants to contaminated water and the Brassica test was
designed for assessment of soil contamination at hazardous waste sites. In
keeping with the intended use of these tests, both are root exposure tests and
have never been used in foliar application testing, the methodology reflecting
crop exposure during pesticide drift. Adopting either the Arabidopsis or
Brassica test into the tier testing scheme should only be done after careful
consideration of the positive and negative features associated with these
tests.
83
-------
Vegetative Growth
LIFE CYCLE
SOYBEAN
2 wk
Germination
t
f
Seed
Reproduction
Flowering
Reproduction
Pod and Seed
Development
Figure 1 —Life cycle of soybean
84
-------
CRITERIA FOR SELECTING LIFE CYCLE TEST SPECIES
After working with Arabidopsis and Brassica as test species and examining the
advantages and disadvantages of various characteristics of each plant, we have
developed a set of criteria of desirable characteristics for selecting a test
species. The criteria are outlined below.
a. Short life cycle (seed to seed)
b. Seed and germination requirements
1. Available seed source
2. Large seed (convenient for planting)
3. High seed viability
4. Non-dormant
c. Structural features favoring chemical testing
1. Strong stem (no staking needed)
2. Single stem axis
3. Even distribution and position of leaves along stem
4. Fairly large leaves and leaf area
5. Small plant size to permit economical growth and treatment
of replicates under greenhouse conditions
d. Growth
1. Minimal fluctuation in growth (biomass and form) under
varying greenhouse conditions (winter versus summer)
e. Flowering and seed set features
1. Consistent and uniform flower and seed set on control
plants
2. No photoperiod requirements
3. Self-pollinating
f. Endpoint measurement
1. Easy to harvest and compare seed and vegetative biomass
yields
2. Easy to make morphological comparisons between treated
and control plants
g. Taxonomic importance
1. A plant which is a member of a taxa which is economically
and ecologically important
85
-------
There is a definite advantage in selecting a species that reaches maturity
rapidly so that experiments can be completed in the shortest time possible,
and when plant growing space is limited it is cost effective to select plants
with a fast turnover.
In plant testing, a seed source is necessary that can provide a readily
available seed on a continuous basis for a long period. Seed size should be
large enough for ease of planting and ease of harvest. Handling extremely
small seed is too labor intensive. Seed should have a high germination rate
and not require any special conditions to break dormancy.
When large numbers of plants are required, then a sturdy plant with a strong
stem is desirable so that staking is not necessary. Plants with several
axillary stems can be a problem because it may be difficult to quantify
morphological characteristics. Morphological effects are easier to quantitate
if the leaves are distributed evenly along the stem, and the leaves are large
enough to easily measure length and area. Small plants have a definite
economic advantage because more individuals can be grown in less space,
thereby improving the accuracy of statistical analysis.
To obtain reproducible test results over time, plants are required that have
the minimum fluctuation in growth from season to season.
Test plants must also have uniform flower and seed production from experiment
to experiment. Any unusual conditions that are required to induce flowering
and seed set would make that species impractical for testing. The test plant
must also be self-fertile. It would not be cost effective if large numbers of
plants had to be hand pollinated.
A desirable test plant should have large seed production, indehiscent siliques
or pods and be easy to harvest.
There is considerable controversy over whether it is more desirable to select
either an economically important species or an ecologically important species.
TEST SPECIES EVALUATION
Using the test species criteria listed above, we have evaluated the
Arabidopsis and Brassica and two additional species that have been suggested
as good test candidates for life cycle testing. The evaluation is summarized
in the following table.
86
-------
TABLE 3 - Evaluation of test species
Criteria
Test Species
A. Short cycle
days
B. Seed requirements
size
germination
dormancy
C. Structure
D. Growth
plant height
biomass
seed production
E. Flowering
self fertile
F. Endpoint measure
Previously Used
Arabidopsis Brassica
45
yes
veg
seed
36
no
veg
seed
Proposed
Buckwheat Bean
50
G. Taxonomic importance moderate high
no
veg
seed
low
55
0.12 mm
90% germ
non d
rossette
raceme
silique
30-40 cm
1 9/P
1.3 mm
90% germ
non d
stem
raceme
silique
70-80 cm
3 g/P
3 mm
90% germ
non d
stem
achene
60 cm
20 g/p
3 mm
85% germ
non d
stem
pod
70 cm
10 g/p
yes
veg
seed
high
CONCLUSIONS
1. A cost effective and accurate life cycle test should be designed and
tested for use in the tier testing scheme of Subdivision J.
2. It may be uneconomical and scientifically unsound to adopt either the
Arabidopsis or the Brassica life cycle test for Subdivision J purposes.
3. Screening studies need to be conducted to identify plant species which
are best suited for life cycle testing of foliar applied chemicals under
greenhouse and field conditions.
87
-------
4. Research needs to be conducted to determine the best methods and times
for the application of chemical and measurement of response during life
cycle testings.
REFERENCES
Atkinson, D. 1985. Glyphosate damage symptoms and the effects of drift. In:
The Herbicide Glyphosate, eds. E. Grossbard, D. Atkinson. London:
Butterworths. p.455-465.
Breeze, V.G. and Timms, L.D. 1986. Some effects of low doses of the
phenoxyalkanoic herbicide mecoprop on the growth of oilseed
rape (Brassica napus L.) and its relation to spray drift damage. Weed Res.
26, 433-439.
Chernicky, J.P. and Slife, F.W. 1986. Effects of sublethal concentrations of
bentazon, fluazifop, haloxyfop, and sethoxydim on corn (Zea mays). Weed Sci.
34, 171-174.
Helsel, Z.R., Ratcliff, E. and Rudolph, W. 1987. Maleic hydrazide effects on
soybean reproductive development and yield. Agron J. 79, 910-911
12.
Hurst, H.R. 1986. Response of cotton to selected herbicides applied to
simulate drift. Bull. Miss. Agric. For. Exp. Stn., Mississippi State, Miss.
Bulletin 946. 7 p.
Lyon, D.J. and Wilson, R.G. 1986. Sensitivity of fieldbeans (Phaseolus
vulqaris) to reduced rates of 2,4-D and dicamba. Weed Sci. 34, 953-956.
Pimentel, D. and Levitan, L. 1986. Pesticides: Amounts applied and amounts
reaching pests. Bioscience 36, 86-91.
Ratsch, H.C., Johndro, D.J. and McFarlane, J.C. 1986. Growth inhibition and
morphological effects of several chemicals in Arabidopsis thaliana (L.) Heynh.
Environ. Toxicol. Chem. 5, 55-60.
Riley, C.M. and Wiesner, C.J. 1989. Off-target deposition and drift of
aerially applied agricultural sprays. Pestic. Sci. 26, 159-166.
Schroeder, G.L., Cole, D.F. and Dexter, A.S. 1983. Sugarbeet (Beta vulqaris
L.) response to simulated herbicide spray drift. Weed Sci. 31, 831-836.
Shimabuku, R.A., Ratsch, H.C., Wise, C.M., Nwosu, J.U. and Kapustka, L.A. A
new plant life-cycle bioassay for assessment of the effects of toxic chemicals
using rapid cycling Brassica. Presented at ASTM 2nd Symposium on Use of
Plants for Toxicity Assessment. April 23-24, 1990. San Francisco, CA. [ASTM
STP 1115, J.W. Gorsuch, W.R. Lower, M.A. Lewis and W. Wang, In Press]
88
-------
Weaver, R.J., Swanson, C.P., Ennis, W.B., and Boyd, F.T. 1946. Effect of
plant growth-regulators in relation to stages of development of certain
dicotyledonous plants. Bot. Gaz. 107, 563-568.
Weidhamer, J.D., Triplett, G.B. and Sobotka, F.E. 1989. Dicamba injury to
soybean. Agron. J. 81, 637-643.
89
-------
SESSION IV: FIELD TESTING (TIER III, SUBDIVISION J)
The mechanics of how tier III (field testing) should be conducted, what
endpoints should be measured, and how the results should be interpreted are
all issues of debate. To provide a basis for discussing these issues, the
first three papers were presented as examples of how field studies have been
conducted to evaluate the influence of herbicides on nontarget, agricultural
plants. The fourth paper described a novel test system to evaluate the impact
of chemicals on a mixed population of species.
90
-------
INVESTIGATING HERBICIDE SENSITIVITY
THRESHOLDS
by
R.H. Callihan* and LW. Lass
University of Idaho
Content: Construction and use of a continuous curve logarithmic applicator
is described here. Interpretation of plant response data in five
crop species is offered to show how plant biology differences
affect time of appearance of and recovery from symptom expression
resulting from sublethal near-threshold levels of herbicide
exposure. Alfalfa (Medicago sativa L.), potato (Solanum tuberosum
L.), pea (Pisum sativum L.), lentil (Lens culinaris L.) and sugar
beet (Beta vulqaris L.) crops were used to examine how differences
in plant height, biomass and other criteria of plant response were
mediated during the growing season by the herbicide sulfometuron
(2-[[[[(4-6-dimethyl-2-pyrimidinyl) amino] carbonly]amino]
sulfonyl] benzoic acid) at doses between 2.2 and 0.07 g/ha, and to
suggest how propagule size and growth form affect responses.
INTRODUCTION
Most pesticides are applied at dosages substantially above those necessary to
effectively control the pest, due to natural limitations upon selective
targeting of pests. Most herbicides, for example, are usually distributed
over the entire target area in which weed suppression is desired. The
proportion of herbicide that is actually absorbed by the weeds is normally
small and variable because the weed size and density may be so small that
interception by soil or other species may far exceed uptake by the weeds.
When a herbicide is applied to a field, the operational target is the field.
The managerial target is the weed population, and enough herbicide is applied
to kill or suppress it. Some of the herbicide may impinge on the weed, some
may land on the soil surface, some may move from the target area to disperse
in the atmosphere, and some may arrive at a non-target area. The amount of
herbicide that moves to a non-target site is seldom sufficient to kill or
suppress weeds at that site, but is often sufficient to produce sublethal
effects or even lethal effects on non-target plants. When foliar-applied
* Presenter
91
-------
herbicides move off target during application they normally do so in much
smaller droplets (below 50 rim diameter) than those that are quickly deposited.
Smaller droplet size is a more effective physical form than the larger droplet
sizes that impinge on the target area. That is to say, in off-target drift,
the solution is in a fine aerosol form that tends to more thoroughly contact
leaves and other vegetative organs exposed to the air currents in which the
aerosols move.
When such off-target movement transports a herbicide to the vegetation-free
soil surface of a newly planted field, coverage of portions of that field may
be very uniform. This uniformity may enhance the herbicidal effect somewhat,
but a greater effect may occur from hydraulic incorporation of the herbicide
into the upper soil horizon in which crop seeds are planted and through which
the embryonic plants must emerge. If the herbicide is confined in that
surface layer of the soil, the concentration may be sufficient to elicit
symptoms, even in small amounts. Crop sensitivity is greatest when organs are
in the formative stages, and plants may then respond to doses far below
herbicidal doses.
Establishment of threshold doses requires determination of the criteria of
plant responses and application of those criteria to plants exposed to dosages
both above and below the threshold. Determination of the most discriminating
criteria can only be done by observing responses at those dosages, so research
in this area should involve observations of various kinds while seeking to
identify the thresholds. The expectation is that one of the hypothetically
appropriate criteria will prove to be appropriate in fact.
Our studies utilize primarily field experimentation to provide applicable
data. We have worked with sulfonylurea as a soil-active herbicide to minimize
the confounding effects of climatic and botanic variation. Our threshold dose
investigations were done with a logarithmic sprayer. The concept of
logarithmic spraying is not new; descriptions of equipment for that purpose
are in the literature (4).
Sulfometuron is widely used for roadside weed control, Conservation Reserve
Program plantings, conifer release, forestry site preparation, industrial turf
and other non-crop uses (1). The high activity of sulfometuron results in a
high likelihood of symptom expression when nearby desirable crops are exposed
to low-level off-target movement of the herbicide.
Sulfometuron applications have been involved in several instances of exposure
that have produced effects on potato, lettuce ((Lactuca sativa L.), pulses,
onion (Aliiurn cepa L.), sugar beet, alfalfa and other crops in the irrigated
western U.S.A. Data describing these effects have been neither well
documented in the literature nor readily available to the scientific
community. Preparation of recommendations and regulatory guidelines for
sulfometuron use and injury management is therefore difficult without
resorting to an attitude of paranoia.
92
-------
MATERIALS AND METHODS
Husbandry
We prepared a silt loam soil by conventional seedbed tillage procedures, then
planted five test crops in 6 x 30 m plots in a randomized complete block
design using commercial equipment. The crops were planted in late spring
(June 2) to minimize the risk of exposure to accidental drift of herbicides
from local grain farming operations. Seeding rates were 2,000 kg (38,000
propagules)/ha of potatoes, 1 kg (22,000 seeds)/ha for sugar beets, 120 kg
(480,000 seeds)/ha for peas, 60 kg (540,000 seeds)/ha for lentils and 14 kg
(3,000,000 seeds)/ha for alfalfa. Peas, lentils and alfalfa were planted in
17 cm rows with a 3 m wide drill; sugar beets were planted in 56 cm rows with
a plate planter; and potatoes were planted in 90 cm rows with a two-row
assist-feed plate planter. Weed suppression was with pre-emergence treatments
one day after planting, with 2.2 kg/ha EPTC (s-ethyl dipropyl carbamothioate)
for alfalfa, 3.9 kg/ha EPTC plus 0.56 kg/ha metribuzin (4-amino-6-(l,l-
dimethylethyl)-3-(methylthio)-l,2,4 triazin-5(4H)-one) for potato, 3.4 kg/ha
cycloate (s-ethyl cyclohexylethyl carbamothioate) for sugar beets, and by
supplemental hoeing and hand pulling in all plots. All other management
practices were done by conventional husbandry practices.
Experimental Treatments
Sulfometuron was applied prior to crop emergence four days after the above
herbicide treatments, with a logarithmic sprayer mounted on a tractor
traveling 0.6 m/s. The propellant was C02, with a pressure of 97 kg/cm2. The
resultant output was 667 L/ha.
Logarithmic Sprayer Design
The logarithmic sprayer components included a concentrate tank constructed
from a 10-cm diameter cam-lock brass male industrial hose connector, with a
brass plate welded over the hose end to form a 500 ml tank with interior
dimensions 76 cm x 13 cm. The tank lid consisted of a 10 cm diameter female
brass hose connector end cap. The solution exited the tank through a 1.25 cm
pipe screwed into a drilled, tapped 1.25 cm hole in the center of the brass
plate. The pipe led to a quarter-turn system shutoff valve attached to a six-
hole manifold 15 cm from the tank. The solution was distributed from the
manifold through six 8-mm diameter equal-length plastic hoses to six flat fan
(Teejet 8004) boom nozzles spaced 50 cm apart on the spray boom.
The diluent inlet to the solution tank was through a 10 cm-long, 10 mm
diameter copper manifold pipe screwed into a 10 mm hole drilled and tapped in
the tank lid, so that the pipe was positioned in the center of the concentrate
tank when the lid was closed. This pipe had sixteen 1 mm holes drilled in
four rows along the length of the pipe to produce theoretically instantaneous
mixing for constant dilution during the operation of the sprayer. The water
diluent was fed to the solution tank inlet pipe from a hose leading from a
93
-------
C0.,-pressurized 12-L tank. The diluent hose was connected to the concentrate
tank cap with a 1-way shutoff quick connector.
Sprayer Calibration
The sprayer was calibrated in 3 replicates by collection of the water diluent
containing Rhodamine B red dye as the tractor sprayed the dye solution through
the logarithmic system over a test path. The dye was mixed at a concentration
of 2 g/L, sprayed, and collected in 15 cm petri dishes spaced at 30 cm
intervals for the first 10 m, 1m intervals for the next 10 m, and 2 m
intervals for the last 10 m. The petri dish contents were rinsed with 20 ml
water into 25 ml screw-capped vials and the dye concentration was determined
with a spectrophotometer measuring light transmission at 560 nm. The actual
logarithmic dye concentration curve was compared with a theoretical log-decay
curve derived from the equation C, = C0 * ert/v where C0 is the initial dye
concentration in the log sprayer's concentrate tank; C, is the dye
concentration expressed as a percentage of C0 at any time t in seconds after
the sprayer is turned on at the beginning of tractor travel along the test
path; r is the rate of diluent flow in ml/s; v is the volume of the
concentrate tank in ml; and e is 2.71828. C0 and C, may be converted to g/ha
when comparison with agricultural application rates is desired, by taking into
additional consideration absolute concentration in g/L and width of the
sprayed swath.
Sprayer Operation
Correct operation of the logarithmic sprayer system depended upon filling the
concentrate tank completely full with the solution at concentration C0. The
cap was then attached to the tank, and the filled diluent hose can be
connected. The system was then pressurized with C02. The tractor forward
motion was started, after which the system shutoff valve was opened to begin
operation. The solution began to hit the soil surface, at apparent maximum
pressure, 0.5 m after the valve was opened. The tractor continued to move
over the 30-m path, and the system was closed after the path had been
traversed. After the 30-m path was completed, the petri dish contents were
collected, the concentrate tank pressure was relieved by the quick disconnect
coupler at the cap, the tank was emptied and the process was repeated.
Sulfometuron was mixed at a concentration of 4.3 x 10"3 g/L and applied with
the logarithmic sprayer to one 3 x 30 m half of each of the five 6 x 30 m
plots (one crop species per plot) in each replicate. The plots were
immediately irrigated with approximately 0.6 cm water with a fixed-position
(solid-set) irrigation system with sprinklers spaced on a 15 x 22 m grid over
the experiment. Soil moisture was maintained above 60% available soil
moisture thereafter by irrigating as needed, which resulted in a total water
application of about 50 cm between planting and harvest.
Cross-contamination was minimized by applying the sulfometuron after all other
field operations were complete, and no further human or equipment entry was
allowed into the area until after two irrigations had moved the herbicide into
94
-------
the soil. Disease control was accomplished by aerial application of
fungicides.
Pea, alfalfa and sugar beet heights were measured and symptom ratings
estimated at each of seven dosage levels two, four, and six weeks after
treatment. An additional alfalfa measurement was made after 12 weeks, and
sugar beet height measurements were made after 10 and 12 weeks. Potato vine
lengths were measured after 4, 6, 10 and 12 weeks. Lentil and pea plants were
harvested and dried after six weeks, alfalfa plants were harvested after ten
weeks, sugar beet plants were harvested after 15 weeks, and potatoes were
harvested after 16 weeks.
RESULTS
Logarithmic Sprayer Calibration
A plot of the actual curve of the dye concentrations of the solution samples
collected in the petri dishes followed the theoretical log-decay curve very
closely after C, = 0.75, i.e., the peak concentration occurred 2.74 m after
the system valve was opened (Fig. 1). C0 was not reached due to the time lag
in filling and pressurizing the boom completely. This lag occurred during the
time in which the concentration ranged from 100% (CJ at t=0 down to 75% at
t=4.5s, 2.7 m from t=0. To reach C0=l, it was necessary to mix the solution
to a concentration where C0=1.32. This adjusted for the delay and reduction
in the peak concentration. When this was done, the adjusted C0 occurred at
approximately 2.7 m.
When sulfometuron was mixed to provide an adjusted C0 (base dosage) of 2.2
g/ha, and where t=4.5 s at a point 2.74 m from theoretical t=0, the actual
dosage applied at 7.3 m was 1/2 of the adjusted base; at 12.5 m the dosage was
1/4; at 17.4 m it was 1/8; at 24.4 m it was 1/16 and at 28 m it was 1/32, or
3%, of the adjusted base dose.
Plant Response
Observations indicate that each species responded to very low doses of
sulfometuron in measurable ways. Plant response data are reported elsewhere
(3) and are therefore not included here. It is apparent that responses in
these crops to sulfometuron occur at much lower doses than previously
considered.
Shoot growth of all crops except potato tended to respond to doses as low as
0.07 g/ha in early growth stages, and to recover from symptoms as the season
progressed. Potato shoots showed transient indication of exposure to 0.55
g/ha or more; however the tubers showed persistent symptoms of exposure to
sulfometuron. Although commercial propagules of this crop are large (40-60
g/seedpiece) in comparison with those of the other species tested, the
treatments produced morphogenic effects such as longitudinal cracking and
periderm thickening. The effects were substantial at 2.2 g/ha sulfometuron,
95
-------
Spectrophotometric
Absorbance (%)
100
CQ (theoretical)
^-^CQ (adjusted)
Measured
Theoretical (Ct =
v = 500.00 ml
r = 20.25 ml/s
speed = 0.6 m/s
0 A 5
Peak
10 15 20
Meters Traveled
o.03
25 A 30
t=47 s.
Figure 1 - Change in solute concentration during application through the logarithmic
sprayer.
but were not reliably observed below 0.27 g/ha. This phenomenon has not yet
been explained in terms of acetolactate synthase (ALS) inhibition, which is
the generally accepted main mechanism of action (2). No distinctive
morphogenic effects other than stunting were observed on the other crops. It
was apparent that the perennial non-determinate nature of alfalfa and potato
allowed those two crops to survive treatment. The sensitivity and modes of
survival of these two species are very different. Alfalfa, which germinates
from small seeds (2 mg/seed) but which produces a long perennial taproot,
shows great sensitivity early in the season but suffers little stand loss and
recovers well. Potato, which grows from a large seedpiece, exhibits no
symptoms until tubers can be observed, and appears to show only morphogenic
symptoms at the near-threshold levels used in these studies. Pea, lentil and
sugar beet displayed suppression symptoms; and although plant recovery
96
-------
subsequently occurred, reduced biomass yield at the higher doses appeared to
persist until harvest.
CONCLUSIONS
Threshold levels can be successfully studied with logarithmic dose sprayers.
Precision in application is not as high as can be obtained with constant-dose
sprayers, but logarithmic applicators allow observation over a continuous
range and provide an efficient means of estimating threshold levels. More
detailed studies may be conducted after the approximate thresholds are set.
A much greater range of doses than was discussed here may be tested by simply
reconfiguring the logarithmic dose system through changing either concentrate
tank volume or flow rate, or by changing both.
Plant responses to near-threshold herbicide doses may deserve more analytical
attention in the future as herbicide families become more diverse. Such plant
response information will be very helpful for the science of diagnostics,
which is becoming more important in modern agriculture. Agriculturists have
in the past carefully studied phytotoxicity and associated phenomena insofar
as it has related to doses approximating those in conventional use. Much
plant physiological information has been gained by those studies. It is
apparent that quantities once thought to be subclinical are not so, and
deserve attention.
REFERENCES
Anonymous. 1987. Oust herbicide (product label). E.I. Dupont De Nemours and
Co., Inc. Agricultural Products Department, Wilmington.
Beyer, E.M., McDuffy, M.J., Hay, J.V. and Schlueter, D.D. 1989. Sulfonylurea
herbicides. In: Herbicides: Chemistry, Degradation and Mode of Action. Vol.
3. 117-189 pp.
Lass, L.W., Callihan, R.H. and Hiller, L.K. 1990. Dose-response of peas,
lentils, potatoes, sugar beets and alfalfa to sulfometuron. in: Proceedings,
Western Society of Weed Science. 43:14-15.
Wiese, A.F. 1986. Herbicide application. 1-27 pp. In: N.D. Camper (ed.),
Research Methods in Weed Science, Ed. 3. Southern Weed Science Society,
Champaign. 486 pp.
97
-------
RESEARCH REPORT ON
1988 POTATO-HERBICIDE INJURY RESEARCH
by
Philip Westra*, Gary Franc, Brian Cranmer
and Tim d'Amato
Colorado State University
Content: A field experiment conducted to determine the influence of Oust,
Glean, Amber, Ally, Harmony Extra, and Assert on potato was
described. The measurements used to determine vegetation injury
and crop yield were discussed.
INTRODUCTION
Confirmed cases of Oust herbicide injury to potatoes in 1987 in the San Luis
Valley prompted this research which was designed to document foliar and tuber
injury caused by foliar applications of Oust, Harmony Extra, Assert, Amber,
Glean, and Ally herbicides. Oust was the only non-crop land herbicide used in
this study. Harmony Extra and Assert were included because of their barley
marketing potential in the valley; this research simulated drift or
misapplication of these two products. Glean and Amber were included to
broaden our understanding of sulfonylurea herbicide effects on potatoes.
Russett burbank and centennial russett potatoes were evaluated in this study
because of their market dominance and importance in the valley. All
herbicides were applied July 1 during tuber initiation to one set of plots,
and on July 14 during tuber bulking to a second set of plots.
METHODS AND MATERIALS
Potatoes were planted on May 18, 1988. Seed stock was the highest research
quality available from the Colorado State University Center Research Station.
Potatoes were planted in plots 14 feet in length, but consisted of two rows of
potatoes planted 34 inches apart; one row (14 plants) of russet burbank, and
one row (14 plants) of centennial russet potatoes. One plant from each end of
the plot was discarded to eliminate border effects.
* Presenter
98
-------
The study consisted of three replications of a randomized complete block
design. There were 60 plots with each plot containing two potato cultivars,
thus yielding a total of 120 subplots.
Oust, Glean, Ally, Harmony Extra, and Assert were applied on two dates
(July 1 and July 14, 1988) to study separate plots of potatoes which were
respectively in the tuber initiation and tuber bulking stages at the time of
application. All herbicide treated plots received only one herbicide
application. Treatments administered at the rates shown in Table 1 were
applied over the top of the potato plants with SS11001 flat fan tips (new)
under 15 psi using a carbon dioxide powered back pack sprayer. An electronic
metronome was used to calibrate walking speed. Water for spraying came from
the CSU campus at Ft. Collins. All herbicides were weighed out on a Mettler
H10 balance, or measured to 0.1 ml accuracy with a 1 ml pipette. The sprayer
was rinsed 4 times with 1) water, 2) bleach solution, 3) ammonium solution and
4) water, between treatments. Environmental conditions at the time of
application were excellent and there was no evidence of herbicide drift or
movement in the research plot area. Various data were collected during the
growing season and at harvests which occurred on August 18 and September 22,
1988.
RESULTS AND DISCUSSION
Vegetation Analyses for Injury
TWO WEEKS FOLLOWING JULY 1 APPLICATION (OBSERVATIONS ON JULY 14)
The high rate of Oust and Harmony Extra, plus Amber and Assert caused a
significant increase in visual injury symptoms as well as chlorosis in both
potato varieties. The high rate of Oust and Harmony Extra significantly
reduced potato plant height for both varieties. Additionally, Amber and
Assert significantly reduced potato plant height for the centennial russett
variety.
By two weeks after application, foliar effects could be observed and measured
for Oust, Harmony Extra, Amber, and Assert. Glean and Ally produced no
significant effects on either variety.
THREE WEEKS FOLLOWING JULY 1 APPLICATION AND ONE WEEK FOLLOWING JULY 14
APPLICATION (OBSERVATIONS ON JULY 22)
Flower Number
Virtually all of the herbicides significantly reduced flower numbers at both
times of application, although the reduction was more severe for the July 1
application than for the July 14 application. Flower number reduction was
very striking in the field, averaging 37% reduction for russett burbank, and
47% reduction for centennial russett across the entire study.
99
-------
Table 1. Application rates for each of the herbicides used.
TRT
NO
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
TREATMENT
NAME
CHECK
OUST
OUST
Glean
Glean
Harmony EXTR
Harmony EXTR
Amber
Ally
ASSERT
CHECK
OUST
OUST
Glean
Glean
Harmony EXTR
Harmony EXTR
Amber
Ally
ASSERT
FD
DF
DF
DF
DF
DF
DF
DF
DF
L
DF
DF
DF
DF
DF
DF
DF
DF
L
AI
#/gai
75
75
75
75
75
75
75
60
2.5
75
75
75
75
75
75
75
60
2.5
RATE
.071
.141
.035
.071
.142
.282
.071
.018
.47
.071
.141
.035
.071
.142
.282
.071
.018
.47
RATE
UNIT
oz/A
oz/A
oz/A
oz/A
oz/A
oz/A
oz/A
oz/A
Ib/A
oz/A
oz/A
oz/A
oz/A
oz/A
oz/A
oz/A
oz/A
Ib/A
GROW A
STGE C
JULY 1
JULY 1
JULY 1
JULY 1
JULY 1
JULY 1
JULY 1
JULY 1
JULY 1
JULY 1
JUL 14
JUL 14
JUL 14
JUL 14
JUL 14
JUL 14
JUL 14
JU1 14
JUL 14
JUL 14
ML
TRT
0.
0.
0.
0.
0.
0.
0.
0.
1.
0.
0.
0.
0.
0.
0.
0.
0.
1.
OR G/
. MIX
00733
01456
00361
00733
01466
02911
00733
00232
94165
00733
01456
00361
00733
01466
02911
00733
00232
94165
1
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
2
216
208
201
209
204
217
212
214
220
203
218
202
207
219
206
205
215
211
213
210
3
314
309
318
320
305
304
316
303
315
313
307
302
317
306
308
312
311
319
301
310
100
-------
Canopy Cross Section
Significant differences in potato canopy height and row closure were obvious.
By measuring the height and width of the canopy, a canopy cross section could
be calculated. All of the herbicides except Ally and Glean at the low rate
caused significant reductions in canopy cross section. Oust at the .141 oz
ai/A caused the most reduction. Canopy reduction from the July 1 application
was more severe than from the July 14 application. Russett burbank cross
section was reduced 24% and centennial russett cross section reduced 29%,
averaged across all herbicides.
Plant Chlorosis and Stem Discoloration
Neither of these variables was evaluated because the degree of damage was
minimal or inconsistent across replications. In general, Oust, Assert, and
the high rate of Harmony Extra caused detectable, slight chlorosis, on the
order of 10 - 15% lighter colored leaves. Foliage color following application
of all other herbicides was normal, or very nearly normal. None of the
herbicides, at the rates tested, caused obvious yellowing or highly chlorotic
foliage. Although purple stem discoloration was noted in some plots, the
degree of discoloration was inconsistent and did not warrant detailed
evaluation.
A striking foliar symptom noted in all Oust treated plots, particularly at the
higher rate, was a foliar symptom which looked like drought stress or some
sort of viral or psyllid injury. This is apparent in detailed photographs
taken on July 22. The effect was consistent across replications, and was
quite severe for the high rate of Oust. Dr. Gary Franc (personal
communication) first thought that we had psyllid injury in certain plots,
which turned out to be the Oust treated plots. This reinforces the conclusion
that of all the herbicides tested, Oust caused the most noticeable and
striking changes in potato foliar characteristics. On July 22, photographs
were taken of both potato varieties for all treatments in this study.
Reproduction/Yield Analyses
PRELIMINARY HARVEST ON AUGUST 18, 1988
A preliminary harvest of three plants was conducted on August 18, 1988 from
the leading edge of each plot. In general, the conclusions drawn from the
August 18, 1988 harvest very closely paralleled the conclusions from the final
harvest made on September 22, 1988. Therefore, the data presented here will
be from the final harvest (September 22, 1988) which consisted of an average
of nine plants per plot.
FINAL HARVEST ON SEPTEMBER 22, 1988
Tuber Numbers
For the Russett Burbank variety, the July 1 application of both rates of Oust
caused a significant increase in tuber number; tuber number was increased 217%
following application of the high rate of Oust (Table 1). It was visually
101
-------
striking during harvest to see the proliferation of small tubers caused by the
July 1 application of Oust; this is evident in tuber photographs from the
final harvest. July 14 applications of Oust did not cause a significant
increase in tuber number. For the centennial russett variety, Oust at the
high rate applied July 1 and July 14 caused a significant increase in tuber
number. No significant change in tuber number was observed for any other
herbicide.
Average Tuber Weight
Russet Burbank
Oust, Harmony Extra, and Amber applied July 1, as well as Oust and Harmony
Extra at the high rate applied July 14 significantly reduced average tuber
weight. Oust at the high rate applied July 1 reduced average tuber weight by
84%.
Centennial Russet
Oust and Assert applied July 1, and Oust, Harmony Extra, Amber, and Assert
applied July 14 significantly reduced average tuber weight. Centennial
Russett average tuber weight was more sensitive to Assert than was Russett
Burbank average tuber weight. Oust at the high rate applied July 1 reduced
average tuber weight by 71%.
Tuber Quality
Russet Burbank
Normal Tubers. Oust, Harmony Extra, and Amber applied July 1 as well as Oust,
Harmony Extra, Amber, Ally, Assert, and Glean at the high rate significantly
reduced the percentage of normal tubers harvested. All other treatments did
not significantly lower the percentage of normal tubers harvested.
Cracked Tubers. Assert and Harmony Extra applied July 1 and July 14, as well as
Oust and Amber applied July 14 significantly increased the percentage of
cracked, abnormal tubers harvested. Tuber cracking was not a predominant
symptom from the July 1 application of Oust.
Folded Tubers. Oust applied July 1 and July 14 as well as Harmony Extra applied
July 1 caused a significant increase in folded tubers which suggests that
tuber growth and development was abnormal following application of these
herbicides.
Popcorn Tubers. Only Oust and Harmony Extra at the high rate applied July 1
caused the formation of very abnormal popcorn tubers which were small in size
and covered with numerous bumps and knobs.
Knobby Tubers. Oust applied on July 1 and July 14 was the only herbicide to
cause a significant increase in the percentage of medium sized tubers with
large knobs and protrusions on the tuber surface.
102
-------
Minuscule Tubers. Oust applied July 1, and Oust at the hjgh rate applied July 14
caused a significant increase in the percentage of very small, minuscule
tubers.
Centennial Russet
Normal Tubers. Oust, Amber, Assert, and Harmony Extra applied July 1 (only
Harmony Extra at the high rate) and July 14 caused a significant reduction in
the percentage of normal tubers harvested. Oust at the low rate applied July
14 reduced normal harvested tubers by 97%.
Cracked Tubers. Oust and Assert applied July 1 and July 14 as well as Harmony
Extra and Amber applied July 14 significantly increased the percentage of
cracked, abnormal tubers harvested. Cracked tuber symptomology was most
exaggerated for Oust treatments.
Folded Tubers. Oust at the low rate, Harmony Extra at the high rate, and Assert
applied July 1 significantly increased the percentage of abnormal, folded
tubers harvested. No other treatments had significant effects on percentage
of folded tubers harvested.
Popcorn Tubers and Knobby Tubers. None of the herbicides caused any significant
effects in these two classifications, indicating that this symptomology was
not characteristic of centennial russett potato response to any of the
herbicides tested.
Minuscule Tubers. Harmony Extra at the high rate applied July 1 caused a slight,
but significant increase in the percentage of very small tubers harvested.
CONCLUSIONS
1. In general, the July 1 application of herbicides during tuber initiation
was more damaging to yield and tuber quality than the July 14 application
during tuber bulking phase.
2. Oust damage symptomology to russett burbank tubers shifted dramatically
from the July 1 application to the July 14 application. The early application
caused folded, knobby, popcorn tuber symptomology with very few tuber cracks
evident, and the proliferation of many small tubers. The late application
symptomology was predominantly tuber cracking.
3. The order of increasing severity of injury to potatoes in this study was:
UT. Check < Ally < Glean < Amber < Harmony Extra < Assert < Oust
4. Tuber symptoms and tuber damage were more obvious and more severe than
foliar symptoms or foliar damage following application of these herbicides.
103
-------
5. In light of the perceived weakness of the sulfonylurea herbicides on plants
in the Solanaceae or nightshade family, the severity of Oust damage to
potatoes (a member of the nightshade family) was somewhat surprising.
6. Oust, even at the lowest rate tested, was extremely damaging to potato
tubers. Its level of tuber damage was several orders of magnitude greater
than the other sulfonylurea herbicides tested. The effects of Oust on tuber
size and tuber quality virtually eliminated the production of any marketable
tubers. Oust and growing potatoes are an extremely bad mix. This indicates
that under no circumstances should Oust be allowed to contaminate environments
where potatoes are grown.
7. Assert, either drifting or at field label rates, should never come into
contact with the foliage of growing potatoes as it causes totally unacceptable
tuber cracking which results in nonmarketable tubers. Assert primarily caused
tuber cracking.
8. Harmony Extra, either drifting or at field labeled rates, should never come
into contact with the foliage of growing potatoes as it causes totally
unacceptable tuber folding which results in non-marketable tubers. Harmony
Extra primarily caused folded tubers.
9. Small amounts of Ally, Glean, or Amber drifting onto growing potatoes
likely would cause slight to minimal potato tuber injury. Of these three,
Ally and Glean would cause the least injury.
10. The Russett Burbank variety was more sensitive to the herbicides in
general, and specifically to Oust, than the centennial russett variety. If
potatoes had to be planted back into Oust contaminated soil, the use of
russett burbank potatoes would be a poor choice; use of centennial russett
would be the preferred choice.
11. Although some of the herbicides significantly reduced tuber yields, a more
objectionable aspect was the effects of some of these herbicides on potato
tuber quality; some herbicides produced tubers which were totally
nonmarketable.
12. This research suggests that some of these herbicides, and especially Oust,
may adversely affect potato tubers at very low concentrations. This raises
the possibility of herbicides such as Oust being able to adversely affect
potato tuber growth at concentrations below current analytical detection
limits. This interaction of potatoes with herbicides which have biological
activity at extremely low concentrations warrants further research.
13. It must be emphasized that in the 1988 study, all herbicides were applied
with a backpack sprayer to the leaves of actively growing potato plants. The
data contained in this report, and in the preliminary data summary, SHOULD NOT
BE USED to draw conclusions about the effects of low carryover levels of some
of these herbicides in following years. That issue would most properly be
addressed by new research on the effects of low soil levels of these
herbicides on potatoes.
104
-------
SYMPTOM EXPRESSION WITH SELECTED HERBICIDES
ON FOUR PERENNIAL PLANT SPECIES
by
Robert Parker
Washington State University
Content: A field investigation to determine the vegetative response of four
nontarget perennial plants (cherry, rose, grape, and alfalfa) to
chlorsulfuron, thifensulfuron, glyphosate, bromoxynil, 2,4-D amine,
and glyphosate + 2,4-D (Landmaster BW) was described.
INTRODUCTION
This work is in response to allegations that herbicide drift from Horse Heaven
Hills wheat ranches caused severe crop damage in the Badger Canyon and lower
Yakima Valley beginning in 1987. The specific objectives were to: 1)
demonstrate, under field conditions, symptoms from herbicides that are considered
possible causes of herbicide drift injury in the Badger Canyon and lower Yakima
Valley and determine the time after application injury is first observed and
follow subsequent plant response and 2) measure the extent to which herbicides
can be taken up by foliage from dust, and the effect of dew on the uptake of the
herbicide from the soil. Only objective 1 will be presented in this paper.
METHODS AND MATERIALS
Field experiments were conducted at the Roza Unit of Washington State
University's Irrigated Agriculture Research and Extension Center near Prosser,
Washington on one-year-old Rainier cherry trees, Lemberger grapes, assorted rose
varieties, and Vernal alfalfa. The silt loam soil had pH 8.1 and contained 12
ppm of available N (NH4+ + N03-), 9 ppm of extractable P, 110 ppm of exchangeable
K, and 8000 ppm of organic mater. Cultural practices, pruning, insect, and
disease control were according to Washington State University recommendations.
The cherries, grapes, and roses were furrow irrigated and the alfalfa was
sprinkle irrigated.
Single application of herbicides were made with a carbon dioxide-pressurized
knapsack sprayer equipped with three (cherry and alfalfa) or two (grape and rose)
8006 flat fan nozzles and calibrated to deliver 374 L/ha at 15 psi. Herbicides
were applied over the top of rose and alfalfa plants, and sprayed on one side
of the cherry and grape by positioning the spray boom to direct the spray on the
plant towards the center of the plants. To block spray drift to surrounding
plants, plastic shields were placed to block spray particles from moving to
105
-------
adjacent plants during application and plots were sprayed during the early
morning when the air was calm. Rates used were 0.26, 0.88, 2.63, and 8.76 g
ai/ha of thifensulfuron and chlorsulfuron; 4.2, 14.01, 42.03, and 140.1 g ai/ha
of bromoxynil; 4.2, 14.01, 42.03, and 140.1 g ae/A glyphosate; 11.21, 37.36,
112.08, and 373.62 g ae/ha 2,4-D; and 3.2 + 5.3, 10.5 + 17.5, 31.5 + 52.5, and
105.1 + 175.1 g ae/ha of glyphosate + 2,4-D (Landmaster BW). Rates represented
1/100, 1/30/ 1/10, and 1/3 of the maximum use rate in the state of Washington
for wheat or fallow fields. In addition to these treatments, alfalfa received
a combination of chlorsulfuron with thifensulfuron and chlorsulfuron or
thifensulfuron with bromoxynil, glyphosate, or 2,4-D. Herbicide rates in the
combination treatments were 1/30 of the high use rate. Herbicides were applied
as a liquid in 374 L of water per ha with Ortho X-77 at 0.25% (v/v) as a
surfactant in all treatments. Nontreated plots were included for each herbicide.
Applications were made at the fourth leaf in cherry and grape of April 20, 1990
and May 3,1990, respectively. Alfalfa regrowth (after first cutting) was 15 cm
tall and rose were leafed out and 40 cm tall when treated (April 30, 1990 for
roses and June 1, 1990 for alfalfa). Individual plot size was two plants for
cherry, grape, and roses with two plants serving as a buffer between adjoining
plots (Figure 1). Each plot of alfalfa was 1.5 x 1.8 m. Treatments were
replicated three times in cherry, grape, and roses and four times in alfalfa.
Each experiment was a randomized complete block design with a factorial treatment
arrangement. Treatment responses were determined by conventional analyses of
variance. Differences among means were tested by least significant differences.
Injury symptoms on all species were observed and photographed during the entire
1990 growing season. Plant injury was evaluated visually with 0 = no visible
injury to 100 = complete plant death.
Leaf area measurements were made on three branches from the treated side of
cherry trees four months after application with a photoelectric meter. On the
same branches internode length was measured. Gain in plant height and stem
diameter were measured four months after herbicide application in cherry.
Grape clusters were hand harvested when grapes in the nontreated controls reached
a 21 degree brix level. Total number and weight of clusters per plot were
recorded. Berries were stripped from 15 clusters and the juice extracted and
evaluated for soluble solids, pH, and color.
Alfalfa was observed for symptoms and plant heights were measured before each
cutting for the remainder of the growing season. When alfalfa reached the early
bloom stage of growth, forage from aim wide strip 1.8 m long was mowed, dried
at 75 C for four days and dry matter yield was determined. After sampling, the
entire experimental area was uniformly mowed.
RESULTS
The development of symptoms over time have been recorded for each herbicide in
a written and photographic form during 1990 and will be observed into 1991 to
determine if any residual effects are apparent. This experiment will be repeated
during 1991 on plants established in 1990 with similar evaluations to follow.
106
-------
All data collected during the 1990 growing season are preliminary and are not
presented in this report. A final report will be written when the experiment
is completed.
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
BO
X
Bl/10
X
Bl/100
X
Bl/30
X
Bl/30
X
Bl/10
X
BO
X
Bl/3
X
Bl/30
X
Bl/100
X
X
X
Bl/100
X
Bl/10
X
Bl/30
X
80
X
Bl/3
X
GL1/30
X
6LO
X
GL1/100
X
GL1/3
X
GL1/10
X
X
X
X
X
Gl/100
X
GO
X
61/30
X
61/10
X
61/3
X
HI/10
X
HI/30
X
Hl/3
X
HI/100
X
HO
X
X
X
LO
X
LI/10
X
11/3
X
LI/30
X
LI/100
X
DO
X
Dl/3
X
Dl/100
X
01/10
X
01/30
X
X
X
X
X
HI/10
X
HI/30
X
HO
X
Hl/3
X
HI/100
X
LI/10
X
LI/30
X
LI/100
X
LO
X
Ll/3
X
X
X
GL1/3
X
6L1/30
X
6LO
X
GL1/100
X
GL1/10
X
60
X
61/30
X
61/100
X
61/10
X
61/3
X
X
X
X
X
Dl/30
X
Dl/10
X
DO
X
Dl/3
X
Dl/100
X
HI/10
X
HI/30
X
HO
X
HI/100
X
Hl/3
X
X
X
LI/10
X
Ll/3
X
LI/30
X
LI/100
X
LO
X
6L1/10
X
GL1/30
X
GLO
X
GL1/100
X
GL1/3
X
X
X
X
X
Dl/100
X
DO
X
Dl/10
X
Dl/3
X
Dl/30
X
61/3
X
61/10
X
61/100
X
61/30
X
GO
Figure 1 -Treatments applied on Rainier cherry trees on April 20,1990 in 374 L/ha water
at15psi. Concentration of high use rate applied: 0,1/3,1/10,1/30, and 1/100. Two trees
per treatment, three replications.
G = chlorsulfuron (Glean)
GL = glyphosate (Roundup)
H = thifensulfuron (Harmony)
D = 2,4-D Amine
L = glyphosate + 2,4-D (Landmaster BW)
B = bromoxynil (Buctril)
High Use Rate
0.375 oz ai/A
0.38 Ib ai/A
0.375 02 ai/A
1.0lbae/A
40.0 oz product/A
0.38 Ib ai/A
107
-------
IMPACT OF AIRBORNE PESTICIDES
ON
NATURAL PLANT COMMUNITIES
by
T. Pfleeger
U.S. Environmental Protection Agency
Content: A field method for studying the influence of pesticides on small
manmade plant communities was presented. The manner in which plant
communities were established, application of chemical and assessment
of impact were all discussed.
INTRODUCTION
The constituency for natural plant communities is small and without economic
persuasion. Therefore it follows that the impact airborne pesticides have had
on natural plant communities has received little attention. This is not to
suggest that natural plant communities are unimportant, but rather the contrary
is true. They play a central role in the dynamic stability of ecosystems
providing such things as energy and habitat structure for the assemblages of
organisms that inhabit them. What is known about pesticides and natural plant
communities has been generally limited to power line right of ways and roadsides.
Some airborne pollutants such as ozone (Miller, 1984) and sulfur dioxide
(Winterhalder, 1984; Legge, 1980} have received substantial attention and it has
been shown that they have a major phytotoxic impact on natural plant communities.
The impacts that low levels of drifting pesticides could be having on natural
plant communities may be subtle ecological effects in comparison to the blatant
phytotoxic effects caused by high concentrations of inorganic pollutants from
smelters and other industrial sources.
The current plant test protocols (Subdivision J), developed by EPA under
authority of FIFRA (Federal Insecticide, Fungicide and Rodenticide Act) use a
tiered approach for testing pesticide effects on nontarget plants. The first
two tiers use the same tests, seed germination, root elongation and vegetative
vigor, with tier II being more detailed. These tests generate dose response data
on specific phases of plant growth. Tier three is suppose to be a field test,
however, results from a tier III have never been submitted to OPP (R. Petrie and
C. Lewis, personal communication).
The existing literature on toxicity testing from single species to ecosystem
testing has primarily been done on aquatic organisms and more recently this has
included terrestrial wildlife species. Terrestrial plant testing and vascular
108
-------
plants in general have not received the attention in relation to the contribution
they make to the functioning of ecosystems.
The overall view from the literature is as test complexity increases, accuracy
and cost increase while precision and reproducability decrease (Levin and
Kimball, 1984). The only exception found to this general rule is the concept
that cost increases as test complexity does. Van Voris et al. (1985), in one
of the few multispecies terrestrial plant tests, compared the cost of their test
to comparable single species tests and concluded that cost differences were
insignificant. A similar finding was shown by Perez and Morrison (1985) and
Cairns (1983) between single species tests and multi-species test systems.
However, no where in the reviewed literature was the idea that costs were
comparable between ecosystem tests and single or multiple species tests.
Multi-species tests generally investigate ecological properties such as density,
biomass, productivity, food web connectivity, symbiosis, herbivory, parasitism,
competition, predation, food availability, nutrient processing, and community
structure and function (Mammons, 1981). The environmental conditions of the test
are generally more realistic than in single species test. Replication of the
test is possible. Control and containment of the chemical and the number of
species within an experiment is possible.
The objective of this paper is to demonstrate an example of a simple, low cost
test method that might qualify as a tier III test to protect natural plant
communities. To accomplish this, low concentrations of organic chemicals were
used in this study to 1) develop a methodology for studying their effects on
plant communities, 2) determine their influence on community composition and
abundance in model plant communities, and 3) determine their effects on
interspecific competition.
METHODS AND MATERIALS
Plant Materials and Conditions
The plant species were gathered as seeds in soil from the Oregon State University
Botany and Plant Pathology Farm located just east of Corvallis, Oregon. The
field containing the seeds has been disturbed annually for over ten years with
no direct application of agricultural fertilizers or pesticides (Lewis Tate,
personal communication, 1986). The disturbance, any combination of plowing,
discing or rototilling beginning in the late spring and continuing intermittently
to early fall, prevented plants from maturing during the summer and therefore
selected for winter annuals.
Soil containing the seed bank was collected from the top 5 cm of the field in
the late summer of 1987 and 1988, when aboveground vegetation was absent. The
soil was sieved through a 6 mm screen and mixed with a commercial potting soil
(Promix) in a 50:50 ratio by volume in the fall of 1987 and a 40:60 ratio in the
109
-------
fall of 1988. Promix prevented the farm soil from hardening and diluted the
seed density.
Fifteen raised beds were constructed from 2x8 inch lumber and enclosed an
inside volume of 0.6 m high and 0.9 m square with a soil block of 0.49 m3. The
beds were sufficiently high to minimize root interactions from adjacent vegeta-
tion. The wooden frames were filled to within 5 cm of the top with unfertilized
bulk soil or 'garden loam', purchased locally. Fertilizer was added to the top
of the garden loam in 1987. This was not repeated the following year because
fertilization caused excessive growth in certain species, making it difficult
to harvest individual plants. The 'garden loam' was covered with 1.5 cm of the
seed bank soil mixture. The beds were irrigated until the fall rains began.
The 12 most common plant species that emerged from the seed bank represent eight
families (Table 1); most are widely distributed throughout the United States and
other parts of the world.
Three agricultural chemicals, atrazine (2-chloro-4-ethyl-amino-6-isopropylamino-
s-triazine), 2,4-D (isoocytl ester of (2,4-dichlorophenoxy) acetic acid) and
malathion (o,o-dimethyl dithiophosphate of diethyl mercaptosuccinate) were
selected as treatments, based on their widespread use in the United States and
the large amount of published research that has been done with these chemicals
(Table 2). Chemical treatments were randomly assigned to beds, in triplicate.
The chemicals were applied after plants had emerged and were less than five cm
tall. During chemical application, all beds were covered with black plastic
except the one receiving treatment. The treatments were applied using a hand
held sprayer with water as the carrier. The control beds received an equal
amount of carrier (water) as did the treatments.
Parameters Measured
Poa annua and Calandrinia ciliata were chosen as target species due to their
resistance to atrazine, high relative abundance, and taxonomic dissimilarity.
Target individuals of the two species were chosen for neighborhood analysis using
randomly selected coordinates and a portable grid. The individual of the
desired species closest to the coordinates became the target. Ten individuals
of each species per bed were chosen. No targets were located within a 10 cm
buffer zone around the outside of each soil block.
Percent cover was measured using nested circular quadrants with diameters of 10
and 20 cm, centered on the target individuals. For cover measurements, plant
parts that grew into the neighborhood were included and plant parts that grew
outside the neighborhood were excluded. The proportion of the neighborhood
covered by each species was recorded. Total cover equaled 100 percent and
included bare ground. This technique allowed for the repeated nondestructive
sampling of the same neighborhoods. Percent cover was measured four times.
Following the final cover measurements each year, all the target individuals were
harvested along with their 10 and 20 cm neighborhoods. All plants rooted within
the neighborhood were harvested at the soil surface. Plants were sorted by
species, dried to constant weight at 60° C, and weighed.
110
-------
Table 1 - Plant species that were most abundant in the artificial plant communities.
Nomenclature follows Hitchcock et al. (1969).
FAMILY
Asteraceae
Brassicaceae
Caryophyllaceae
Geraniaceae
Labiatae
Poaceae
Portulacaceae
Scrophulariaceae
SPECIES
Sencio vulqaris L.
Capsella bursa-pastoris (L.)
Moench
Draba verna L.
Cerastium viscosum L.
Spergula arvensis L.
Stellaria media (L.)
Cyrill
Erodium circutarium (L.)
L'Her
Lamium purpureum L.
Poa annua L.
Poa bulbosa L.
Calandrinia cil iata
(R. & P.) DC.
Veronica persica Poir.
COMMON NAME
groundsel
shepherd's purse
whitlow grass
annual mouse
- eared chickweed
spurry
chickweed
filaree
red dead-nettle
annual bluegrass
bulbous grass
red maids
creeping speedwell
CODE
SEVU
CABU
DRVE
CEVI
SPAR
STME
ERCI
LAPU
POAN
POBU
CACI
VEPE
111
-------
Table 2 - Composition, source and properties of the three organic chemicals used
as treatments.
ATRAZINE
2,4-D
MALATHION
Chemical formula C8H14C1N5
Manufacturer
Product name
Recommended
application rate
Percent of
Ciba-Geigy
AAtrex SOW
2.5 Ibs / acre
low = 8%
recommended rate high = 16%
used
Actual chemical low = 16.7 mg/m2
application rate high = 33.4
mg/mi
C,6H,6C1203
Albaugh
Lo-Vol 4D
2 pts / acre
low = 10.6%
high = 106%
low = 8.2 ul/m2
high = 81.9
ul/m2
Helena
CYthion
2 pts / acre
low = 106%
high =1060%
low = 81.9 ul/m2
high = 819.3
ul/m2
Total biomass of each bed was determined after the neighborhoods had been
harvested. Total biomass was determined by summing the biomass within the
neighborhoods plus the biomass remaining in the bed after the neighborhood
harvest.
The aboveground biomass data were analyzed using a one-way analysis of
variance (ANOVA) procedure with a protected LSD multiple range test (alpha <
0.05, N = 9). The biomass data were log transformed. The ANOVA and multiple
range test were used to determine if differences existed among treatments by
species for each chemical. In order to analyze changes between sampling
periods, means of the cover data (N = 60) from the 20 cm neighborhoods of both
targets for each treatment were plotted and examined.
Cover data used in the multiple regression analysis were transformed using the
square root of the arcsin of each value. The regression analysis was
performed on each treatment at each sampling period (alpha < 0.1 and 0.05, N =
30). The log transformed final aboveground biomass of the target species was
the response variable and the transformed cover values of each species within
either the 10 or 20 cm neighborhoods were the predictor variables [log biomass
= f(arcsin cover)]. The full model (y = B0 + 6,x, ... B8x8, where y = target
biomass, 6 = coefficients fitted for the regression, x, = cover of STME, xz =
cover of VEPE, etc.) was used, even though some species were not significant
to the model. This was done so that different treatments could be compared.
Following the work of Weldon and Slauson (1986), the importance of competition
was determined by the magnitude of R2 in the regression analysis.
112
-------
RESULTS
Biomass
Total biomass of the plot decreased with increasing atrazine application in
1988. The total aboveground biomass at harvest was significantly reduced
(P=.024) by the high concentration treatment (Figure 1). The low treatment
did not differ significantly from either the control or the high treatment.
The individual species constituting the communities demonstrated four distinct
patterns of biomass change when treated with atrazine (Figure 1). Stellaria
biomass decreased with increased chemical application. Veronica and Lamium
biomass was significantly reduced only at the high dose. Calandrinia.
Capsella. and Erodium demonstrated no significant change in biomass. However,
two of these species, Calandrinia and Capsella, showed a non-significant
biomass increase with increasing levels of atrazine. Finally, biomass of both
Poa species, the major monocots of the community, increased with an increase
in dose.
When 2,4-D was applied in 1989, the resulting communities had significantly
less biomass than the controls (P=.036) (Figure 1). However, there was no
significant difference between the low and high treatments. A two fold
increase in atrazine reduced biomass to 77 percent of the lower treatment
whereas a tenfold increase of 2,4-D reduced it to 88 percent of the low
treatment level.
The species responded differently to 2,4-D than to atrazine treatment (Figure
1). Stellaria. Veronica, and Lamium snowed no change in response to
application of 2,4-D. However, biomass of Calandrinia, Capsella, and Erodium
all decreased with an increased application of chemical. Capsella's response
occurred only with the high treatment level. Although neither Poa species had
a statistically significant difference among treatments (P=.090 and .463),
both produced the greatest biomass in the high treatment.
Malathion caused no significant decrease in community production (Figure 1)
(P=.192), even at five times the recommended dose. Erodium was the only
species that decreased in biomass (P<.0001), reacting even at the low dose.
Capsella showed a similar pattern, but it was not significant due to the large
amount of variability between replicate communities.
Cover
Cover patterns differed by species, chemical treatment, and sampling time,
with greater changes in atrazine and 2,4-D treatments. The atrazine data will
be shown as an example.
ATRAZINE
The atrazine control treatment started with Stellaria as the dominant species,
but by the second sampling period it was a co-dominant (species having similar
high amounts of cover) with Lamium and Veronica (Figure 2). At the fourth
113
-------
MO
Atrozine (1988)
(with fsrtilizar)
TOTAL B»y»SS (g/m2 )
CONTROL * SJ4 A
LOW • 710 AB
MICH « SSO B
STMC VCPC POAN
2. 4-D (1989)
TOTAL BWUASS (g/m* )
CONTROL • 204 A
LOW • 210 •
* MICH • ISA B
STME VCPC POAN LAPU CACI CA8U CRC! POBU
Malothion (1989)
TOTAL BIOUASS (g/m2 )
CONTROL • 294 A
LOW » 225 A
MICH • 252 A
STMC VCPC POAN LAPU CACI CABU ERCI POBU
Species
Figure 1 - Above-ground blomass at harvest. Treatments: ••-control croq -low
i—i -high. Species codes are defined in Table 2. Within each species,
treatments with the same letter do not differ statistically. Statistically
significant differences among treatments were identified using a protected
LSD on log transformed biomass data (alpha = .05, N = 0)
114
-------
60-
50-
40-
30-
20-
10-
0-^
60-
50-
(D40
O 30:
020
10
0
Atrazine 1988
Control
0 10 20 30 40 50 60 70 80 90 100
I
Low
0 10 20 30 40 50 60 70 80 90 100
50-
50-
40-
30-
20-
10-
0-
High
• «^» •
__^, a-.-^: i.-=-4n.-n"_1 1 _ a-
9"~'~- _ _ —a- """"
®" * *
6—~~" """"
— --«
Q
e
Ł
0 10 20 30. 40 50 . 60 70 80 90 100
Julian Days
Figure 2 -- Percent cover over time of the six species with the highest cover values
for the 1988 experiment using atrazine. The data are from 20 cm diameter
neighborhoods. Each symbol Is the mean of 30 samples, 10 from each of
three replicate communities. Species: -^- -STME ~A~ -VEPE- *--LAPU
-&- -POAN and POBU -&- -CACI -0- -CABU. Species code are defined
In Table 2. Julian days are from beginning of the calendar year.
115
-------
sampling, Stellaria was again dominant, due in part to completion of Lamiurn's
life cycle. Poa, Calandrinia and Capsella remained understory species for the
duration of the experiment except in the fourth sampling period, after
Capsella bolted, penetrated the canopy, and increased its cover.
At the low application rate of atrazine, Stellaria, the initial dominant, lost
dominance to Veronica and, to a lesser extent, Lamium (Figure 2). Stellaria
returned as a co-dominant as Lamium completed its life cycle before the fourth
sampling. The other three species remained in the understory for the duration
of the experiment.
The high treatment caused a radical change in how the community was structured
(Figure 2). Two species dominant in the control and low application
treatments, Stellaria and Lamium, were killed. Their death and the low
coverage of Veronica led to a community dominated by Poa, Capsella and
Calandrinia, the three species forming the understory in the control and low
treatment communities.
Competition
Competitive outcome differ by target species, chemical treatment and sampling
time under all senarios tested. Only Poa neighborhoods with atrazine
treatment are shown here as an example of the data.
REGRESSION USING COVER VALUES
1988, Atrazine Treatments, Poa Targets
In the control treatment, Lamium was the only species in the 10 cm Poa
neighborhoods that had interactions that were consistently significant (i.e.,
statistically significant in at least three of the four sampling periods)
(Figure 3). In contrast, however, Lamium was not a dominant species in the
control neighborhoods, as measured by aboveground biomass (Figure 1), or a
consistent competitor in the 20 cm neighborhoods (Figure 3). All major
species had a significant negative effect on target biomass at the second
sampling period in the control treatment; this sampling was also the most
important period of interspecific competition as measured by R2 (Figure 3).
With low atrazine treatment, Stellaria, Capsella, and Erodium were consistent
competitors in the 10 cm neighborhoods, while Lamium was the only species
consistently significant in the 20 cm neighborhoods. Veronica, the major
biomass contributor to the low atrazine community (Figure 1), was not a
consistent competitor. The low treatment had the highest level of
interspecific competition (as measured by R2) at the third and fourth
sampling, later than in the control. The low treatment had more significant
species interactions (19) than either the control (9) or the high (5)
treatments, indicating a dispersion of interspecific competition amongst many
species and over time (Figure 3).
The high treatment had only Erodium as a consistent competitor, with the
second sampling in the 10 cm neighborhood and the first sampling in the 20 cm
116
-------
neighborhood having the most interspecific competition (highest R2). Erodium.
while a consistent competitor in both the high and low treatments, was only a
minor contributor to community biomass under all treatments (Figure 2).
1
Thu
10
10
10
10
Atrazine
Target Species = Poa annua
Control
STME
VEPE
LWU
O8U
025
050
0.25
0.42
Low
10
10
10
10
High
0.30
0.40
025
0.18
TVne
Atrazine
I Target Species = Poa annua
Control
STME VEPE L*PU OCt CABU EFO RAPE
20
20
20
20
0.30
026
0.44
024
Low
20
20
20
20
0.34
026
0.57
0.64
20
20
20
20
High
0.41
029
0.34
024
Figure 3. Species that are cross hatched were significant competitors of the target
species Poa annua as determined from multiple regression analysis. Significance is at
the .1 level. Neighborhood cover was measured four times through the season. Target
biomass was measured at the conclusion of the season. The target neighborhood in
top figure was 10 cm and in the bottom figure the neighborhood was expanded to 20
cm. Species codes are defined in Table 1.
117
-------
DISCUSSION
i
Competition is just one of many species interactions within plant communities.
Other interactions such as allelopathy and herbivory cannot be discounted even
though no evidence was found of either. In any community, numerous
interactions are likely to be occurring, both positive and negative, with
direct and indirect effects. This experiment was designed to enhance the
probability of competition occurring. Synchronous seed germination was
encouraged so dominance and suppression resulting from early emergence were
diminished. Soil fertility was amended to ensure that resources were
available for sufficient plant growth to ensure interactions. Density
dependent mortality occurred, which is generally considered a symptom of
competition. The area where the experiments were conducted was isolated from
major herbivores. The raised beds further isolated the experiments from the
effects of the local soil conditions, including microsite variations. The
conditions and outcomes of the study strongly suggest that competition was a
major contributor to the structure of these communities. Therefore, the
experiments will be discussed in terms of competition.
Other processes besides interactions were probably altered by chemical
treatments. Organic pollutants have the potential to alter community
composition either directly through phytotoxicity or indirectly via secondary
effects. The secondary effects of these chemicals can include increased
nutrient uptake, resulting in increased herbivory and disease. While not a
factor in these experiments, temporary pollen sterility, decreased seed
germination rates and slowed decomposition rates can also be produced by these
pesticides, which would have an effect in natural environments.
Competition
Few studies have looked at the effects of organic pollutants on plant
competition. In this study, interspecific competition was severely altered
following application of organic compounds. The importance of interspecific
competition increased in some situations while it decreased in others. The
onset of competition was delayed in Calandrinia neighborhoods treated with
atrazine, but not in other treatments. A major change following treatment was
in the identity of influential competitors. The addition of a pollutant, even
malathion, an insecticide developed for use on plants, changed the competitive
hierarchy in almost every scenario tested.
These results suggest that plant communities are being subtly altered by
exposure to organic pollutants. Other studies investigating the relation
between competition and anthropogenic stresses have found similar
modifications. Changes occurred in the competitive balance in favor of
ryegrass when grown with clover in a replacement series experiment exposed to
ozone (Bennett and Runeckles, 1977). Bennett and Runeckles (1977) suggest
that the clover was more sensitive to ozone and therefore grew less. The
competitive balance within competing species pairs exposed to UV-B radiation
often changed dramatically (Fox and Caldwell, 1978).
118
-------
The introduction of xenobiotic compounds into the environment has many
potential effects and no particular method can test for all possibilities.
The methodology described here investigated the effects on biomass and plant
interactions of a natural plant community. When a new method is suggested,
many questions arise about its adequacy. For this community level test, one
must select which species to use, environmental conditions to provide,
parameters to measure and analyses to perform. All these factors need to be
considered in evaluating its performance.
The detail from regression analysis may be more than is needed from a
regulatory viewpoint, but it added understanding about competition and the
consequences of chemical addition. For example, Stellaria, the dominant
species in the control treatments in 1988 both by biomass and percent cover,
was a significant competitor to Poa only in the second sampling period (Figure
3). In contrast, it was a significant competitor in all four sampling periods
in the low atrazine treatment, even though it no longer was dominant (Figure
2). Erodium, with low biomass and cover in the atrazine experiment, was a
significant competitor in all treatments, in spite of its rarity (Table 3).
Using only data on species importance and the changes caused by treatment, the
biotic forces structuring the community are likely to be less understood and
potentially misinterpreted.
Single Species and Multi-Species Toxicity Tests
An underlying assumption in the development of this methodology was that
multi-species toxicity tests are a better indicator of the ecological
consequences of the release of organic chemicals than single species
laboratory tests. While this method is an increased level of sophistication
over current single species laboratory tests, it does not evaluate cross
trophic level interactions such as herbivory or disease that may have
significant impacts on plant community structure.
From this consideration of the literature and these experimental results, it
can be concluded that multi-species testing is a necessary addition to
environmental toxicology that will add realism and therefore credibility to
ecological risk assessments. In this multispecies experiment, some species
(Poa spp.) increased in response to higher levels of chemical. This result
would not have been determined from a single species test, nor would the
change in species interactions. The multi-species method used here takes
advantage of laboratory control by having homogeneous soil conditions,
emergence time and replication, while at the same time using naturally
occurring plants and climatic conditions. Numerous test systems (especially
aquatic) have already been developed to evaluate toxicants in more realistic
environments (Cairns and Mount, 1990; Odum, 1984; Hanson and Garton, 1982).
These test systems are in most cases as economical as single species tests,
although whole ecosystem manipulations probably are not. However, in certain
cases whole ecosystem testing may have some advantages (Perry and Troelstrup,
1988), especially in ecosystem restoration (Harris et al., 1990).
The increase in no-till agriculture has dramatically increased the use of
herbicides. The new generation of herbicides, the sulfonylureas, have low
119
-------
mammalian toxicities but extreme phytotoxicity, making them less of a human
and wildlife hazard and their application at low concentrations decreases the
risk of groundwater contamination. This lack of direct mammalian toxicity
along with the ability to genetically engineer herbicide resistant crops
increases the potential for widespread use and a consequent increase in
undesirable modification of non-target plants and communities. A slow
alteration of natural plant communities may be occurring now due to the
widespread release of chemicals either through direct phytotoxicity (Krahl-
Urban et al., 1988) or by the more subtle processes of evolution (Grant,
1972). The probability of unsuspected and probably undesirable change
increases with the continual release of organic compounds without valid
ecological testing.
CONCLUSIONS
1. A test method was developed for evaluating effects of anthropogenic
compounds on plant communities. This field method used species that grow
without cultivation in the geographic region of interest and are therefore
adapted to local environmental conditions. Environmental heterogeneity common
in most field studies was reduced by the use of raised beds and a uniformly
mixed soil containing seeds. Synchronous seed germination was enhanced by
initial watering and covering the beds. The optimal soil fertility remains
to be determined, but it is between the levels used in these experiments.
This method combines the characteristics of laboratory testing (simple,
economical, controlled and precise) and the realism of natural field testing,
providing a test with the benefits of both. The method may be appropriate for
investigating many processes of interest in plant ecotoxicology. Its use in
toxicology testing is enhanced by its small size, making it suitable for
transport and requiring little waste disposal.
2. All compounds tested, atrazine, 2,4-D and malathion, modified species
abundance in the model plant communities. Community productivity
significantly decreased when treated with atrazine and 2,4-D, but not with
malathion. There were four patterns of response exhibited by individual
species: biomass 1) decreased, 2) increased, 3) did not change or 4) decreased
at only the high concentration. Erodium biomass equally decreased at both
malathion concentrations. While some species were severely reduced in cover
and biomass, no species was completely eliminated from any community.
Communities were simplified and their dominance hierarchy was dramatically
altered when exposed to atrazine and 2,4-D and to a lesser extent with
malathion. The dominant species were replaced when treated with atrazine.
With 2,4-D, the dominant species was not significantly affected but
subdominant species were replaced.
3. Treatment with organic compounds altered interspecific competitive
relationships. All chemical treatments changed the identity of consistently
competitive species and the timing of important competitive interactions, when
species importance was measured by the cover surrounding target plants. Ten
120
-------
cm neighborhoods had more indicators of competitive interactions than 20 cm
when cover was the parameter measured. However, when biomass was used to
quantify influence of neighbors, 20 cm neighborhoods accounted for more
competitive interactions than did the 10 cm neighborhoods. Cover was a better
measure of competition than biomass, because it was easy to measure, indicated
more species interactions and was nondestructive, enabling competitive
interactions to be assessed throughout the growing season.
RESEARCH NEEDS
The approach discussed in this paper is preliminary and is not developed fully
enough to be implemented as a tier III test. In fact, little if any research
has been done on the impacts of organic chemicals on natural vegetation let
alone the development of a tier III test to investigate the potential for such
effects. However, the test described does have the potential to look at
questions not addressed in the current tier I and II test protocols. Current
field tests used for aquatic organisms and terrestrial wildlife are expensive
and have endpoints of questionable significance. The test described in this
paper has the advantage of being simple, inexpensive and produces endpoints of
ecological significance. Questions concerning reproductive biology and plant
interactions can easily be accommodated in this test. Some endpoints of
ecological interest are difficult and expensive to determine (i.e. competitive
indices, nutrient cycling) others such as biomass and cover are simple to
gather and interpret.
A field test that does not include ecological interactions does not give any
more information than currently exists in the present greenhouse tests except
how the plant performs under a different physical environment. Without
including biotic interactions into a field test, the test overlooks the
missing component from greenhouse studies and does not maximize the usefulness
of a field test. If a field test is to be developed the following questions
need to be addressed; 1) How does the disruption of one plant community relate
to other plant communities?; 2) What is an unacceptable amount of community
disruption?; 3) What ecologically significant endpoints should be measured?;
4) Is the natural weed flora used in this test the test community to use or
should other floras be used, such as ones that have a positive economic impact
or are representatives of rare and endangered species?; 5) What difference
does geographic location have on the results?; 6) Should the chemical
application rates be the same or lower than field application rates (i.e. is
the test simulating aerial drift?)?; 7) How should the chemical be applied?;
and 8) What formulation of the chemical should be used?. None of these
questions is easy or simple to answer, but without the research to answer
them, terrestrial plant field testing will repeat the expensive and
unrevealing results of aquatic and wildlife field tests.
A research program should be initiated that has as its objective the
development of a tier III field test for terrestrial plants. The first step
in this direction has occurred with the convening of this workshop. A second
gathering of experts is required with a much more focused agenda resulting in
practical set of guidelines for the culturing of test plants and/or plant
121
-------
communities. This would then be refined by experimentation at one location
followed by round robin testing throughout the country. The end product would
be a test that is simple and yet has ecological meaning.
REFERENCES
Aebisher, N.J. 1990. Assessing pesticide effects on non-target invertebrates
using long-term monitoring and times-series modelling. Functional Ecology.
4:369-373.
Bennett, J.P. and Runeckles, V.C. 1977. Effects of low levels of ozone on
plant competition. J. Appl. Ecol. 14:877-880.
Cairns, J., Jr. 1983. The case for simultaneous toxicity testing at
different levels of biological organization. Jji: W.E. Bishop, T.D. Cardwell,
and B.B. Heidolph (eds.), Aquatic Toxicology and Hazard Assessment: Sixth
Symposium. ASTM STP 802, ASTM, Philadelphia. 111-127 pp.
Cairns, J., Jr. and Mount, D.I. 1990. Aquatic toxicology. Environ. Sci.
Technol. 24:154-161.
Dickson, K.L., Duke, T. and Loewengart, G. 1985. A synopsis: workshop on
multispecies toxicity tests. 248-253 pp. in: J. Cairns, Jr. (ed.),
Multispecies Toxicity Testing. Pergamon Press, New York. 261 pp.
Fox, F.M. and Caldwell, M.M. 1978. Competitive interaction in plant
populations exposed to supplementary ultraviolet-B radiation. Oecologia.
36:173-190.
Grant, W.F. 1972. Pesticides - subtle promoters of evolution. Symp. Biol.
Hung. 12:43-50.
Hammons, A.S. (ed.). 1981. Methods for Ecological Toxicology: A Critical
Review of Laboratory Multispecies Tests. Ann Arbor Science Publishers, Inc.
Ann Arbor, MI.
Hanson, S.R. and Garton, R.R. 1982. Ability of standard toxicity tests to
predict the effects of the insecticide diflubenzuron on laboratory stream
communities. Can. J. Fish. Aquat. Sci. 39:1273-1288.
Harris, H.J., Regier, H.A. and Francis, G.R. 1990. Ecotoxicology and
ecosystem integrity: the Great Lakes examined. Environ. Sci. Technol.
24:598-603.
Hitchcock, C.L., Cronquist, A., Ownbey, M. and Thompson, J.W. 1969. Vascular
Plants of the Pacific Northwest. University of Washington Press, Seattle. 5
Vols.
122
-------
Krahl-Urban, B., Papke, H.E., Peters, K. and Schimansky, Chr. (compilers).
1988. Forest decline: cause-effect research in the United States of North
America and Federal Republic of Germany. Assessment Group of Biology, Ecology
and Energy of the Julich Nuclear Research Center. Julich, FDR. 137 pp.
Legge, A.H. 1980. Primary productivity, sulfur dioxide, and the forest
ecosystem: an overview of a case study. In: Proc. Symp. Effects of Air
Pollution on Mediterrean and Temperate Forest Ecosystems, Riverside, Calif.,
U.S. Department of Agriculture Forest Service, June 22-27. Gen. Tech. Rep.
PSW-43
Levin, S.A. and Kimball, K.D. (eds.). 1984. New perspectives in
ecotoxicology. Environmental Management. 8:375-442.
McCahon, C.P. and Pascoe, D. 1990. Episodic pollution: causes,
toxicological effects and ecological significance. Functional Ecology.
4:375-383.
Miller, P.R. 1984. Ozone effects in the San Bernadino National Forest. 161-
197 pp. In: D.D. Davis, A.A. Millen, and L. Dochinger (eds)., Air Pollution
and Productivity of the Forest: Proc. Symp., Washington, D.C. Izaak Walton
League of America, Arlington, VA. Oct 4-5, 1983.
Odum, E.P. 1984. The mesocosm. Bioscience. 34:558-562.
Perez, K.T., and Morrison, G.E. 1985. Environmental assessments from simple
test systems and a microcosm: comparisons of monetary costs. 89-95 pp. In:
J. Cairns, Jr. (ed.)., Multispecies Toxicity Testing. Pergamon Press, New
York. 261 pp.
Perry, J.A. and Troelstrup, Jr., H.H. 1988. Whole ecosystem manipulation: a
productive avenue for test system research? Environ. Tox. Chem. 7:941-951.
Tagatz, M.E. 1986. Some methods for measuring effects of toxicants on
laboratory-field-colonized esturarine benthic communities. In: J. Cairns,
Jr. (ed.), Community Toxicity Testing. ASTM STP 920. ASTM, Philadelphia.
18-29 pp.
Van Voris, P., Tolle, D.A., Arthur, M.F. and Chesson, J. 1985. Terrestrial
microcosms: applications, validation and cost-benefit analysis. 117-142 pp.
jjn: J. Cairns, Jr. (ed.), Multispecies Toxicity Testing. Pergamon Press, New
York. 261 pp.
Weldon, W.C., and Slauson, W.L. 1986. The intensity of competition versus
its importance: an overlooked distinction and some implications. Quart. Rev.
Biol. 61:23-43.
Winterhalder, K. 1984. Environmental degradation and rehabilitation in the
Sudbury area. Excerpt from: Laurentian University Review. Northern Ontario:
Environmental Perspectives. 16(2).
123
-------
DISCUSSION
The following discussions were transcribed from rather faulty tapes and edited
when necessary. The following is a synopsis of key thoughts, opinions,
concerns, and ideas put forth by participants during the final discussion
periods.
FRIDAY AFTERNOON FOLLOWING THE PRESENTATIONS
At the conclusion of the formal presentations, the participants were requested
to review the information which had been presented at the workshop and prepare
two lists of major points and issues. One list dealt with the effectiveness
of Subdivision J testing to protect natural plant communities, and the other
to protect nontarget agricultural crops. The combined lists were presented to
the members of the workshop by Frank Benenati and Frank Einhellig, and the
participants responded with the questions and comments.
The following is a brief summary of the key points of discussion, based on the
entire conversation which is included in Appendix G.
1. Endangered Species
An exchange of comments lead to the general consensus that expansion of the
list of test species to provide more diversity may also include plant species
more closely related to endangered species than those which are currently
used.
2. Status of Current Tier I and II tests.
There appeared to be general agreement that the experiences gained through the
use of the tests may be the basis for making some immediate changes whereas
other changes may require round-robin laboratory testing.
3. Tier III Testing
There was a great deal of disagreement as to why and how Tier III, field
testing should be conducted. Individuals opposing field testing raised the
fundamental questions of: 1) what is the purpose of the test? 2. What is the
endpoint to be measured? and 3) What constitutes unreasonable risk? Persons
arguing in favor of field testing suggested that the Tier III testing to
protect the environment could be patterned after the screening tests which
industry currently uses to develop new products. Individuals acquainted with
this type of screening tests were quick to point out that the screening is
done to determine EC80 results whereas EPA is interested in < EC25 responses.
It was pointed out that the screening tests for new products are not subject
to the GLP standards and the data generated is too variable to satisfy EPA
requirements, and to modify current field testing to accommodate EPA
requirements would be extremely costly. Furthermore, it was suggested that a
124
-------
series of increasingly more complex test systems microcosm, mesocosm, and
field tests may be more desirable than going directly to field tests.
Reference was made to difficulties encountered in large scale testing of
aquatic organisms and birds in hopes that plant field testing could be
initiated in a more expeditious manner without exorbitant expense.
4. Application Problems
It was suggested that the low efficiency of pesticide application is
fundamental to the question of drift damage. It was pointed out that a spray
group task force has been actively addressing a host of issues concerning
spray drift and will report their findings to EPA in the near future.
SATURDAY MORNING COMMENTS ON TENTATIVE
RECOMMENDATIONS
Friday evening a list of "tentative recommendations" was compiled by four
participants; Frank Benenati, Frank Einhellig, Joe Gorsuch, and Hilman Ratsch.
On Saturday morning the tentative recommendations were presented to all
members of the workshop and the following questions, comments, and requested
changes were made.
1) Drop tier I seed germination tests except for those cases where there is
reason to believe germination is a more sensitive indicator of effects.
(General discussion supported this recommendation.)
2) Develop criteria for acceptability of emergence and germination (if
conducted) response in a particular soil medium.
(It was explained that this recommendation called for cut-off times for
the emergence and seedling growth tests.)
3) Analytical determination of the chemical in use will only be required to
conclude a negative finding and terminate the test. (All current
analytical procedures will be required in tier II.)
(After a lengthy discussion, the general consensus was that the
recommendation should be included, but rewritten to clarify its
meaning.)
4) Identify and characterize the nature of the soil required for testing
procedures (perhaps start with OECD guidelines).
(There was general agreement that the recommendation would improve the
guidelines, but it should be modified to include a consideration of soil
pasteurization.)
125
-------
5) Provide better guidelines on experimental design and interpretation of
statistical analysis procedures on a species by species basis (add
reference to Appendix).
(No objection to the recommendation as stated.)
6) Evaluate and expand the current recommended list of test species with
the objective of enhancing the use of more diversity. The intent would
not be to require more species to be tested, but to include
representative genera and families that might be extrapolated to woody
species and/or endangered species, where appropriate.
(No objection to the recommendation as stated.)
Although not related to the recommendation, a lengthy discussion ensued
regarding the nature of the test chemical, formulated versus active ingredient
only. There was no agreement on this issue and it was decided by a hand vote
that a new recommendation would be added to emphasize the need for careful
review of this issue.
7) Range finding for tier II concentration determinations should be used to
determine if foliar or soil application results in more sensitive
responses. The most sensitive exposure route should be used in tier II
tests. If equally sensitive, tier II should use the most relevant route
of exposure.
(The group voted to remove this recommendation.) ,
(8) Optimize test conditions in tier II: i.e., temperature, photoperiod,
light, humidity, C02
(There was considerable disagreement concerning the merit of this
recommendation, but following a lengthy discussion, the group voted to
retain this recommendation in a revised form.)
II. Harmonize differences in test procedures between different regulatory
authorities or governing bodies (OECD, EEC, FIFRA, TSCA, FDA, CERCLA)
and work toward harmony with inter- national communities testing
requirements. Because of these inconsistencies, testing costs for
laboratories maintaining two or more programs are increased.
1) Establish what the inconsistencies are between agencies, e.g.,
a) EC-25 for effect under FIFRA, compared to 1% tolerance for
FDA.
b) Number of species required, number of plants per test, and
number of replicates per test.
c) Nutrient addition (FDA) compared to no nutrient addition
required by FIFA.
d) Photoperiod requirements under FDA, but not specified in
some.
e) Watering regiments that should be optimized.
126
-------
IV.
f) Endpoints that are required: FIFRA does not require shoot
heights, root length, and shoot and root weights, whereas
FDA does.
...ETC.
2) Call for joint efforts to arrive at a consensus on the testing
procedures.
(A short discussion was followed by endorsement of this recommendation.)
Research is needed to improve the efficiency and in some cases, the
validity of testing protocols. Special case needs include: (in no
priorital order)
1) Establish the feasibility of using tissue culture methods as
options for tier I and II work. Tier I might include multiple
exposure concentrations, but testing would be comparable to range
finding in tier II without GLP/analytical determinations. A
special focus should be to use tissue culture as a surrogate for
tests on non-target woody species of concern.
2) Develop efficient life cycle bioassays, both for representative
dicot and monocot species. This should include methods of
chemical applications in these bioassays.
3) Research the possibility and procedures for using mesocosms to
evaluate chemical effects on plant communities.
a) Agroecosystem models versus natural community models.
b) What parameters should be indicators of effects?
c) When should such studies be required?
d) Evaluate the feasibility of using soil core and terrestrial
microcosm chambers and other "off the shelf" technologies.
4) Study the possibilities of using remote sensing procedures and
other technologies to monitor chemical effects in field tests,
including possible effects on nontarget plants and plant
communities.
Item 3 under IV was a focus of discussion. The question was raised, "What is
biologically significant effect?" There was disagreement on what would be a
meaningful endpoint to measure in a field experiment, and what would
constitute an unacceptable effect. There appeared to be mutual agreement that
additional research needs to be done to develop meaningful field study
protocols.
5) Research and a database on culture techniques are needed for plant
species identified for tier III evaluation. These include forest
and understory species and wetland species.
(General agreement that the recommendation was appropriate.)
127
-------
6) There needs to be research done to examine extrapolation of
toxicological data from inter- and intra-surrogates.
III. Design and implementation of field experiments should be clarified.
A. Current tier III requirements appear to be more efficiently accomplished
if divided into two phases, or considered as two separate tiers.
1. Develop protocols or consensus methodology for small plot tests
with regard to critical species, soil type, and other field
variables.
2. Preliminary field tests are carried out to identify sensitive
variables noted above.
(Concern was expressed again that endpoints must be established before
field tests start.)
B. The nature of more extensive tests, conceived as tier IV, can only be
estimated after doing part A. This includes the regional conditions
needed for the studies, species and species assemblages to be included
in the tests, and range of treatment levels expected.
1. Establish the minimum information necessary for conducting a valid
risk assessment.
2. Research needs to be conducted to determine under what conditions
test data are adequate without tier IV information.
(The recommendation was accepted as written.)
In the summary, neither the government nor private sector alone has the
resources or expertise to accomplish the objectives set forth in these
recommendations and goals. A group effort will be required and this must
include mechanisms for the sharing of data and improved communications between
all involved.
There was a general feeling that another meeting was desirable, and it was
agreed that such a statement should be included in the recommendations.
128
-------
FINAL RECOMMENDATIONS
I. Harmonize differences in test procedures between different regulatory
authorities or governing bodies (OECD, EEC, FIFRA, TSCA, FDA, CERCLA)
and work toward adopting universal standard tests for use throughout the
international community. Because of these inconsistencies, testing
costs for laboratories maintaining two or more programs are increased.
1) Establish what inconsistencies exist between agency test
guidelines, e.g.,
a) EC25 for effect under FIFRA, compared to 1-5% tolerance for
FDA.
b) Number of species required, number of plants per test, and
number of replicates per test.
c) Nutrient addition (FDA) compared to no nutrient addition
required by FIFRA.
d) Photoperiod requirements under FDA, but not specified in
some.
e) Watering regiments that should be optimized.
f) Endpoints that are required: FIFRA does not require shoot
heights, root length, and shoot and root weights, whereas
FDA does.
2) Call for joint efforts to arrive at a consensus on testing
procedures.
II. Revisions in tier I and tier II testing are needed to expedite the
procedure and to obtain the most meaningful data. The overall goal is
to reduce the cost, yet maintain the sensitivity of the screening tests.
A priority listing of the suggested revisions includes:
1) Drop tier I seed germination tests except for those cases where
there is reason to believe germination is a more sensitive
indicator of effects.
2) Develop evaluation criteria for seed germination and emergence
response in a defined soil type (see recommendation 4 in this
section). Specific definitions are needed for what constitutes a
germinated seed, an emerged seed, and the length of test-time
needed to conclude a negative result.
3) Simplify and reduce the cost of the tier I screening test by
eliminating the analytical determination of chemical test
solutions (a GLP requirement, Appendix C). The exception will be
when a negative result (no plant response) occurs, then analysis
should be conducted to prove that the chemical was administered at
the stated concentration. No recommendation is made to change the
current tier II requirements for chemical analysis.
129
-------
4) Identify and characterize the nature of the soil required for
testing procedures (perhaps start with OECD guidelines). This
should include a consideration of organic content and soil
pasteurization.
5) Provide better statistical guidelines addressing: 1) experimental
design (number of replicates, etc.), 2) statistical procedures,
and 3) interpretation of statistical results.
6) Evaluate and expand the current recommended list of test species
with the objective of enhancing the use of more diversity. The
intent would not be to require more species to be tested, but to
include representative genera and families that might be
extrapolated to woody species and/or endangered species, where
appropriate.
7) Review the current guideline requirements regarding the nature of
the chemical test product. Address the issue of how closely the
test-product must resemble the end-use formulated product which
usually includes surfactant, stickers, etc.
8 Provide guidelines for minimum test conditions in tier I and II,
i.e., temperature, photoperiod, light, and humidity. The
guidelines must be sufficiently flexible to accommodate the
different physiological needs of various test species.
III. Design and implementation of field experiments should be clarified and
developed in parallel with accompanying research (section IV-3).
1) Current tier III requirements appear to be more efficiently
accomplished if divided into two phases, or considered as two
separate tiers.
a) Develop protocols or consensus methodology for small plot
tests with regard to critical species, soil type, and other
field variables.
b) Preliminary field tests are carried out to identify
sensitive variables noted above.
2) The nature of more extensive tests, conceived as tier IV, can only
be determined after doing part (1, a and b). This includes the
regional conditions needed for the studies, species and species
assemblages to be included in the tests, and range of treatment
levels expected.
a) Establish the minimum information necessary for conducting a
valid risk assessment.
b) Research needs to be conducted to determine under what
conditions test data are adequate without tier IV
information (Section IV).
130
-------
IV. Research is needed to improve the efficiency, and in some cases the
validity of testing protocols. Special case needs include: (in no
priorital order)
1) Establish the feasibility of using tissue culture methods as
options for tier I and II testing. Tier I might include several
different exposure concentrations, comparable to range-finding
tests in tier II, but without GLP/analytical determinations. A
special focus should be to use tissue cultures to test slow
growing woody perennials and endangered species.
2) Develop efficient life-cycle bioassays, both for representative
dicot and monocot species. Methods for the application of
chemicals should be included in these bioassays.
3) Research the possibility and procedures for using mesocosms and
field studies to evaluate chemical effects on plant communities.
a) Understanding agroecosystem models versus natural community
models.
b) What parameters should be evaluated to determine the extent
of effects?
c) When should such studies be required?
d) Evaluate the feasibility of using soil-core and terrestrial
microcosm chambers and other "off-the-shelf" technologies.
4) Study the possibilities of using new technologies, for example,
thermal sensing procedures, to monitor chemical effects in field
tests to predict and identify possible effects on nontarget plants
and plant communities.
5) Research is needed to provide optimal culture techniques for plant
species identified for testing. Species include forest species
(both canopy and understory) and wetland species.
6) Research is needed on the validity and accuracy of intraspecies
and interspecies extrapolation of toxicological data from
surrogate test species to potential nontarget plants.
V. National Agricultural Chemists Association workshop mini-workshop.
There appears to be a clear need for setting up a subsequent workshop
group or dialogue.
In summary, neither the government nor private sector alone has the resources
or expertise to accomplish the objectives set forth in these recommendations
and goals. A group effort will be required and this must include mechanisms
for the sharing of data and improved communications between all involved.
131
-------
APPENDIX A
EPA-54Q/9-82-020
October 1982
Pesticide Assessment Guidelines
Subdivision J
Hazard Evaluation:
IMontarget Plants
Prepared by
Robert W. Hoist, Ph.D.
and
Thomas C. Ellwanger, Ph.D.
Office of Pesticide Programs
Guidelines Coordinator
Robert K. Hitch
Hazard Evaluation Division
Office of Pesticide Programs
U.S. Environmental Protection Agency
Office of Pesticides and Toxic Substances
Washington, D.C. 20460
-------
Subdivision J - Hazard Evaluation: Nontarget Plants
Table of Contents
DISCUSSION
I. Introduction 1
II. Organization 2
III. Major Issues 3
GUIDELINES
Series 120: GENERAL
§ 120-1 Overview 14
§ 120-2 Definitions 17
§ 120-3 Basic test standards 19
§ 120-4 General evaluation and reporting requirements 22
Series 121: TARGET AREA TESTING
§ 121-1 Target area phytotoxicity testing 28
Series 122: TIER 1 OF NONTARGET AREA TESTING
§ 122-1 Seed germination/seedling emergence and
vegetative vigor (Tier 1) 38
§ 122-2 Growth and reproduction of aquatic plants
(Tier 1) 40
§ 122-30 Acceptable methods and references 42
Series 123: TIER 2 OF NONTARGET AREA TESTING
§ 123-1 Seed germination/seedling emergence and
vegetative vigor (Tier 2) 48
§ 123-2 Growth and reproduction of aquatic plants
(Tier 2) 49
Series 124: TIER 3 OF NONTARGET AREA TESTING
§ 124-1 Terrestrial field testing 51
§ 124-2 Aquatic field testing 53
-------
1
SUBDIVISION J ~ HAZARD EVALUATION: NONTARGET PLANTS
DISCUSSION
I. Introduction
The performance requirements and testing and reporting proce-
dures of pesticide chemical, environmental, and toxicity properties
to support the registration of each pesticide under the Federal
Insecticide, Fungicide, and Rodenticide Act (FIFRA) are provided
in two document series. The first is Volume 40 Part 158 of the
Code of Federal Regulations (CFR) which specifies the kind.of data
and information that must be submitted. Section 158.150 specifies
the performance requirements for phytotoxicity (plant protection)
testing. The Agency intends to promulgate 40 CFR Part 158 as a
final rule during 1983.
The second series of documents [Guideline Subdivisions, such as
the present one, published by the National Technical Information
Service (NTIS)] provide the test criteria and reporting procedures
for the various studies. This subdivision, entitled Subdivision J -
Hazard Evaluation: Nontarget Plants, provides detailed information
relating to the phytotoxicity (plant protection) data requirements
listed in 40 CFR Part 158, §158.150. Subdivision J describes the
conditions under which the phytotoxicity data requirements are
applicable, the standards and protocols for acceptable testing,
stated with as much specificity as the current scientific disci-
plines allow, and reporting procedures. Also provided in this
subdivision are circumstances under which an applicant should
consult with the Agency before initiating a study.
The plant protection test protocols and reporting procedures
are provided to the registrants and general public for information
purposes. Results of the phytotoxicity studies found in this Sub-
division will be reported to the Agency on a limited basis. See
paragraphs D.2 (page 7) and E.I (page 8) of the discussion and
§ 120-1(d) and (e) of the guidelines (page 13) which provide state-
ments as to the requirements to submit data for the various studies
of this Subdivision.
The phytotoxicity data submitted along with data on environ-
mental fate and efficacy are used to assess the potential hazard
of pesticides on nontarget plants, both terrestrial and aquatic.
Nontarget plants include crops, ornamentals, and others that are
intentionally sprayed or otherwise treated, and plants outside the
area of intended application (which would include food and cover
vegetation for animals, food, fiber, fuel, and ornamental plants
for man, and endangered and threatened plants).
-------
A purpose common to all tests is to provide data which will be
used to determine the need for (and support the wording for) pre-
cautionary labeling or other statements to minimize the potential
adverse effects to nontarget plants. Generally, the registrant will
provide adequate precautionary labeling with respect to nontarget
plants such as crops, ornamentals, and the like. However, there
may be situations where the Agency will have to develop additional
precautionary labeling. For example, the spraying of herbicides
may not be permitted in the vicinity of critical habitats of
endangered or threatened plants listed by the United States Depart-
ment of Interior.
IX. Organization
The discussion continues with presentation of the major issues
addressed by commenters with the publication of the proposed guide-
lines - Subpart J: Hazard Evaluation: Nontarget Plants and Micro-
organisms, to FIFRA in the Federal Register (45 FR 72948-72978,
November 3, 1980).
The Guidelines portion of this subdivision (p. 11) is divided
into three major parts: General (Series 120); Target area phytotox-
icity (Section 121-1); and nontarget area phytotoxicity (Series 122,
123 and 124). The general section series deals with the overview
and scope of the subdivision including a general discussion of phyto-
toxicity data (§ 120-1), the definitions of specific words used in
the subdivision (§ 120-2), basic standards for testing (§ 120-3),
and the general evaluation and reporting procedures (§ 120-4).
Section 121-1 deals with target area phytotoxicity testing,
which is used to evaluate pesticide toxicity to those plants that
would experience intentional application.
The next three section series (Series 122, 123, and 124) com-
prise the tier testing sequences (Tiers 1, 2, and 3, respectively)
employed to study and report on pesticide toxicity to nontarget
area plants. The effects of the pesticides are determined through
a series of tests as dictated by specific requirements of each
test and tier. The tests are designed to provide guidance for
gathering pesticide effects information on terrestrial and aquatic
plant growth and development. The influences of geographical, sea-
sonal, and species variation are also addressed.
Also contained in a section in Series 122 are detailed proto-
cols for some of the studies found in Subdivision J. At the end
of each protocol are selected references to acceptable methods
that may be used to develop pesticide phytotoxicity data.
-------
Each test section contains an opening paragraph restating the
circumstances and for what products, as found in 40 CFR Part 156,
the data are required. The test sections also contain specific
test criteria, procedures and reporting formats which, in addition
to the respective general testing information, apply to the accom-
plishment of the studies.
The execution of studies in the higher tiers depends on the
results of studies in the lower tiers. The tier system is intended
to reduce repetitive consultation between the registrant and the
Agency about the need for tests of greater complexity. As a result,
the time required to develop data for registration of a pesticide
should be reduced substantially.
III. MAJOR ISSUES
The Agency received comments from numerous persons or groups
regarding the 1980 proposed guidelines and the 1982 draft of this
document. In many cases the commenters provided information on
the applicability and the scientific merit of the various tests.
In response to these public comments, the Agency has modified or
clarified all sections and many paragraphs of these guidelines.
Only the more significant and controversial issues submitted by
the public are discussed in the following pages. Many recommen-
dations were adopted by the Agency which do not warrant discussion
here.
A. General Information.
Several commenters have expressed concern that the Agency,
through proposed Sub part. J and the other proposed subparts, is
trying to investigate whether all pesticides exhibit subtle effects
on the environment. The Agency is required by FIFRA to ascertain
whether a pesticide "...will perform its intended function with-
out unreasonable adverse effects on the environment..." [FIFRA
sec. 3(c)(5)]. The effects may, indeed, be unreasonable and
unacceptable, even if considered subtle by some observers. The
purpose of this and other subdivisions is to provide guidance in
the submission of data and other information. From this combina-
tion of information, an overall environmental risk assessment
concerning the exposure and effects of the pesticide can be made.
Included in this evaluation is a determination as to the possible
effects on endangered and threatened plant species.
The preamble to the November 3, 1980 proposed Subpart J guide-
lines (FR Vol. 45, page 72949) provided examples as to the possible
uses of the information. Also, Subdivision H, Labeling-for Pesti-
cides and Devices, provides the guidance concerning various types
-------
of label limitations, precautionary statements, or restrictions
relating to phytotoxicity.
B. Substitution of Test Data.
From the comments of several groups, it was obvious that the
Agency did not make it entirely clear about the possibility of
substituting existing test data for data produced during the tier
tests (§§ 122, 123, and 124). Zt is not the intent of the Agency to
request completely new or redundant testing where existing test data
would satisfactorily answer the question as to a pesticide's phyto-
toxic properties.
The substitution of test data applies primarily to the testing
of herbicides. The Agency realizes that registrants who desire to
market herbicides and other pesticides have tested their products
extensively for phytotoxic effects. The information to be reported
for Tiers 1, 2 and 3 have generally been generated during these
tests. Therefore, to satisfy the requirements for phytotoxicity data
as found in 40 CFR Part 158, the registrant would simply have to make
the data from these investigative tests presentable and provide them
to the Agency. This will alleviate the need to "skip to Tier 3" for
herbicides or generate new data at great expense and time. ~~"~
To help in this matter, the paragraph on substitution
t§ 163.120-5(c) in proposed Subpart J] was reworded and moved to a
more prominent, suitable location [§ 120-1(e)(4)] in the current
Subdivision J. Also, the beginning of each tier test section
contains a cross-reference to this substitution paragraph.
C. Test Substance.
1. Testing of the same pesticide lot. Several comznenters noted
that the use of the same lot of pesticide throughout all testing is
impractical. This requirement has been modified so that the same lot
is desired only in laboratory studies.
2. Data requirements for manufacturing-use products. Prom
comments to other subdivisions of the FIFRA guidelines, the Agency
has concluded that extending the data requirements to such manufac-
turing-use products is appropriate. The Agency was influenced by
the views of commenters on this issue who generally favored a data
submission requirement which makes the basic manufacturer of an
active ingredient responsible for providing most of the phyto-
toxicity data*
Therefore, a section of 40 CFR Part 158, entitled "Formulators'
Exemption" (§ 158.50), requires a registrant of a manufacturing-use
product to submit (or cite) any data pertaining to the safety of an
-------
active ingredient in its product if the same data are required to
support the registration of an end-use product that could legally
be produced from the registrant's manufacturing-use products.
(An immediate end-use product is a pesticide product bearing label
directions for immediate end-use as a pesticide.) Section 158.50
also provides that such data must be submitted by an applicant for
registration of the end-use product, except that the producer of
the end-use product will generally not have to submit or cite data
pertaining uses to formulate the end-use product. This decision
reflects the Agency's expectation that manufacturing-use product
registrants will be the major source of registration data, and
that end-use product formulatora will, in most cases, need to
supply much less data. This decision is consistent with the pro-
visions of, and Congressional intent behind, sec. 3(c)(2)(D), of
FIFRA, which provides that:
No applicant for registration of a pesticide who
proposes to purchase a registered pesticide from
another producer in order to formulate such pur-
chased pesticide into an end-use product shall be
required to -
(i) submit or cite data pertaining to the safety
of such purchased product; or
(ii) offer to pay reasonable compensation other-
wise required by [§ 3(c)(1)(D) of FIFRA] for use of any
such data.
Implicit in sec 3.(c)(2)(D) is Congress1 expectation that it
would be the registrant of the manufacturing-use product who would
provide significant amounts of data pertaining to the safety of its
product. (See, e.g., Sen. Rep. No. 334, 95th Cong., 1st Sess.,
pp. 8-9.)
Moreover, if data requirements were imposed solely on regis-
trants of end-use products, sec. 3(c)(2)(D) might be read to prevent
the Agency from obtaining data on the grounds that the data pertain
to the safety of a purchased product.
3. Testing a representative end-use product. The Agency seeks
to avoid imposing a burden of duplicative testing on applicants for
registration. Therefore, where 40 CFR Part 158 specifies that the
test substance shall be a representative end-use product, testing
may be performed using the formulation in question (end-use product
being registered) or similar, yet representative, end-use product.
It is not necessary to repeat the test using other similar products.
A representative end-use product is defined in § 120-2(1) as:
A pesticide product that is representative of a major
formulation category (e.g., emulsifiable concentrate,
granular product, wettable powder) and pesticide group
(e.g., herbicide, fungicide, insecticide, etc.) and
-------
contains the active ingredient of the applicant's
product.
The use of a typical end-use product in plant protection test-
ing is needed for tests which determine the extent of phytotoxicity
under actual use conditions. In Subdivision J, all tests in § 121
(Target Area Phytotoxicity) and in § 124 (Nontarget Area Plant
Field Studies) are in this category. Moveover, since manufacturing-
use products may be formulated into end-use products belonging to
several different formulation categories/ testing is required with
a typical end-use product from each formulation category. Accord-
ingly, the test substance section of these tests now contains a
provision which states:
The test substance shall be the end-use product or a
representative end-use product from the same major
formulation category for that general use pattern.
Examples of major formulation categories are: wet-
table powders, emulsifiable concentrates, and granu-
lars. (If the manufacturing-use product is usually
formulated into end-use products comprising two or
more major formulation categories, a separate study
must be performed with a typical end-use product
for each category.)
It should be noted that the submission of data using the
specific end-use product in question is recommended as it would
better describe any phytotoxicity associated with that chemical.
4. Technical grade vs. formulated product. Comments were
received on both sides of the issue as to which test substance,
technical grade or formulated product, to test at the Tier 1 and 2
levels. The Agency has decided to leave these test substances as
they are, i.e., technical chemical to be used at Tiers 1 and 2 and
the representative end-use product to be used in Tier 3. The use
of the technical chemical in Tiers 1 and 2 follows the intent of
the Agency to use existing information to satisfy the data require-
ments of these tiers. A significant amount of initial screening
information is generated using the technical chemical.
In connection with testing of technical material at the Tier 1
and 2 level, there were several comments about the requirement to
make special formulations for these tests. Special formulations
are neither required or desired. The only requirement is the use
of a suitable solvent, if needed, at a level that is not phyto-
toxic to dissolve the material in water or other suitable carrier.
D. Target Area Phytotoxicity Testing.
1. Phytotoxicity and efficacy testing. Several commenters
noted a confusion between those phytotoxicity tests found in proposed
-------
Subpart J and those normally performed in relation to and simul-
taneously with product performance (or efficacy) testing. All
phytotoxicity testing and reporting procedures were removed from
Product Performance (1975 proposal; currently called Subdivision
G) not to imply separate criteria and procedures, but rather to
separate the subjects of phytotoxicity and efficacy. Product
performance testing and target area phytotoxicity testing are
ordinarily and may continue to be conducted simultaneously.
2. Waiver of target area phytotoxicity. The Agency has
determined that target area phytotoxicity data does not need to be
submitted because the registrants are generally willing to accept
the overall responsibility of the product respect to efficacy and
phytotoxicity [FIFRA Sec. 3 (c)(5)]. These data guidelines are
provided to the registrants for those instances where data may be
needed.
3. Weed-free control plots. The weed-free or otherwise "pest-
free" control plots of proposed §§ 120-2(i) and 121-1(c)(1)(iv) were
the subject of several comments. Originally the proposed guidelines
required the maintenance of weed-free and pest-free plots. The
conmenters stated that this is very difficult, impractical, and at
times may be even detrimental to the crops. Therefore, the defini-
tion of "pest-free" has been changed to only recommend control of
pests including weeds in order that healthy desirable plants are
available for testing. For example, the control process of weeds
may be by hand-weeding and/or by use of a commonly-used reference
chemical products(s).
4. Testing not prohibited by the label. As stated in sec.
(2)(ee) of FIFRA, a pesticide may be applied "...employing any
method of application not prohibited by the labeling..." In the
proposed Subpart J guidelines [proposed § 163.121-1(c)(3)], all
equipment types not prohibited by the label would have been eval-
uated with respect to pesticide application and movement in the
environment. Several commenters have stated that testing all
applicable methods not prohibited by the label is impractical and
that either only some of those specified on the label or the "worst
case" situations should be evaluated. The Agency agrees that such
extensive testing is impractical and would provide little additional
information as to the phytotoxic nature of the pesticide. Testing
of the "worst case" is discouraged because of the complicated
determination of that situation. Therefore, use of some methods
of application which are found on the label need only be tested.
If a "worst case" application method can be readily determined
prior to testing, then testing may be limited to that case. Sup-
port for the use of that method should be furnished to the Agency.
5. TanX mixtures and serial applications. Several commenters
stated that the tank mixture (antagonism and synergism) and serial
applications tests were excessive [§ 121-1(b)(5) and (6)]. The
-------
8
Agency in Pesticide Programs PR Notice 82-1 of January 1982 has
eliminated, in most cases, the requirement to submit residue and
compatibility data for tank mixes. In the PR Notice, it was noted
that registrants normally test for these conditions and submit
label statements that allow only certain tank mixtures or serial
applications•
Therefore, the Agency will not require antagonism or synergism
studies on desirable target area plants. There may be times when the
Agency will desire this information to assess phytotoxicity problems
associated with antagonism and synergism.
•
6. Data on fruit and nut trees and pastures and rangelands.
Data on the yields of fruit and nut trees and on population shifts
in pastures and rangelands were addressed as being excessive and
unattainable by several commenters. It was noted that the yields of
fruit and nut trees are variable from year to year and that the data
required in § 121-1(c)(2)(iii) would be meaningless. The Agency has
now corrected this by asking for the comparison of yields and growth
of treated trees to simultaneous controls not to just preapplication
measurements of the treated trees.
The reporting of general population shifts in pastures and
rangeiands [121-1(c)(2)(ii)] was included to determine if the de-
sired species are replacing those plant species being controlled
and if other undesirable species were in turn replacing the desir-
able species. This is a desirable ecological research parameter
but is not necessary in the evaluation of pesticidal phytotoxicity
in the registering of pesticides. Therefore, the requirement has
been removed.
7. Subsequent planting (rotational crops). Commenters noted
that the evaluation of subsequent planting was excessive and required
in another section. The other study, found in Subdivision N [§ 165-
2] , is designed to evaluate soil residues and the uptake by edible
crops or forage of persistent pesticides. The studies in Subdivi-
sion J [§ 121-(c)(6)] are used to evaluate the phytotoxic effects
of persistent pesticides, primarily herbicides. Therefore, this
test will be regained in this subdivision.
E. Nontarget Area Phytotoxieity Testing
1. Data requirements for nontarget area phytotoxicity tests.
The Agency in the public draft of this NTIS document proposed that
the phytotoxicity testing be required on a case-by-case basis. A
number of commenters requested that the requirements for nontarget
area phytotoxicity be deleted in their entirety because it was felt
that the information submitted could be classified as "nice to know"
rather than as necessary to know for a registration decision.
-------
The Agency is retaining Subdivision J nontarget area phytotoxi-
city tests for those situations where such information is desired.
The Subdivision provides a set of standards and reporting formats
for the tests and data when they are requested. Several examples
when the data may be required are: (1) hazards posed to endangered
or threatened plants listed by the United States Department of
Interior, Fish and Wildlife Service; (2) initiation of a rebuttable
presumption against registration (RPAR) where a phytotoxicity
problem may exist; and (3) where a specific phytotoxicity problem
arises when general open literature data are not available.
The Agency will inform the registrant of the chemical in ques-
tion concerning the phytotoxicity problem and the specific data
required to address the problem.
2. Terrestrial species selection. In the proposed Tiers 1 and
2, seed germination/seedling emergence and vegetative vigor tests
(proposed §§ 163.122-1 and 163.123-1), ten specific kinds of plants
were to be tested. This made the guidelines somewhat inflexible and
did not readily permit the use of much screening test data already
generated by companies. The selection now states that soybeans,
corn, and a dicot root crop are to be tested, and that seven other
test species are to be a balance of monocots and dicots. Corn and
soybean were retained due to their economic significance and the
quantity of pesticide research performed using these species. By
increasing this flexibility of species selection, tests that are
normally performed by the developer/registrant during screening and
initial field testing may often be used. This change will result
in a significant cost reduction for this test.
3. Aquatic species selection. Several commenters noted that
inclusion of five aquatic species at the Tier 1 and 2 level can lead
to expensive and unnecessary testing. They suggested that only one
species, probably Selenastrum capricornutum, be tested at the Tier
1 level.
After careful consideration, the Agency decided that this
species selection was indeed unnecessary and that the selection
could be based on use pattern. Selenastrum will be tested for all
terrestrial or aquatic outdoor uses. If an outdoor aquatic use
pattern is anticipated, the other four aquatic species would also
be used.
The aquatic species selection was based on those species that
have been extensively tested and for which the growth parameters
have been strictly determined and specific strains are readily
available. For these reasons Lemna gibba G3 is chosen over Lemna
minor and Selenastrum capricomutum over Chlorella vulgaris. The
diatoms are used because they have been shown to be very sensitive
to water pollutants. Anabaena flos-aguae is chosen as a represen-
tative of a group of plants that can fix atmospheric nitrogen.
-------
10.
The overall selection was made to obtain a broad representation
of aquatic plants and provide some insight into variations of effects
on aquatic plants• The increased diversity of plant types required
in Tier 3 (dicots, monocots, ferns, etc.) addresses the fact that
plants other than algae inhabit aquatic areas. Again this test is
to note the variation of effects (i.e./ tolerance or resistance)
to the pesticide.
4. Dosages or application levels. Many commenters to the
proposed Subpart J guidelines stated that three times the label rate
was an unrealistic quantity to be assessed for nontarget area
phytotoxicity. This statement was ba?. •. on information from actual
uses and exposures. In response to these comments, the maximum
dosage or application level was set at the maximum label rate.
Again comments were received that this rate was excessive and that
the rate should be based on environmental exposure.
It was not the intention of the Agency to perform these tests
after environmental exposures had been determined or modeled. If the
registrant, however, decides to perform these tier tests after deter-
mination of the environmental exposure, then a rate equal to at least
three times the exposure as found in the adjacent nontarget area may
be used. It must be remembered that the adjacent nontarget area can
be the adjacent desirable plant of another species 0.1 meter or 100
meters distant. Therefore, the use of this exposure level must be
supported with appropriate data.
On the other hand the use of the maximum label or environment
exposure rate does not preclude the voluntary testing and submission
of phytotoxicity data where the tests were performed using higher
rates. It is noted that dosages used during manufacturing screening
tests would have a greater tendency to exceed this required dosage
or application level, and would thereby increase the probability of
acceptance of these screening tests.
F. Plant Mutagenicity Testing.
Since proposing the concept of a plant mutagenicity testing
scheme in Subpart J, many registrants and other researchers have
expressed concern that these tests would not provide meaningful data,
Also, no incidence of plant mutagenicity has been substantiated for
target area crops or nontarget area plants.
Several conmenters suggested that this set of tests undergo an
extensive series of evaluations before this type of testing be in-
cluded in any finalized ruling. Also, commenters and others pro-
vided references which question the validity of using plant muta-
genicity studies to evaluate human mutagenicity.
-------
11
Upon evaluation of these comments, the Agency has decided to
withdraw the requirement for the plant mutagenicity studies until
extensive testing can be performed to show the more substantial
usefulness for this requirement.
G. Tier 3 Field Studies.
Several commenters noted confusion in the requirements of and
the differences between the Tier 3 aquatic and terrestrial field
studies and the Tier 4 geographical and seasonal field tests. This
confusion was generated by the tier progression statements where one
progressed from Tier 2 to either Tier 3 or 4, depending upon a com-
plex set of progression requirements.
To eliminate this confusion, all field studies were combined
at the Tier 3 level with respect to either terrestrial field or
aquatic field testing. Geographical or seasonal considerations
are included in the Tier 3 tests. There is no longer any Tier 4
testing.
B. Nitrogen Fixation Studies.
All testing of microorganisms was removed from Subdivision J,
except for testing of algae. Therefore, testing of the nitrogen
fixation potential as affected by pesticides was removed from
Subdivision J. This study will be considered for inclusion in pro-
posed Subdivision S dealing with pesticide-microorganism effects.
Comments received will be used in the development of these require-
ments when this subdivision is prepared.
I. Sorptlon Study.
The requirement for a sorption study as proposed in Subpart J
was based on a theory of possible mode of exposure of aquatic
vegetation to pesticides. These pesticides would be carried by
runoff water from adjacent agronomic fields or sites of pesticide
application. However, recent studies have shown that this was not
the probable mode of exposure. Rather the exposure has been attri-
buted to a concentrated "slick" of pesticide floating on the water.
The Agency has since determined that it can determine either
of these exposures from existing or provided data. Therefore,
this section was deleted in its entirety.
-------
12
J. Spray Drift Studies.
Spray drift can affect not only nontarget plants but also
nontarget animals and humans. Because of the broad spectrum of
adverse effects from spray drift, the Agency has removed this sec-
tion series from Subdivision J and will include it in proposed
Subdivision R on Pesticide Aerial Drift Evaluation. Comments
received on spray drift will be addressed in this new subdivision.
K. Tier Progression.
Commenters in general agreed that the EC10 value for the Tier
1 and 2 progression criteria is too stringent because the variation
of plant growth and development response within a treatment of a
study will normally exceed 10 percent. Through testing at EPA
laboratories and evaluation of testing submitted to EPA, the Agency
has determined that the proposed tier progression criteria for
terrestrial plant studies were excessive and at times not definable.
For example/ in the case of providing height and weight on all plants
tested, the variation within any one group would preclude an analysis
of the possible effects. Therefore, the criteria have been revised
to the simple criterion of a detrimental effect of 25 percent or more
(EC2S) on one or more plant species employing the maxiinv™ label rate.
If, upon statistical analysis of the results, it has been
determined that the variation or error within the species is signifi-
cant enough to overshadow a detrimental effect of 25 percent, then
the tests must be repeated. If the population size was sufficiently
large to not warrant retesting, then an explanation as to why addi-
tional tests were not performed must be provided.
Commenters also stated that the EC50 value for the aquatic plant
testing was not realistic but rather an EC90 or EC95 is more appro-
priate. The Agency, however, has decided not to change this pro-
gression criteria for the following four reasons:
- Good general agreement does not exist among researchers
on the value that would best describe a possible "worst
case" or one from which the population can readily
recover.
- The EC50 value is used as a "trigger" to require studies
and would be more indicative of normal situations. Also,
EC50 values have been commonly obtained for many aquatic
plants, whereas the EC90 or EC95 values are not well
based, statistically.
- The Agency has reduced the number of species at the Tier 1
and 2 levels, basing their inclusion on use pattern.
-------
13
- The mjtyimmn dose level has been reduced to the maximum label
rate or to 3 times the maximum expected environmental exposure.
L. Statistical Analysis
Several commenters stated that for the results to be statis-
tically significant more replicates and/or a greater population
size would be required. A basic part of scientific analyses is to
have sufficiently large populations in order that the results be
meaningful. The Agency is maX-r.c the selection of population size
flexible as each study would re. re a different number of indivi-
duals. It should be noted that e • 'v. species has a different seed
germination and survivability rate >.hich has a direct bearing on
the statistical significance of the results. The Agency encourages
the use of the largest possible populations for each of the tests
in order to approach the 90 to 95% level of confidence with a
significance level of less than 0.10. The following references
are provided concerning sample size selection.
Casagrande, J.T., Pike, M.C., and Smith, P.G. 1978. An improved
approximate formula for calculating sample sizes for comparing
two binomial distributions. . ,.ome tries 34:483-486.
Fleiss, J.L. 1973. Statistical Methods for Rates and Propor-
tions. John Wiley and Sons, Inc. New York.
Snedecor, G.W., and Cochran, W.G. 1967, Statistical Methods, 6th Ed.
Iowa State Univ. Press. Ames, Zowa.
-------
14
SUBDIVISION J — HAZARD EVALUATION: NONTARGET PLANTS
GUIDELINES
Series 120: GENERAL
5 120-1 Overview.
(a) General* (1) Scope. This subdivision deals with data
submittal to support registration of all outdoor use pesticides that
come in contact with plants. This subdivision addresses testing for
adverse pesticidal effects to nontarget plants, including those which
are within the pesticide application target area (such as crop plants
which are growing with weeds or are hosts for insects and disease
organisms), and those which are outside the target area (such as
typical adjacent crop plants, desirable ornamentals, garden plantings,
important wildlife food and cover species, and forestry, lurcher, and
conservation plantings and endangered and threatened plant species).
This subdivision addresses plant toxicity with respect to that
resulting from either direct exposure (i.e., application of a pesti-
cide to a plant) or from indirect exposure (i.e., exposure resulting
from movement of the pesticide through the environment as from
runoff, soil erosion, spray drift, etc.).
(2) Organization. (i) This subdivision contains two broad
areas of testing procedures:
(A) Toxicity to plants in the target area {§ 121-1); and
(B) Toxicity to plants outside of the target area (section
series 122, 123, 124).
(ii) These data should be derived from tests and reported in
a manner which complies with the general test standards contained in
§ 120-3 and the general reporting requirements contained in § 120-4
as well as the specific standards and reporting requirements of each
section listed in paragraph (a)(2)(i) of this section.
(b) "When required" and "test substance" requirements. The
registration applicant should be careful to distinguish between the
•when required" and the "test substance" paragraph requirements of
each section of this subdivision:
(1) The "when required" paragraphs restate the circumstances,
as found in 40 CFR Part 158, § 158.150, and specify the categories
of products for which data must be generated to support registration
applications. The test data are ordinarily provided to support the
-------
15
registration of each end-use product with the prescribed use pattern
and each manufacturing-use product used to make such an end-use
product.
(2) The "test substance" paragraphs state the kind of pesti-
cide material that must be used in each test. The test substance
for studies in this subdivision may be the technical grade chemical,
or a representative end-use product. Generally, each of these
test substances is prepared by the basic manufacturer of a pesticide
chemical.
(c) Testing to meet requirements. Since studies found in this
Subdivision would ordinarily be conducted by the basic manufacturer,
pesticide formulators would not often be expected to conduct such
tests themselves to develop data to support their individual prod-
ucts. (See 40 CFR § 158.50 concerning the formulators' exemption.)
They may do so if they wish, but they may also merely rely on the
data already developed by the basic pesticide manufacturer.
(d) Target area phytotoxicity testing waiver of requirements.
(1) The Administrator has determined that efficacy test data include
target area phytotoxicity testing data, and that data submittal for
such testing may be waived, by his authority under FIFRA Sec. 3(c)(5),
for most kinds of pesticide products. (See 44 FR 27938-27940, Friday
May 11, 1979.) Such products generally include all pesticides whose
uses result in direct or indirect application to plants in the target
area such as agricultural, lawn, and garden use.
(2) Even though the Administrator will ordinarily waive the
requirement for submittal of target area phytotoxicity test data as
indicated in paragraph (b)(1) of this section, he reserves the
authority to require such data on a case-by-case basis whenever the
Administrator deems that such data are necessary to evaluate the
acceptability of a product for registration. If it is determined
that data phytotoxicity for a pesticide are necessary, the Agency
will promulgate the specific target area phytotoxicity data require-
ments by letter to a specific registrant or by general notice.
(3) Thus, the guidelines .:: this subdivision should be used
by registration applicants as phytotoxicity test standards and
phytotoxicity data reporting requirements when target area phyto-
toxicity data are submitted to support registration applications.
The guidelines may also be used to provide guidance on testing
to support the claims and directions for use on product labeling
for products for which target area phytotoxicity data submittal is
waived.
(e) Nontarget area phytotoxicity testing. (1) Data require-
ments . Data concerning the determination of outdoor pesticidal
-------
16
effects on non-target area plants shall be required on a case-by-
case basis. (See 40 CFR § 158.150.) For example, if it is deter-
mined that the application of a pesticide will have an effect on
an endangered or threatened plant listed by the United States
Department of Interior, or if particular phytotoxicity problems
arise for which open literature data are not readily available,
phytotoxicity data may be requested. Hontarget area phytotoxicity
data will not be waived for pesticides that are under review for
or are in a cancellation or suspension proceeding, or against
which a rebuttable presumption against registration (RPAR) notice
has been issued. The Agency will promulgate the nontarget area
data requirements for RPAR and other requests by letter to a specific
registrant or by general notice.
(2) Testing scheme. Tests in the lower tiers (1 and 2) are
designed to screen those technical chemicals to determine the
potential to cause adverse effects on seed germination, vegetative
vigor, and aquatic plant growth and reproduction. The higher
tier (3) is designed to broaden the knowledge concerning any
detrimental effects on non-target plants of either technical
chemicals or formulated products. The criteria to proceed from
one tier to the next are contained in the "Tier progression" para-
graph of each section.
(3) Waivers. Waivers of specified nontarget phytotoxicity
test data or protocols may be requested. The request for waiver
must address the product application methodology, the pesticide
product's biological, chemical, and physical properties, and the
known phytotoxic properties of the pesticide product.
(4) Substitutions. If the pesticide or the active ingredient
of the pesticide (e.g., herbicides) has been extensively tested
using screening tests or other evaluation systems that are similar
in intent to any tests of Tiers 1, 2, or 3, the data from those
tests may be submitted in lieu of the required data of the tier
tests. The term "extensively tested" means testing of at least
the plants or plant families represented in §§ 122-Kb) (2) and
122-2(b)(2) under environmental conditions suitable to determine
any phytotoxic effects. The reports should be submitted as
provided in paragraphs (c) of §§ 122-1, 122-2, 123-1, 123-2, 124-1,
and 124-2. The Agency will reserve the right to require testing
as provided in Tiers 1 through 3 if the submitted test data do not
prove to be adequate to assess a pesticide's phytotoxic nature.
(f) Relation to other pesticide evaluation tests. (1) The
data requirements of tests of other subdivisions are imposed so
that duplicative testing is avoided to meet the requirements
40 CFR Part 158. Where data are submitted to fulfill the require-
ments of one subdivision, cross references to that data should be
made by the registrant if the data are also required elsewhere.
-------
17
(2) The registration applicant is referred to Subdivision H
"Labeling for Pesticides and Devices" for requirements on pesticide
labeling. One of the important objectives of the testing programs
required in Subdivision J is to develop sufficient data to support
appropriate and adequate precautionary labeling statements and
instructions for use, with respect to nontarget plants. Applicants
should read the appropriate paragraphs of § 100-9 and section series
104 of Subdivision H dealing with phytotoxicity and nontarget plant
effects.
§ 120-2 Definitions.
Terms used in this subdivision shall have the meanings set forth
in FZFRA at § 162.3, sec. 3 regulations, at § 60-2 of Subdivision 0,
and at § 90-2 of Subdivision G« In addition, for the purposes of
this subdivision:
(a) The term "algae" includes all chlorophyllous Thallophyta
other than the Bryophyta. It includes the blue-green algae
(Cyanobacterium or Cyanophyta), green algae (Chlorophyta), golden
algae and diatoms (Crysophyta), brown algae (Phaeophyta), red algae
(Rhodophyta), and golden-green algae (Xanthophyta).
(b) The term "aquatic plants" includes those plants that are
totally aquatic (free-floating or attached, submersed, and immersed)
and those which are semi-aquatic such as swamp and wetland plants.
(c) The term "desirable plants" means those plants that are not
to be detrimentally affected during pesticide application. They may
include crops, ornamentals, or wild plants inside or outside of the
area of intended application.
(d) The term "ECx" means that external pesticide concentration
required to cause a detrimental change or alteration (in a nontarget
plant) expressed as a percent (x) in comparison to untreated control
plants. An EC25 and EC50 are the concentrations required to effect
a 25 and 50 percent detrimental change, respectively, on nontarget
plant growth or activity,
(e) The term "EDx" means that internal pesticide concentration
or dosage required to detrimentally affect plant growth and
differentiation (in a nontarget plant) expressed as a percent (x) in
comparison to untreated control plants.
(f) The term "Ix" means that pesticide concentration required
to effect a detrimental change (usually inhibition) in enzymatic
activity in a plant expressed as a percent (x) in comparison to the
specific en2ymetic activity in untreated control plants. For example,
-------
18
ISO is used to indicate a 50 percent reduction in the activity of the
enzyme in question*
(g) The term "microorganism" means any of those organisms
classified as algae, fungi (Myxomycota and Eumycota), and bacteria
(Schizomycota).
(h) The terms "nontarget plant" and "nontarget microorganism"
mean any plant and microorganism species not considered to be pests
in the location in which it is growing. These species are not
intended to be controlled, injured, Jellied, or detrimentally-affected
in any way by a pesticide. "Nontarget plants" include desirable
or pest host plants such as crops or ornamentals within the target
area, and desirable plants outside the target area.
(i) The term "pest-free" means as free of pests as reasonably
possible. For all pesticide phytotoxicity tests, damaging insects
and surrounding weeds should be controlled so that healthy desirable
plants are available for testing. With this action detrimental
effects can be attributed to the pesticide in question, not to another
pesticide, or to weeds, or damaging insects.
(j) The term "phytotoxicity" or "plant toxicity" means unwanted
detrimental deviations from the normal pattern of appearance, growth,
and function of plants in response to pesticides and to other toxic
chemicals that may be applied with the pesticide. The phytotoxic
response may occur during germination, growth, differentiation, and
maturation of plants, and may be of a temporary or long-term nature.
Phytotoxic responses include adverse effects on growth habit, yield,
and quality of plants or their commodities to the extent that a
relationship between cause and effect can be established.
(k) The term "plants" includes vascular and nonvascular plants,
algae, and fungi.
(1) The term "representative end-use product" means a pesticide
product that is representative of a major formulation category (e.g.,
emulsifiable concentrate, granular product, wettable powder) and
pesticide group (e.g., herbicide, fungicide, insecticide, etc.) and
contains the active ingredient of the applicant's product.
(m) The term "target area" means the area intentionally treated
with a pesticide when label use directions are followed.
(n) The term "target area plants" means all plants located
within the target area, and includes both desirable and undesirable
species.
-------
19
§ 120-3 Basic test standards.
(a) Scope. This section contains test standards that apply to
all studies in this subdivision. If a specific test of this subdivi-
sion contains a standard on the same subject/ that specific test
standard shall take precedence in the performance of that particular
study.
(b) general. The experimental design, execution of the
experiments, classification of the organism, sampling, measurement,
and data analysis in support of an application for registration must
be accomplished by use of sound scientific techniques recognized by
the scientific community. The uniformity of procedures, materials,
and reporting must be maintained throughout the toxicity evaluation
process. Refinements of the procedures to increase their accuracy
and effectiveness are encouraged. When such refinements include
major modifications of any test procedure or standard, the Agency
should be consulted before implementation. All references supplied
with respect to protocols or other test standards are provided as
recommendations.
(c) Personnel. (1) All testing and evaluation must be done
under the direction of personnel who have the education, training,
and/or experience to perform the testing and evaluation in accordance
with sound scientific experimental procedures.
(2) To help assure consistency in the development of data, one
person should be responsible for each particular phase of the study.
(d) Test substance. (1) Plant hazard evaluation tests to sup-
port the registration of a pesticide shall employ either the tech-
nical of the active ingredient -or the formulated end-use product(s),
as specified in the following series of sections in this subdivision:
121, 122, 123, and 124.
(2) The composition of the test substance shall be determined,
including the name and quantity of contaminants and impurities in
order to account for 100 percent of the test sample in accordance
with § 61-1 of Subdivision 0. If the test substance is a formulated
product, it shall be within the limits, if any, certified in accordance
with § 62-2.
(3) Samples from the same lot of the test substance should be
used throughout a particular laboratory test or study. Field tests
may use samples from several lots due to the volume and geographical
requirements. The samples should be stored under conditions that
maintain their purity and stability. In the case of formulated
products, storage should be under conditions as found in commonly-
recognized storage practices.
-------
20
(4) If a carrier, vehicle, or adjuvant is used to dissolve,
dilute, or modify the physical characteristics of the test substance
for any study, it should be chosen to possess as many of the follow-
ing characteristics as possible:
(i) It should not interfere with the metabolism (degradation)
of the test substance;
(ii) It should not alter the chemical properties of the test
substance; and
(iii) At levels used in the study, it should not produce
physiological or toxic effects to plants*
(5) Where the test substance does not readily dissolve in water,
for example in Tier 1 and 2 tests, acetone, alcohol, or other suitable
solvent may be used to facilitate dissolving the substance in water
or other suitable carrier. Other adjuvants should not be used.
(6) In addition to or in lieu of data required by this subdivi-
sion, the Agency may require, after consultation with the applicant,
data derived from testing to be conducted with:
(i) An analytically pure grade of an active ingredient;
(ii) The technical grade of an active ingredient;
(iii) An inert ingredient of a pesticide formulation;
(iv) A contaminant or impurity of an active or inert ingre-
dient;
(v) A metabolite or degradation product of an active or
inert ingredient;
(vi) The pesticide formulation;
(vii) Any additional substance which enhances the phytotoxic
activity (up to and including synergistic effects) of the product
for which registration is sought; or
(viii) Any combination of the test substances mentioned in
paragraphs (d)(5)(i) through (vii) of this section.
(e) Nontarget plant test species. (1) The organism species
or groups to be tested are specified in the following series of
sections of this subdivision: 121, 122, 123, and 124.
(2) Healthy plants must be used.
-------
21
(3) Either cultivated crop, ornamental, or wild indigenous
plants may be used; endangered or threatened species as determined
by the Endangered Species Act of 1973 (Public Law 93-205} shall not
be used.
(4) Test organisms that are obtained from natural systems and
which are to.be used for testing should be maintained under condi-
tions similar to their natural or normal cultural environment•
(5) The population size of each replicate or treatment should
be large enough to assure meaningful results. Sample sizes should
be selected which will yield results that are statistically signifi-
cant at the 90 to 95% level of confidence with a significance
level of less than 0.10. The sample size for each plant species
in the tier tests (section series 122 and 123) should be of suffi-
cient size to statistically support the 25 or 50% (EC25 or EC50)
progression criteria.
(f) Nontarget organism safety. While performing field tests,
all necessary measures should be taken to ensure that nontarget
plants and animals, especially endangered or threatened species,
will not be adversely affected either by direct hazard or by impact
on food supply or food chain.
(g) Controls. Control groups are used to assure that effects
observed are associated or attributed only to the test substance
exposure. In phytotoxicity evaluations, all treated plots, plants,
and commodities must be compared directly to untreated control plots,
plants, and commodities. The appropriate control group should be
similar in every respect to the test group except for exposure to the
test substance. Within a given study, all test organisms including
the controls should be from the same source. To prevent bias, a
system of random assignment of the test plants to test and control
groups is required. Where a carrier, vehicle, or adjuvant other
than water is used, appropriate experiments and controls should be
included to distinguish the possible action of the carrier, vehicle,
or adjuvant.
(h) Equipment. (1) All equipment used in conducting the test,
including equipment used to prepare and administer the test substance,
and equipment to maintain and record environmental conditions, should
be of such design and capacity that tests involving this equipment
can be conducted in a reliable and scientific manner. Equipment
should be inspected, cleaned, and maintained regularly, and be
properly calibrated.
(2) The application equipment used in testing products in small
field plot studies should be designed to simulate conventional farm
equipment. This can be accomplished by using the basic components
of commercial application equipment in the design of the small-plot
equipment. For example, nozzle types, sizes, and arrangements on
-------
22
small plot sprayers can be identical to those used by growers on
commercial ground sprayers; or single-row commercial granular
application equipment mounted on a garden tractor for small plot
trials should produce results comparable to a multiple of such
units on a large tractor• For large-scale field trials, commer-
cial application equipment should be used. Specific details as
to descriptions of equipment design/ adjustment/ and operation
should be provided in test reports.
§ 120-4 General evaluation and reporting requirements.
(a) General. (1) Experimental use permits may be required
for the terrestrial testing of pesticides under field conditions
involving more than 10 acres, such as in studies described in
§§ 121-1 and 124-1. A permit may be required for aquatic field
testing of pesticides of more than one acre for studies described
in §§ 121-1 and 124-2.
(2) The report should include a detailed and accurate descrip-
tion of test procedures, materials, results and analysis of the
data, a statement of conclusions drawn from the analysis, and a
tabular summary and abstract of results. When they have been
determined, the primary and secondary modes of action with respect
to plant morphogenic and biochemical levels should be reported.
(3) The metric system should be used in test reports. The
U.S. standard measures may be used to preclude extensive conver-
sion to the metric system. The two systems shall not be mixed
(e.g., g/sq. ft.).
(4) The English language shall be used in all test reports.
English translations must be provided with foreign language reports.
(b) Test materials and methods. (1) Dates. Report the
actual dates of the studies including date(s) of initiation (plant-
ing, transplanting, and cultural practices), applications),
observations, and harvest.
(2) Laboratories. The names of the laboratories or institu-
tions performing the tests should be included.
(3) Personnel. Name and title of each investigator, and the
name, address/ and phone number of the employer should be reported.
-------
23
(4) Test substance. Identification of the test substance
shall be provided, including:
(i) Chemical name, molecular structure, and qualitative and
quantitative determination of its chemical composition;
(ii) Relevant properties of the substance tested, such as
physical state/ pH/ and stability; and
(iii) General identification and composition of any vehicles
(e.g., diluents, suspending agents, and emulsifiers} or other
materials used in the testing of the substance*
(iv) Appropriate portions of this reporting requirement may
be satisfied by cross-referencing to Subdivision D (§ 61-1, §§ 64-1
thru -21).
(5) Dntreated control (check) plots. Detailed descriptions
of plots and plants used as controls for comparisons of toxic
effects should be included for each test. Untreated control (check)
plots should be treated and evaluated in the same manner as the
treatment plots with respect to other pesticides or chemical
(fertilizers, etc.) and cultural practices.
(6) Test organisms. The description should include the iden-
tification of the test organisms (genus, species, and cultivar or
variety, as appropriate), rationale for selection of the species
employed, and location of plant collection areas including their
physiographic data. When plant species other than those identified
for specific studies have been tested, their degree of suscept-
ibility to the pesticide should be included in the test report.
This susceptibility should be reported in terms of EC values as in
the regular test plant reports.
(7) Location. Geographic location, including relation to the
target sites, should be reported.
(8) Substrate conditions, (i) For aquatic pesticide applica-
tions, the following physiographic conditions should be reported:
(A) Type of aquatic site, such as lake, pond, reservoir,
stream, or irrigation ditch with flow rate (if moving water);
(B) Size (area and depth or volume or length, width, and depth
of the treated areas, and of the whole site), as is appropriate to
the type of application and the type of target organism(s);
(C) Water quality including pH and temperature and hardness,
alkalinity, or salinity, where possible;
-------
24
(D) Turbidity (visual), conductivity (if possible), and
dissolved oxygen (for submerged plants only); and
(E) Soil texture, including that of soils along the immediate
shoreline or ditchbank and the submersed soil where the target pests
are present (with the percent organic material in the soil also
reported). (Recommended methods and soil texture classifications may
be found in the Walkley-Black Procedures in Soil Sci. 63:251, 1947,
and the Soil Survey Manual, U.S. Oept. Agr. Handbook No. 18, 1951,
Pig. 1, and Soil Sci. Soe. Amer. 26:305-317, 1962.)
(ii) For terrestrial pesticide applications, the following
physiographic conditions should be included:
(A) The edaphic conditions and characterization including soil
type and texture, and approximate pH and temperature;
(B) Where the presence of a fragipan or shallow bedrock may
lead to restricted leaching or soil waterflow, the depth of that
restriction; and
(C) The degree and direction of slope and its orientation to
the row direction if the slope will lead to excessive runoff.
(9) Environmental conditions, (i) For growth chambers and
laboratory experimentation, the light quality, light quantity (lux
or Einsteins m~2s~1}, air temperature, humidity, photo- and thermo-
periods, and watering schedules should be reported.
(ii) For greenhouse and field experiments, the approximate
light quantity (usually expressed in degree of cloudiness), high and
low daily air temperatures, relative humidity, and photoperiod (day
length) should be reported. The environmental conditions of the
specific field site are required only for the day of application.
Area or specific field environmental conditions may be used for long
term studies. Rainfall is to be reported for the duration of field
experiments.
(10) Application. (i) General. The test substance application
method should be reported, including dosage rates, application
equipment (nozzle, orifice, pressure), time and number of applications
(with reference to season and stage of growth), spray dilution, spray
volume per unit area, and adjuvants;
(ii) Application rates. Dosages should be reported in units
of active ingredient or acid equivalent as appropriate. Rates may
be expressed as units of ingredient per unit of land area to be
treated, units of concentration (such as parts per million), units
per flow rate, or units of ingredient per unit volume applied to
obtain a specified degree of foliage coverage (such as "to runoff").
If a product is applied more than once within a year or growing
-------
25
season, each rate and the interval between applications should be
indicated. If products are applied in a tank mixture or are applied
serially, rates and intervals, as appropriate, should be reported
with identification and formulation for each product.
(iii) Tl"f"g of applications. When the test substance,
particularly a herbicide, plant regulator, desiccant, or defoliant,
is applied to any desirable nontarget plants within or adjacent to
the target area, the plant's stage of growth or development at
application should be described in test reports.
(iv) Serial applications. In addition to the detrimental
effects of the pesticides, the times of application (or application
interval) should be indicated for each product or tank mix involved
in the serial application.
(c) Observations. (1) Observations should be reported to
include all variations, either inhibitory or stimulatory, between the
treated test organisms and the untreated control test organisms.
Such variations may be phytotoxic symptoms (chlorosis, necrosis, and
wilting), formative (leaf and stem deformation) effects, and/or growth
and development rates. Observations should include the stage of
development and dates when adverse results occurred and subsided or
recovered. Any lack of effects by the pesticide should also be
reported.
(2) Observations should be reported in sufficient detail as to
allow complete evaluation of the results. This evaluation, to be
performed by the registrant, should include the degree or extent of
effects exerted by the pesticide in question for each replicate and
variable.
(3) The detrimental or adverse effects to be considered and
reported during the observation period of terrestrial studies include:
(i) Stand or plant population;
(ii) Overall vigor of the plants expressed as height, weight,
diameter, length, or other similar aspect of growth;
(iii) Phytotoxicity or visible symptoms such as discoloration,
malformation, desiccation, or defoliation;
(iv) Lodging of plants;
(v) Effect on root growth and structure;
(vi) Development delay or acceleration with respect to
maturation; and
-------
26
(vii) Yield of the crop or commodity that is treated as com-
pared to those of crops or commodities of untreated check plots.
(4) Where pesticides are applied to aquatic systems and
influence plant growth and development in aquatic systems/ the
effects of that pesticide on nontarget plants in the system and
along the immediate border should be evaluated and reported, includ-
ing vigor of the plants, phytotoxicity or other visible symptoms,
and delay or acceleration with respect to vegetative growth, flower-
ing or speculation, and maturation.
(5) Uniform scoring procedures should be used to evaluate the
observable toxic responses.
(6) At least two methods of evaluation (such as quantitative
and qualitative determinations) should be used in the evaluation of
pesticide effects on growth, reproduction, and yield of plants in
greenhouse and controlled chamber experiments. When direct measure-
ments cannot be made, such as in large field evaluations, a zero
to one hundred (0-100) or zero to .ten (0-10) rating scale should
be used, where zero (0) indicates no injury and one hundred (100)
or ten (10) indicates a total effect or kill produced by the test
substance. An explanation of the steps of the rating scale em-
ployed should be included with the report. Other rating scales
(0 to 4; 0 to 9) may be used but are not conducive to statistical
analysis.
(7) Observation reports should include the basic data used
for the statistical analysis [see paragraph (d) of this section].
Such data should include the actual values used to determine any
percentages of effects. Raw data (chromatographs, field reports,
and analysis data) may also be included to substantiate the basic
data that are required.
(d) Statistical analysis. (1) When test results such as
efficacy, phytotoxicity, or yield indicate adverse effects on
crops and other nontarget test organisms, statistical analysis is
required in the evaluation the response(s). The statistical
analysis should consist of:
(i) The tabulation of the response data at each treatment
level;
(ii) The determination of 25 or 50 percent detrimental effect
levels (e.g., EC25, EC50, as appropriate) and the 95 percent con-
fidence limits, where possible, for each; and
(iii) The estimated non-discernible effect level. This is the
level at which there would be no significant effect on the intended
yield, quality, or aesthetics of the crop or plant which might be
exposed.
-------
27
(2) Statistical analysis is also useful in evaluation of
interactions resulting from studies supporting tank mixtures or
serial applications (See 121-1(b)(5) and (6)].
(e) References. Copies of references or Literature used in
modifying the test protocol, performing the test, making and inter-
preting observations, and compiling and evaluating the results
should be submitted. Copies of unpublished literature should also
be included. Copies of the recommended literature referenced in
these guidelines are not required.
(f) Special test requirements. In addition to the data
required in this subdivision, data from other tests may be required
by the Agency for making judgments regarding safety to nontarget
plants. Such data will be required where there are special prob-
lems, such as a proposed pattern of use, mode of phytotoxic action,
or a unique chemical property. Methods are usually derived from
those already described or cited in other subdivisions of these
guidelines.
-------
28
Series 121: TARGET AREA TESTING
$ 121-1 Target area phytotoxicity testing.
(a) When required. (1) General. (i) Data concerning the
phytotoxic effects of a pesticide on desirable target area plants
generally will be waived by 40 CFR Part 158 to support the registr-
ation of each end-use product intended for outdoor and greenhouse
applications or outdoor planting of treated material [see § 120-1(d)].
In certain situations noted in § 120-1(d), the Agency may request
phytotoxicity data from studies provided for in this section.
(ii) The data requirements of this section need not be ful-
filled for herbicides which provide long-term or total vegetation
control, e.g., clean yard chemicals, desiccants and defoliants.
(2) Experimental use permits. The registration applicant is
also reminded that an experimental use permit may be required in
order to conduct field studies described in this section. See
Subdivision I for information concerning experimental use permits.
(3) Simultaneous testing. The target area phytotoxicity tests
and reporting as described in this section may be performed simul-
taneously with the appropriate product performance tests described
in Subdivision G (Series 90 through 96).
(b) Test standards. In addition to the general standards
set forth in § 120-3, the following standards for the target area
phytotoxicity testing apply:
(1) Test substance. The test substance shall be the end-use
product or a representative end-use product from the same major
formulation category for that general use pattern. Examples of
major formulation categories are: wettable powders, emulsifiable
concentrates, and granulars. (If the manufacturing-use product is
usually formulated into end-use products comprising two or more major
formulation categories, a separate study must be performed with a
typical end-use product for each category.)
(2) Test species. Those desirable target area or pest host
plant species as listed on the label (for example, the crop plant or
ornamental) which will be within the target area should be tested.
The plant cultivars to be tested should include representatives of
the cultivars that are most likely to be used.
(3) Applications levels, (i) The minimum, maximum (or the
greatest allowable concentration), and 2 times the maximum label
-------
29
applice.v_r-. level ox rate should be tested. Levels greater than
2 times the label rate nay also be Included. The estimated non-
discernible effect (or no-effect) level should also be determined.
(ii) The multiples of the application rate to be tested are
those various quantities of the formulation in the label-recommended
quantity of carrier (such as water) to be used per land or aquatic
use area.
(4) Adjuvants. Products with labeling which allows or recom-
mends the addition of separately-packaged adjuvants to the spray
tank should be supported with data indicating any detrimental
effects (such as increased crop phytotoxicity) which may result
from their addition to the pesticide, especially a herbicide,
plant regulator, desiccant, or defoliant. If a range of adjuvant
rates is recommended, the maximum rates within that range should
be evaluated in conjunction with the intended pesticide product.
(5) Tank mixtures. When tank mixtures are recommended on
product labeling, a study may be required on a case-by-case basis
to demonstrate the extent of antagonism and synergism with respect
to detrimental effects on nontarget plants by the products of tank
mixtures. Antagonism and synergism are best evaluated in adjacent
plots where possible interactions are subjected to statistical
analysis. See § 164-4 of Subdivision N for possible combined test-
ing.
(6) Serial applications. Data requirements for serial appli-
cation(s) of one or more pesticide(s) preceding or following
another pesticide on the same crop area in the same growing season
are identical to those described in paragraph (b)(5) of this
section for tank mixes with respect to phytotoxicity, when such
serial applications are recommended on the label. See § 164-4 of
Subdivision N for possible combined testing.
(7) Site. The test should be performed in greenhouses or
wherever the product is intended to be used.
(8) Protocol. The protocols, methods, or practices should be
those employed for the anticipated registered use of the pesticide
product. Specific points of information that should be addressed
concerning use patterns, application methodology, cultural prac-
tices , responses, and subsequent planting are found in paragraph
(c) of this section.
(c) Reporting. In addition to the information required by
§ 120-4, the test report should include the following information
with respect to phytotoxicity to the plants within the target area
(with the exception of weeds). This information should include
the method of application, cultural practices, plant responses,
subsequent plantings, and use patterns that may be involved.
-------
30
(1) General information, (i) Timing of applications. When
crops or desirable target area plants are or will be involved in the
application of any pesticide, their stage of growth or development
at application should be described in the test report. •'
(ii) Meteorological conditions. Where meteorological condi-
tions cause detrimental effects on plants which in turn allow the
pesticide to further adversely affect the plants, the specific
factor(s), such as temperature, wind conditions, precipitation, or
daylength, affecting product activity should be measured and
reported. Bdaphic factors, such as soil moisture content and
temperature, which are directly affected by meteorological con-
ditions, should also be reported. Soil moisture may be observed
and expressed in terms of dry and cracked, waterlogged, or other
similar conditions. Organic matter content of the soil should
also be reported.
(iii) Spray dilutions. In foliar applications, when a pesti-
cide is applied as a diluted spray and the quantity is dependent
upon the number of trees per area or density of vegetation, the
total spray volume per unit area, and the concentration of the
applied pesticide should be reported.
(iv) Untreated controls (checks). In phytotoxicity evalua-
tions, all treated plots, plants, and/or commodities should be
compared directly to untreated control plots, plants, or commod-
ities. All quality and/or yield evaluations of pesticide-treated
plants or commodities should be compared to control plants or
commodities receiving the same pesticides (e.g., herbicides,
insecticides, fungicides) except the one being evaluated. Detailed
descriptions of plots and plants used as control treatments for
comparisons of detrimental side effects should be included for
each test. Since such control plots are established to evaluate
any direct detrimental effects of the pesticide on the crop or
commodity rather than to evaluate efficacy, any detrimental
effects on the crop or commodity resulting from pests should be
controlled. In other words, the control plots should be both
untreated by the pesticide in question and as pest-free as reason-
ably possible. If, in addition to the untreated control plots,
plants, and/or commodities, a registered product is applied (as a
standard) for comparison of detrimental effects, data should
indicate the standard product's name, active ingredient, dosage
rate, and phytotoxicity results. Where infestations of weeds
occur in check (or test) plots, the degree of infestation and
species of weed(s) should be reported.
(2) Use patterns. .When the following use patterns are found
on the label, the corresponding information as detailed below should
be reported.
-------
31
(i) Dse in field crops. Effects of pesticides on desirable
target area plants should be evaluated and reported. The extent
and duration of the effect should be expressed in terms of stand
and vigor, recovery, yields, and degree of phytotoxicity.
(ii) Use on pastures and rangelands. Effects of pesticides
on desirable target area plants should be evaluated and reported.
Severity and duration of adverse effects on desirable plant species,
expressed in terms of stand and vigor reductions, recovery, and
changes in yields, should be reported. Data should be submitted
addressing reseeding intervals which minimize adverse effects on
reseeded plants, and animal grazing recommendations which allow
recovery of desired plant species. If the applied pesticide kills
all vegetation in the treated area for an extended period of time
resulting in bare spots, the registrant should record the duration
of this effect, estimated soil loss by erosion and any changes in
vegetation cover (desirable or undesirable).
(iii) Use on and around fruit and nut trees. Applications of
pesticides on and around fruit and nut trees require evaluation and
reporting of detrimental effects on foliage, and changes in growth
compared to preapplication measurements and simultaneous controls.
Pesticide applications to bearing fruit and nut tree areas also
require evaluation and reporting of detrimental effects on yields
and commodity (produce) quality for the year of and the year after
application. Supporting data should address, for all trees, the age
of the trees, the transplant-to-application interval, and the maxi-
mum allowable extent of contact between the pesticide (with par-
ticular reference to herbicide spray drift) and trees. For ground
sprays, unless the pesticide is broadcast over the entire orchard
floor, data should indicate the application technique (band, spot,
shielded, or directed spray application) and the size of the
treated ground area around the tree trunk. Assessment of root
sucker treatments should be made where applicable. For foliar
sprays, the data should include the volume of finished spray applied
per unit of land area, concentration of product in the spray solu-
tion, and the extent of foliage coverage (such as volume of finished
spray per tree or application to the point of runoff).
(iv) Use on lawns and turf. Evaluation of effects of pesti-
cides on representative species or cultivars of desirable lawn and
turf plants should include such factors as color, density, percent
cover, growth rate, rooting, and tillering. If use on bentgrass
is intended, this highly susceptible species should be evaluated.
Data should address use on newly-seeded lawns by demonstrating
safety to representative species and cultivars of desirable lawn
plants to be named on the label as kinds on which the product is
safe to use, with seeding-to-application intervals (if appropriate).
Data should also address use of an appropriate application-to-
reseeding interval for each of these desirable lawn plants that
may be reseeded. Interactions between herbicide application and
-------
32
lawn cultural practices (such as raking, mowing, mowing height,
watering, and fertilizing) should be evaluated for possible
adverse effects on desirable lawn species. In situations where
fertilizer and a pesticide are applied serially and both types
of products nay contact the emerged crop foliage (such as in turf
or lawns), the interval between application of the pesticide and
the fertilizer should be reported, as well as any resultant phyto-
toxic effect/ stunting, or discoloration, and recovery time for
the injured desirable species•
(v) Use around ornamentals. Phytotoxicity data in support of
use on or around an ornamental should include an evaluation of the
sensitivity of representative cultivars of that species. Since it
has been documented that cultivars and varieties of the same species
vary in their susceptibility to injury, the limited nature of test-
ing should be addressed in product labeling. Test data should iden-
tify the method of application as to directed spray and/or topical
applications. Growth stage of the ornamentals and the transplant-
-to-application interval (when applicable) should be indicated
in the test data. Information should be submitted on specialized
nursery cultural practices employed in tests, such as use of
artificial soils, mulches, containerized stock, and other pesti-
cides.
(vi) Use in forest management. The effects of the pesticide
on desirable plant species commonly present in forest management,
in addition to the desirable forest trees, should be indicated in
the report with any detrimental or adverse effects that the pesti-
cide may cause. Special attention should be given to pesticidal
effects on noncompetitive ground cover species that aid in the
land management practices such as erosion control. Appropriate
testing and assessment techniques adapted to the size of the plot
should be used to determine the effect of pesticides on all plants.
(A recommended reference is: Phillips, E.A. 1959. Methods of
Vegetation Study. Holt, Rhinehart, and Winston, Inc.: New York,
N.Y. 107 pp.)
(3) Application methodology. All methods of pesticide appli-
cation specified on the label should be evaluated and reported.
Specific detail as to descriptions of equipment design, adjustment,
and operation should be provided in test reports involving aerial
applications and applications using conventional farm equipment
(such as tillage or planting equipment), irrigation systems,
mechanical incorporation, directed sprays, mist blower (air
blast, air carrier), subsurface placement, or band rather than
broadcast distribution.
(i) Aerial application. Guidance and the data requirements
for testing aerial applications will be provided in a subdivision on
spray drift exposure assessment.
-------
(ii) Irrigation system application. (A) For irrigation sys-
tem applications, multiple plots and subplots within a treated
field should be examined and the results reported for crop phyto-
toxicity (expressable as yield quantity, quality, and timeliness
of harvestable commodity) as an indication of pesticide hazard.
Data from such plots should be reported for each individual plot
and not simply averaged together. It is important that, in addi-
tion to the standard requirements for conventional applications,
submitted data should include soil texture, percent soil organic
matter, relative soil moisture content (dry, medium, or wet) at
application, acre-inches of water applied, and precipitation quan-
tities within one week after application.
(B) For overhead sprinkler irrigation systems, plots should
be placed at both extreme ends of the lateral as well as in at
least one area where the sprinkler patterns overlap. On a center
pivot, one might have to use several "pie" sections for treatment
subplots in one half with the second half as the control. The
concentration of active ingredient at several nozzles along the
lateral should also be determined and reported.
(C) For surface irrigation systems such as flood, furrow, drip,
and surge, the following data should be submitted. Concentrations
of active ingredients in water should be determined for the study
plots where the treated water enters the field, and at the lower end
of the field or where the water exits. When furrow irrigation is
used, data should indicate the spatial relationship between crop rows
and furrows. If pest control in furrow irrigation applications is
intended only for the furrow itself and not the bed between the
furrows, the data should so indicate.
(iii) Directed sprays. When sprays are directed toward or away
from certain portions of the soil or plants, data should indicate
nozzle arrangements, nozzle orientations, the extent of spray contact
with soil or plants, and application height.
(iv) Mist blower applications. Guidance and the data require-
ments for testing mist blowers (air blast and air carriers) will be
provided in a subdivision on spray drift exposure assessment.
(v) Subsurface soil applications. When pesticides are ap-
plied directly beneath the soil surface (injected through shanks
or spray blades, or gravityfed), test reports should include infor-
mation on the application equipment. For example, for injection
equipment, the following should be specified: application device
spacing, depth of operation, injection pressure, speed of opera-
tion, volume of liquid or gas applied per unit area for general
broadcast applications or linear row distance for band and row
applications, and the number and placement of injectors with
respect to plant rows.
-------
34
(vi) Other aquatic applications. When a pesticide is applied
to a natural aquatic system other than an irrigation system, the
following application information should be included:
(A) Target site where the pesticide was applied (for example,
to weed foliage, to surface of water, to bottom of water body, into
water, to ditchbanJe, to shoreline, or to forests);
(B) Description of the equipment used to apply the pesticide
(for example, ground-spraying device, pumping device, boat, blower,
helicopter, or fixed-wing aircraft);
(C) Description of any water level changes used in conjunc-
tion with the pesticide application, such as drawdown operation or
drainage of conveyance system, including the extent of water
level change, the time of the change in relation to the pesticide
application, and the duration of the change in water level; and
(D) The timing of the application in relation to the calendar
date and the stage of growth of the target and nontarget organisms.
(4) Cultural practices. Cultural practices for a given use
pattern or application method vary with production areas and fre-
quently from grower to grower within an area. The effects of
cultural practices on the product's possible detrimental effects
should, therefore, be addressed.
(i) Irrigation. Irrigation and watering practices should be
studied as a variable if the product is to be used in irrigated
areas or greenhouses, respectively. The influence of different
irrigation practices should be studied in the use area. Irrigation
data should include a description of equipment and techniques used
in water application, the number and timing of irrigations, and
quantity of water in acre-inches (hectare-centimeters) applied at
each irrigation. Also, describe the chronological relationship
between irrigation applications and application of the pesticide,
such as herbicide, plant regulator, desiccant, or defoliant.
Where flood irrigation is utilized (such as in rice production),
depth, duration, and any "flushing" should be described for each
test. When irrigation is used to activate a pesticide in the
absence of precipitation, the minimum and maximum application-to-
irrigation interval (producing the desired efficacy level) should
be reported. Since crop safety is often influenced by pesticide
placement in the soil profile, and irrigation may directly affect
such placement, label-recommended or label-allowed irrigation
practices should be supported by crop safety data (phytotoxicity
and yield). When irrigation practices result in loss of pesticide-
contaminated water (as in runoff or drainage) from the target area,
data should be submitted addressing effects of such water on non-
target plants.
-------
35
(ii) Moving. Mowing operations may enhance detrimental effects
from pesticides intended for use on lawns/ turf, golf courses, median
strips, pastures, rangeland, and hay and forage crops. Mowing just
prior to or just after a pesticide application may, by mechanically
injuring desirable plants or by decreasing growth rates, increase
injury to desirable plants (especially young shoots). Mowing just
prior to application may be a requirement for plant regulators in-
tended to maintain the neat appearance of grassy areas by retarding
grass growth. In situations where mowing is routinely a part of
cultural practices, or may influence detrimental effects, such
practices should be reported in test results.
(5) Target area plant responses. The detrimental effects on
crops, commodities (produce), or any other desirable plant species
or commodity within the target area should be evaluated and reported.
The following are some of the characteristics that should be addressed:
(i) Stand. Crop stand counts, reported as percentage of
untreated control crop stands, should be submitted to support pesti-
cides applied prior to crop emergence.
(ii) Vigor. Crop vigor (or stunt) ratings or measurements
(plant height, weight, diameter, or length) in treated areas should
be compared to plants in check plots in which commercially acceptable
levels of pest control are maintained. Vigor ratings should be
reported at the point of maximum stunting. Zf stunting is observed,
it is important that subsequent evaluations be made to document the
degree of recovery.
(iii) Planting depths. A range of planting depths within the
range recommended for the crop should be included in preliminary
studies with preplant and preemergence (to crop) applications. Data
obtained from these trials should reflect any effects of varying
planting depths on the incidence of crop Injury that might be
encountered under commercial use conditions. In subsequent trials,
commercial planting equipment at recommended depth settings should
be used. Zf in preliminary studies the planting depth is found to
be a critical variable, crop emergence data should be taken from all
trials.
(iv) Lodging. The effect of pesticides on lodging of target
area crops such as soybean, wheat, corn, sorghum, rice, or sugarcane
should be Indicated. Observed percent cf treated plants affected and
the severity or approximate degree of ang."? of lodging in treated
plots should be compared to that in weed-i.ee check plots.
(v) Phytotoxicity. Evaluations of visible symptoms of pesti-
cide injury (such as discoloration, malformations, desiccation,
defoliation, or death) to crop plants should be at least visually
assessed and reported. These symptoms should be compared to results
-------
36
in check plants untreated with the pesticide in question. Evalua-
tions should be performed at the time injury is first observed and
at periodic intervals thereafter to document the degree of recovery.
(vi) Development. Effects of pesticides on plant development
(such as delayed emergence, prolonged vegetative growth, delayed or
decreased flowering or fruit set, or delayed maturation) should be
indicated in test results. If such effects are outgrown by or before
the usual harvest date, such recovery should be reported.
(vii) Yields. Effects of pesticides on yields should be
reported. Yield data can confirm that there are no lasting detri-
mental effects on the desirable target area plants due to the
pesticide application. Yield data may also be used to evaluate
benefits derived from the application. When yields are evaluated
in relation to crop safety or phytotoxicity, yields from treated
plots should be compared to yields from untreated plots. Compari-
sons of treated and untreated (control) plot yields, when expressed
as weight of seed (grain and dry beans) or hay, should be based
upon equivalent moisture contents (percent moisture) acceptable
for commodity storage. In the case of weed control, yields from
weedy check plots may be reduced as a result of weed competition
and may mask crop injury due to herbicide application. Therefore,
herbicide yield comparisons should be drawn from the treated plots
and weed-free plots. The maintenance of weed-free control plots
may be accomplished by some other weeding practice or by use of a
commonly-used (reference) herbicide. When any adverse effects
indicated in paragraphs (c)(5)(i) through (vi) of this section
occur, the ultimate indication of their impact can usually be
evaluated at harvest.
(6) Subsequent planting. The effects of pesticides on desir-
able plants subsequently planted in the area within six months of
application should be evaluated and reported. Subsequent planting
may include emergency replanting of crops or trees within the
target area where crop failure may have occurred and where the
planting of rotational crops (including cover crops) takes place
after the harvesting of the crop present during the pesticide
application.
(i) Emergency replanting. If pesticide labeling states that
crops may be safely replanted after an initial crop failure, the
•submitted data should support: the crops suitable for replanting;
pesticide application-to-replanting intervals; additional pesticide
applications recommended or allowed; recommended soil tillage; and
soil and meteorological conditions under which replanting is or is
not recommended. For example, when the original pesticide was
applied in bands, as in the case of certain herbicides, replanting
may be recommended to take place only between the treated bands.
-------
37
(ii) Rotational crops (including cover crops). If detrimental
effects are observed, results of studies evaluating severity and
duration of effects on the injured rotational crops should be sub-
mitted. To determine the duration of phytotoxic effects, susceptible
rotational crops should be planted at varying time intervals after
pesticide application. Such studies may be combined with field
studies designed to evaluate soil residues. [See § 165-2 of Sub-
division N.]
-------
38
Series 122: TIER 1 OF NONTARGET AREA TESTING
§ 122-1 Seed germination/seedling emergence and vegetative vigor
(Tier 1).
(a) When required. (1) Data on the toxic effects of a pest-
icide on seed germination or seedling emergence and vegetative
vigor are required by 40 CFR Part 158 on a case-by-case basis to
support the registration of each end-use product intended for
outdoor pesticide application, and each manufacturing-use product
which legally could be be used to make such end-use products.
[See § 120-1(e).]
(2) Studies of this section need not be conducted for pesti-
cides applied by systems where the chemicals are not readily
released into the environment. Examples of these systems are:
tree injection, subsurface soil applications, recapture systems,
and wick applications and swimming pools.
(3) Portions of this Tier 1 test may be combined with the
respective parts of the Tier 2 test (§ 123-1) and performed as one
test.
(4) See § 120-1(e) concerning substitution of testing and data
submission requirements.
(b) Test standards. In addition to the general test standards
set forth in § 120-3, the following standards for the seed germina-
tion or seedling emergence and vegetative vigor studies apply:
(1) Test substance. The technical grade of the active ingre-
dient shall be tested. Where a technical grade does not exist,
the manufacturing-use product or an end-use product with the highest
percentage of the active ingredient shall be used.
(2) Species. The following plant species and groups should
be tested:
(i) Dieotyledoneae; Six species of at least 4 families, one
species of which is soybean (Glycine max) and a second of which is
a root crop.
(ii) Monocotyledoneae; Four species of at least 2 families,
one species of which is corn (2ea mays).
(3) Application levels. One concentration level equal to no
less than maximum label rate should be tested. If it can be deter-
mined that the maximum quantity that will be present in the non-
-------
39
target area is significantly less than the maximum label rate, a
concentration equal to no less than 3 times that maximum quantity
may be tested. The phrase "the maximum label rate" means the
maximum recommended amount of active ingredient in the recommended
minimum quantity of carrier such as water to be used per land
area. For purposes of calculating the dose level in the seed
germination study, 1 pound of active ingredient per acre should be
considered to be equal to 3 ppov in the solution which is applied
to seeds. (Mote: a 1 Ib. ai/acre application to a 3 inch soil
depth would equal 7.5 ppmw in the soil solution*)
(4) Number of plants. At least 3 replicates, each with
5 plants, should be tested per dose level for the vegetative vigor
tests. At least 3 replicates, each with at least 10 seeds, should
be tested per dose level for the seed germination study. Larger
populations and more replicates may be needed to increase the
statistical significance of the test.
(5) Site. The seed germination/seedling emergence studies
should be conducted under controlled conditions in growth chambers
or greenhouses. The vegetative vigor test may be performed in a
growth chamber, greenhouse, or in small field plots.
(6) Duration. (i) Seed germination, if performed using petri
plates or seed germination paper, should be assessed after 5 days.
Seedling emergence should be observed weekly, or more frequently,
for at least two weeks after germination.
(ii) The effect of vegetative vigor should be observed weekly,
or more frequently, for at least two weeks. If abnormal symptoms
occur, the observations should be continued until the plant dies
or fully recovers.
(7) Protocols. The protocols for these tests outlining the
acceptable environmental conditions, procedures, and some pertinent
references are found in § 122-30(a) through (c).
(c) Reporting. In addition to the information required in
§ 120-4(b), the test report should include the following informa-
tion.
(1) The number of seeds tested and the number germinated or
emerged per dosage level for each replicate;
(2) Descriptions'of the appearance and the growth and develop-
ment of the seeds and emergent plants, indicating any abnormalities
and expressions of phytotoxicity; and
(3) Tabulation of the results indicating the percentage
effect level for each species as compared to untreated control
plants.
-------
40
(4) Data on weight and height or other growth parameters may
also be submitted.
{d) Tier progression. (1) If the results of the seed
germination/seedling emergence test(s) have indicated an adverse
effect greater than 25 percent on one or more plant species, then
seed germination or seedling emergence tests at the Tier 2 level are
required (see § 123-1).
(2) ' If the results of the vegetative vigor test(s) have indi-
cated an adverse effect greater than 25 percent on one or more
plant species, then vegetative vigor tests at the Tier 2 level are
required (see § 123-1).
(3) If less than a 25 percent detrimental effect or response
is noted for either seed germination/seedling emergence or vegeta-
tive vigor tests, no additional testing of the respective tests
at higher tiers is ordinarily required. The Agency, after review
of the data, may require certain additional tests to determine a
more definite nondiscernible effect level.
§ 122-2 Growth and reproduction...of aquatic plants (Tier 1).
(a) When required. (1) Data on the toxic effects of a pesti-
cide on growth and reproduction of aquatic plants are required by
40 CFR Part 158 on a case-by-case basis to support the registration
of each end-use product intended for outdoor pesticide application,
and each manufacturing-use product which legally could be used to
make such end-use products. [See § 120-1(e).]
(2) Studies of this section need not be conducted for pesti-
cides applied by systems where the chemicals are not readily
released into the environment. Examples of these systems are:
tree injection, subsurface soil applications, recapture systems,
and wick applications.
(3) Portions of this Tier 1 test may be combined with the
respective parts of the Tier 2 test (§ 123-2) and performed as one
test.
(4) See § 120-1(e) concerning substitution of testing and
data submission requirements.
(b) Test standards. In addition to the general test standards
set forth in § 120-3, the following standards for the studies of the
growth and reproduction of aquatic plants apply:
-------
41
(1) Test substance. The technical grade of the active ingre-
dient shall be tested. Where a technical grade does not exist,
the manufacturing-use product or an end-use product with the highest
percentage of the active ingredient shall be used.
(2) Species, (i) Selenastrun eapricornutum (a) freshwater
green alga) should be tested regardless of the intended outdoor use
pattern.
(ii) If the intended use pattern is for outdoor aquatic pest
control at sites other than swimming pools, the following species
should also be tested:
Lemna gibba (duclcweed);
Skeletonema costatum (marine diatom);
A freshwater diatom (unspecified species); and
Anabaena flos-aquae (blue-green alga).
(3) Application levels. The quantity of test substance to be
tested should be equivalent to the maximum label rate as though it
were directly applied to the surface of a 15-cm or 6-inch water
column. The application of 1 Ib active ingredient per acre or 1.1 kg
per hectare is equal to 735 parts per billion (ppb) in a 6-inch or
15-cm water column. If it can be determined that the maximum quan-
tity that will be present in the nontarget area is significantly
less than the maximum label rate, a concentration equal to no less
than three times that maximum quantity may be tested.
(4) Number of plants. At least 3 replicates, each with 5 vas-
cular aquatic plants (Leana gibba - stage: 3 fronds per plant) should
be tested per dose level. The recommended quantities of algal plant
material to be used are provided in the recommended references of
the protocols provided in § 122-30(d) through (h). Larger popula-
tions and more replicates may be needed to Increase the statistical
significance of the test.
(5) Site. All studies provided for in this section should be
conducted under controlled conditions In growth chambers.
(6) Duration. (1) Lemna studies should be conducted for at
least 14 days with observations at least every three days.
(ii) Algal studies should be conducted for at least five days with
daily observations. Observations may continue until the occurrence
of maximum standing crop of the controls.
-------
42
(7) Protocols. The protocols for these tests outlining the
acceptable environmental conditions and procedures and some
pertinent references are found in § 122-30(d) through (h).
(c) Reporting. In addition to the information required by
$ 120-4(b)(1) through (6), and (8), (c), (d), and (e) of this
subdivision, the test report should include the following:
(1) Lemna. The change in growth expressed as the number of
original plants and fronds and the additional plants and fronds
produced;
(2) Algae. Growth should be expressed as the cell count per
ml, biomass per volume, or degree of growth as determined by
spectrophotometric means; and
(3) Tabulation of the results indicating the percentage effect
level versus time as compared to the control.
(d) Tier progression. (1) If a detrimental effect or response
on plant growth and development for any aquatic plant species for the
maximum label rate is greater than 50 percent with respect to the
controls, testing at Tier 2 is required. See § 123-2.
(2) If less than a 50 percent detrimental effect or response
is noted, no additional testing at higher rates is required. The
Agency, after review of the data, may require certain additional
tests to determine a more definite nondiscernible effect level.
§ 122-30 Acceptable methods and references.
The following test protocols have been developed to provide
guidance in the performance of pesticide plant hazard evaluation
testing:
(a) Seed germination. (1) Protocol, (i) Seeds are germinated
between sheets of sterile filter paper or germination paper moistened
with the chemical; or the seeds are germinated in acid-washed quartz
sand or in "standard" soil that has been sprayed or otherwise treated
with a known quantity of the chemical. The seeds may be surface-
sterilized.
(ii) Use at least ten seeds per dish. The seeds are incubated
for at least five days. The test temperature should approximate the
optimum temperature for the species and variety used.
(iii) The seeds are observed after five days or more frequently.
Seed germination is reported as the number of germinated seeds
-------
43
compared to the number planted. The radicle should be 5 mm in
length for a germinated seed.
(2) Recommended references.
(i) Horowitz, M. 1966. A rapid bioassay for PEBC and its
application in volatilization and adsorption studies. Meed Res.
6:22-36.
(ii) Kratky, B.A., and G.F. Warren. 1971. The use of three
simple, rapid bioassays on forty-two herbicides. Weed Res. 11:257-
262.
(iii) Truelove, B., (ed). 1977. Research Methods in Weed
Science. 2nd Ed. Southern Weed Science Society. Auburn Printing
Inc., Auburn, AL 221 pp.
(b) Seedling emergence. (1) Protocol, (i) Seeds may be
germinated in pots using acid-washed sand or a standardized soil.
At least 10 seeds per pot should be used. The seeds may be surface-
sterilized. The soil or support medium is sprayed or otherwise
treated with a known quantity of the chemical. The test conditions
should approximate those optimal conditions for the species and
varieties considered. The seeds should be incubated for at least
14 days. The seeds are observed after 10 and 14 days, and seedling
emergence is recorded as the number of emerged seedlings.
(ii) This test may be extended by 14 days to assess the effect
of soil applied pesticides on vegetative vigor.
(2) Recommended reference.
Truelove, B., (ed). 1977. Research Methods in Weed Science.
Southern Weed Science Society. Auburn Printing Inc., Auburn, AL 221
pp.
(c) Vegetative vigor - foliar spray. (1) Protocol, (i) The
foliar spray can be applied by any acceptable method using labora-
tory-, greenhouse-, or field-grown plants. The plant should be 1 to
4 weeks post-emergent in order to gain young foliage. Types of
sprays and methods of foliar applications may be found in the
reference below. Detrimental effects are to be reported as severity
of phytotoxicity (percent or rating), abnormal changes in growth
and development, and/or abnormal changes in plant morphology as
compared to untreated controls. Direct measurements of height and
weight may also be made and reported.
(ii) Vegetative vigor of seedlings treated with soil-applied
pesticides may be evaluated by extending the period of observation
of the seedling emergence study.
-------
44
(2) Recommended reference. Truelove, B., (ed). 1977.
Research Methods in Weed Science. Southern Weed Science Society.
Auburn Printing Inc., Auburn, AL 221 pp.
(d) Leana gibba; Growth conditions. (1) Species and type.
Lemna gibba G3. Source: Dr. Charles Cleland, Smithsonian Radia-
tion Biology Laboratory, Rockville, MD 20852 (limited supplier)
(2) Protocol. The following are acceptable conditions for the
growth and maintenance of Lemna gibba G3.
(i) Environmental conditions.
Light Intensity: 5 klux (approx. 100 uE m~2s~1)
Light Quality: warm white fluorescent
Photoperiod; continuous light
Thermoperiod: continuous 25 ± 2*C
(ii) Culture conditions.
Liquid culture
Nutrients: M type Hoagland's medium without EDTA or
sucrose (Hillman, 1961 a & b)
pH 5.0 ± 0.1 after autoclaving
(iii) Procedures. The vessel size-to-medium quantity ratio
should be 5 to 2. Maintain the Lemna stock under axenic conditions.
The tests may be performed under non-axenic conditions as long as non-
organic media are used. Sucrose (10 g/1) and EDTA (9 mg/1) may be
added if flowering is desired.
(3) Recommended references.
(i) Davis, J.A. 1981. Comparison of static-replacement and
flow-through bioassays using duckweed, Lemna gibba G3. U.S.
Environmental Protection Agency. Washington DC (EPA 560/6-81-003).
(ii) Hillman, W.S. I961a. Experimental control of flowering
in Lemna XXX. A relationship between medium composition and the
opposite photoperiodic responses 'of L^ perpusyilla 6746 and L^
gibba G3. Amer. J. Bot. 48:413-419.
(iii) Hillman, W.S. 1961b. The Lemnaceae, or duckweeds.
Bot. Rev. 27:221-287.
(e) Selenastrum capricomutum; Growth conditions. (1) Species.
Selenastrom capricomutum Printz. Source: EPA Corvallis Laboratory,
Corvallis, OR 97330
(2) Protocol. The following are acceptable culture conditions
for the growth and maintenance of Selenastrmn capricomutum.
-------
45
(i) Environmental conditions.
Light Intensity: 4 klux (approx. 60 uE m"2s~1)
Light Quality: cool white fluorescent
Photoperiod: continuous light -
Thermoperiod: continuous 24 ± 2*C
(ii) Culture conditions.
Liquid culture
Nutrients: U.S. EPA (1978) medium (EDTA shall not be
used in the experimentation medium.)
pH 7.5
(3) Recommended references.
(i) Environmental Protection Agency, National Eutrophica-
tion Research Program. 1971. Algal Assay Procedure: Bottle Test.
(AAP-.BT). National Environmental Research Center, Corvallis, OR
97330
(ii) Miller, W.E., J.C. Greene, and T. Shiroyama. 1978. The
Selenastrum capricornutum Printz algal assay bottle test. U.S.
Environmental Protection Agency, Corvallis, OR 97330 (EPA 600/9-78-
018).
(iiij organization for Economic Cooperation and Development
(OECD). 1981. Alga, Growth Inhibition Test. OECD Guidelines for
Testing of Chemicals — Ecotoxicology Test No. 201. OECD, Paris,
France.
(f) Skeletonema costatum: Growth conditions. (1) Species.
Skeletonema costatum.
(2) Protocol. The following are acceptable culture conditions
for the growth and maintenance of Skeletonema costatum.
(i) Environmental conditions.
Light intensity: 4 klux (approx. 80 uE m~2s~1)
Light quality: cool white fluorescent
Photoperiod: 16/8 hr day/night
Thermoperiod: 20 +_ 2*C continuous
(ii) Culture conditions.
Liquid culture
Nutrients: Walsh and Alexander (1980) medium
pH 8
-------
46
(3) Recommended references.
(i) U.S. Environmental Protection Agency. 1978. Bioassay
procedures for the ocean disposal permit program. U.S. EPA Labora-
tory, Gulf Breeze, FL 32561 (EPA-600/9-78-010).
(ii) Walsh, G.E., and S.V. Alexander. 1980. A marine algal
bioassay method: Results with pesticides and industrial wastes.
Water, Air, Soil Pollut. 13:45-55.
(g) A Freshwater Diatom; Growth conditions. (1) Species. (To
be selected.)
(2) Protocol. The following are acceptable culture conditions
for the growth and maintenance of Navicula seminulum or other selected
freshwater diatom.
(i) Environmental conditions.
Light intensity: 4.3 klux (approx. 85 uE m~2s-1)
Light quality: cool white fluorescent
Photoperiod: continuous light
Thermoperiod: continuous 24 ^ 20C.
(ii) Culture conditions.
Liquid culture
Nutrients: U.S. EPA (1971) medium
pH 7.5
(3) Recommended reference.
Environmental Protection Agency, National Eutrophication
Research Program. 1971. Algal Assay Procedure: Bottle Test
(AAP:BT). National Environmental Research Center, Corvallis,
OR 97330
(h) Anabaena flos-aquae; Growth conditions. (1) Species.
Anabaena flos-aquae (Lyngb.) DeBrebisson. Source: EPA Corvallis
Laboratory, Corvallis, OR 97330
(2) Protocol. The following are acceptable culture conditions
for the growth and maintenance of Anabaena flos-aquae.
(i) Environmental conditions.
Light intensity: 2 klux (approx. 40 uE m~2s-1)
Light quality: cool white fluorescent
Photoperiod: continuous light
Thermoperiod: continuous 24 ^ 2°C
-------
47
(ii) Culture conditions.
Liquid culture
Nutrients: O.S. EPA (1978) medium (EDTA should not
be used in the experimentation medium.)
pH 7.5 (not to be exceed 8.5)
(3) Recommended references.
(i) Carr, N.G., and B.A. Whitton, eds. 1973. The Biology
of Bluegreen Algae. University of California Press, Berkeley.
676 pp.
(ii) Environmental Protection Agency, National Eutrophication
Research Program. 1971. Algal Assay Procedure: Bottle Test.
(AAP:BT). National Environmental Research Center, Corvallis, OR
97330
(iii) Miller, W.E., J.C. Greene, and T. Shiroyama. 1978. The
Selenastrum capricomutum Printz algal assay bottle test. O.S.
Environmental Protection Agency, Corvallis, OR 97330 (EPA 600/9-78-
018).
-------
48
Series 123: TIER 2 NONTARGET AREA TESTING
§ 123-1 Seed germination/seedling emergence and vegetative vigor
(Tier 2). .
(a) When required. (1) Additional data on the phytotoxic
effects of a pesticide on seed germination/seedling emergence or
vegetative vigor, respectively, are required by 40 CFR Part 158 on
a case-by-case basis when a 25 percent phytotoxic effect to one
or more plant species is noted as a result of the respective Tier
1 tests* These data are required to support the registration of
each end-use product intended for outdoor application.
(2) Portions of this Tier 2 test may be combined with the
respective parts of the Tier 1 test (§ 122-1) and performed as one
test.
(3) See § 120-1(e) concerning substitution of testing and data
submission requirements.
(b) Test standards. In addition to the general test standards
set forth in § 120-3, the test standards for this section shall be
the same as those contained in the Tier 1 studies [§ 122-Kb)] with
the following modifications:
(1) Dosages. The following dosages should be tested: (i) At
least 5 dosages should be tested;
(ii) The dosages should include a subtoxic (
-------
49
is greater than the EC25 for one or more terrestrial plant species
tested. (Tier 3 testing involves evaluation of the pesticide under
field conditions.) See § 124-1.
5 123-2 Growth and reproduction of aquatic plants (Tier 2).
(a) When required. (1) Additional data on the phytotoxic
effects of a pesticide on growth and reproduction of aquatic plants
are required by 40 CFH Part 158 on a case-by-case basis to support
the registration of each end-use product intended for outdoor pesti-
cide application, if the results of the Tier 1 tests required by
§ 122-2 have indicated an adverse effect greater than 50 percent
on growth and reproduction of any aquatic plant.
(2) See § 120-1(e) concerning the substitution of testing and
data submission requirements.
(b) Test standards. In addition to the general test standards
set forth in § 120-3, the test standards for this section shall be
the same as those contained in the Tier 1 studies [§ 122-2(b)] with
the following modifications:
(1) Dosages. The following dosages should be tested: (i) At
least 5 dosages should be tested;
(ii) The dosages should include a subtoxic «EC50) and a
nontoxic concentration;
(iii) The highest dosages should be less than the 1-fold
concentration tested in § 122-2(b)(3); and
(iv) The dosages should be of geometric progression of no more
than 2-fold. For example, the test concentration series may be: 0.1,
0.2, 0.4, 0.8, and 1.6 Xg/ha/15 cm (a 2-fold progression).
(2) Plant species. At least those plant species of Tier 1
[(§ 122-1 (b)(2)] which exhibited phytotoxic effects should be
tested. The use pattern/plant species combinations of § 122-2(b)(2)
should be followed.
(c) Reporting. In addition to the information required by
§ 122-2(c), the test report should include the determination of
the 50 percent detrimental effect level.
(d) Tier progression. Testing at the Tier 3 level is required
if:
(1) The maximum recommended application quantity [where 1 kg/ha
(0.892 Ib/A) equals 0.655 ppm in 15 cm (6") of water] or the antic-
ipated environmental exposure is greater than the EC50 for any one
aquatic plant species tested; and
-------
50
(2) The pesticide is expected to be applied to a fresh water,
estuarine, or marine aquatic system by either direct application or
direct discharge of treated water (except swimming pools), or the
pesticide is to be used within a forest system. (A forest system is
considered equivalent to an aquatic system, since it ordinarily
contains brooks, streams, and rivers. See $ 160-3(c), (d), and (e)
of Subdivision N for full explanation of pesticide aquatic use
patterns.) See § 124-2 (Tier 3) where evaluation of the pesticide
under field conditions is employed. Pesticides with terrestrial
uses only need not be tested.
-------
51
Series 124: TIER 3 NONTARGET AREA TESTING
§ 124-1 Terrestrial field testing (Tier 3)
(a) When required. (1) Data on the phytotoxic effects of
the end-use product on seed germination, vegetative vigor, and
reproduction potential under field use conditions are required by
40 CFR Part 158 on a case-by-case basis to support the registration
of each end-use product intended for outdoor application. The
maximum recommended application quantity or anticipated environ-
mental exposure is to be equal to or greater than the EC25 for one
or more terrestrial plant species as found in the Tier 2 tests
(§ 123-1).
(2) The data requirements of this section need not be ful-
filled for pesticides applied by systems where the chemicals are
not readily released into the environment. Examples of these
systems are: tree injection, subsurface soil applications, recap-
ture systems, and wick applications.
(3) See § 120-1(e) concerning substitution of testing and data
requirement submission.
(b) Test standards. In addition to the general test standards
set forth in § 120-3, the test standards for this section shall be
the same as those contained in § 122-Kb) of this subdivision, with
the following modifications:
(1) Test substance. The test substance shall be the end-use
product or a representative end-use product from the same major
formulation category for that general use pattern. Examples of
major formulation categories are: wettable powders, eamlsifiable
concentrates, and granulars. (If the manufacturing-use product is
usually formulated into end-use products comprising two or more
major formulation categories, a separate study must be performed
with a typical end-use product for each category.)
(2) Application levels. The dosages tested should be the same
as those employed in the Tier 2 test [§ 123-1(b)(1)].
(3) Species, (i) Representatives of the following plant
groups are to be tested, subject to the limitations of paragraph
(iii) below:
(A) Dicotyledonae (dicots), representatives of three families;
(B) Monocotyledonae (monocots), representatives of three
families;
-------
52
(C) Vascular Cryptogaaae (ferns and allies), representatives of
two families;
(D) Bryophyta (mosses) or Hepatophyta (liverworts), one repre-
sentative (for wetland use patterns only); and
(E) Gymnospermae (conifers), one representative.
(ii) Plant species used for testing Tiers 1 and 2 can be used
to satisfy the monocot or dicot test plant requirements of this
section.
(iii) If any of the plant groups are not likely to be exposed
to the pesticide under normal conditions of use, testing of such
groups is not required. Justification for elimination of a test
species or group should be included in the test report.
(iv) Additional plant species may be required if the general
selectivity of the pesticide cannot be readily identified.
(4) Test conditions* ' Plants are to be grown under field-use
conditions similar to those of the natural habitat of the plants in
use.
(5) Duration. The test duration should be of sufficient length
to assess multiple applications directed by the label. Observations
should continue for at least two weeks after the last application and
for a maximum of four weeks to note any recovery or death.
(6) Season of application. The test substance is to be applied
over a period of time or season according to the proposed label
instructions.
(7) Test locations. The pesticide should be tested in those
geographic locations where it is expected to be used, as based on
proposed label use sites. Where important species diversity and
physiographic differences occur within a region of intended applica-
tion, regional testing may be inadequate, and testing at a more
specific region or biome level may be required. United States
regional areas of potential testing include:
Northeastern temperate deciduous;
Southeastern temperate deciduous;
Northern grassland (prairie);
Southern grassland (prairie);
Northwestern (and Alaskan) conifer forest and high desert;
Southwestern chaparral Mediterranean and low desert; and
Hawaiian and Caribbean tropical regions.
(c) Reporting. In addition to the information required in
§§ 120-4 and 122-1(c) of this subdivision, the test report should
-------
S3
include the test conditions employed (including the soil and
environmental conditions) end the determination of the 50 percent
detrimental effect level. *
124-2 Aquatic field testing (Tier 3).
(•) When required. (1) Data on the phy to toxic effect! of the
product on growth and reproduction of an expended number of aquatic
plants are required by 40 CFR Part 158 on a case-by-case basis to
support the registration of each end-use product intended for outdoor
pesticide application, when*
(i) The anticipated environmental exposure is greater than the
ECSO for any one aquatic plant species tested in Tier 2 tests (f 123-
2)i and
(ii) The pesticide is expected to be applied to a fresh water,
estuarine, or marine aquatic system by either direct application or
direct discharge of treated water (except svimning pools), or the
pesticide is to be used within a forest system. [See f 160-3(c),
(A), and (e) of Subdivision N for description* of these aquatic uses.)
Pesticides with only terrestrial uses need not be tested.
(2) See { 120-1(e) concerning substitution of testing and data
requirements submission.
(b) Teat standards. In addition to the general test standards
set forth in | 120-3 of this subdivision, the test standards for this
section shall be the same as those in ) 122-2(b), with the following
modifications:
(1) Test substance. The test substance shall be the end-use
product or a representative end-use product from the same major
formulation category for that general use pattern. Examples of
major formulation categories are: wettable powders, eaulsifiable
concentrates, and granular*. (If the manufacturing-use product is
usually formulated into end-use products comprising two or More major
formulation categories, a separate study must be performed with a
typical end-use product for each category.)
(2) Application levels. The dosages tested should be the same
aa those specified in the Tier 2 aquatic test standards [| 123-
(3) Species, (i) Aquatic plant representatives of the
following plant groups are to be testedi
(A) Dleotyledonae (dicots), one representative!
-------
54
(B) Monocotyledorae (nonocots), representatives of three
families;
(C) Vascular Cryptogamae (ferns and allies), representatives
of three families;
(D) Algae (including Cyanophyta), • representative of each
Division) and
(E) Bryophyta (mosses) or Hepatophyta (liverworts), one
representative (not required for true aquatic use patterns, rather
for wetland use patterns).
(ii) Plant species used for testing Tiers 1 and 2 can be used
to satisfy the monocot and dicot test plant requirements of this
section.
(ill) Additional plant species nay be required if the general
selectivity of the pesticide cannot be readily identified*
(4) Environmental conditions, (i) Plants may be grown in
either native soil, water, or other substrate of similar nature to
that of the indigenous area or under other conditions similar to the
natural habitat.
(11) Reduction of light intensity by natural or constructed
light shade may be necessary to simulate the reduced light inten-
sities found with certain plant communities such as deeply submerged
eitea or shaded waters.
(ill) Other natural conditions should also be maintained where
plants are removed from their natural habitat. Soil, water, and air
temperatures should approximate those of the natural habitat. For
estuarine and marine habitats, the following conditions should, to
the extent possible, simulate the natural environment: tidal action,
water turbidity, flow rates, salinity, and degree of exposure.
(Iv) Tests should be performed either in enclosed, controlled
areas of a lake, pond, or swamp, or in large water cultures such as
aquaria or plastic wash tubs. Tests are not to be performed in
dynamic or flowing water where the release of the chemical cannot be
contained or its escape prevented.
(v) The field studies should be conducted using:
(A) Acceptable protocols as may be found in the following
recommended reference:
True love, B., 1977, Research Methods in Weed Science, 2nd Ed.
Southern Weed Science Society, Auburn Printing Inc., Auburn, ALi or
-------
55
(B) A protocol with prior approval of the Agency*
(5) Duration. The test duration should be of sufficient
length to assess aultlple applications directed by the label*
Observation* should continue for at least two weeks after the last
application and for a aaxiaum of four weeks to note any recovery
or death.
(6) Season of application. The test substance is to be Ap-
plied over the period of tiste or eeaeon according to the proposed
label instructions.
(7) Test locations. The pesticide should be tested in those
geographic locations where it is expected to be used, as based on
proposed label use sites. Where important species diversity and
physiographic differences occur within a region of intended appli-
cation, regional testing may be inadequate, and testing st a nore
specific region or bloae level say be required. United States
regional areas of potential testing includei
Northeastern tenperate deciduous;
Southeastern temperate deciduous/
Northern grassland (prairie))
Southern grassland (prairie);
Northwestern (and Alaskan) conifer forest and high desert)
Southwestern chaparral Mediterranean and low desert; and
Hawaiian and Caribbean tropical reglona.
(c) Reporting. In additi-. to the information required by
H 120-4 and 122-2(c) of this subdivision, the test report should
Include the test conditions (including soil, water, and environ-
mental conditions) and the determination of the 50 percent detri-
mental effect level.
-------
-------
APPENDIX B PB87-1022UO
EPA 540/9-86-132
June 1986
HAZARD EVALUATION DIVISION
STANDARD EVALUATION PROCEDURE
NON-TARGET PLANTS:
SEED GERMINATION/SEEDLING EMERGENCE
TIERS 1 AND 2
Prepared by
Robert W. Hoist, Ph.D.
Standard Evaluation Procedures Project Manager
Stephen L. Johnson
Hazard Evaluation Division
Office of Pesticide Programs
United States Environmental Protection Agency
Office of Pesticide Programs
Washington, D.C. 20460
REPRODUCED BY
U S. DEPARTMENT OF COMMERCE
NATIONAL TECHNICAL
INFORMATION SERVICE
SPRINGFIELD. VA. 22161
-------
TABLE OF CONTENTS
Paqe
I. INTRODUCTION
A. Purpose of the Standard Evaluation
Procedure 1
B. Background Information 1
C. Objective of Seed Germination/Seedling
Emergence Tests 1
1. Tier 1 Test 1
2, Tier 2 Test 2
II. INFORMATION TO BE SUPPLIED 2
III. DATA INTERPRETATION 2
IV. THE DATA EVALUATION PROCESS
A. Identify Data Gaps 3
B. Assess the Appropriateness and Adequacy
of the Data 3
D. Report Preparation 4
D. Conclude if the Requested Action is
Supportable 4
V. APPENDICES
Appendix 1: Information Requested of the
Registrant 5
Appendix 2: Specific Questions for the
Reviewer 9
Appendix 3: Sample Standard Format for
Preparation of. Scientific
Rev i ews 12
REFERENCES 13
-------
NON-TARGET PLANTS;
SEED GERMINATION/SEEDLING EMERGENCE - TIERS 1 AND 2
I. INTRODUCTION
A. Purpose of the Standard Evaluation Procedure
This Standard Evaluation Procedure is designed to aid Ecologi-
cal Effects Branch (EEB) data reviewers in their evaluations of
preliminary (Tier 1) laboratory seed germination/seedling emergence
studies submitted by registrants in the assessment of pesticide
effects on non-target plants. This document is also designed to aid
EEB reviewers in their evaluations of laboratory/greenhouse/small
field plot (Tier 2) seed germination/seedling emergence s-tudies sub-
mitted by registrants for the same purpose. s
B. Background Information
Seed germination/seedling emergence studies are designed to pro-
vide phytotoxicity data on a pesticide. These phytotoxicity data are
needed to evaluate the effect of the level of pesticide exposure to
non-target and terrestrial plants and to assess the impact of pesti-
cides on endangered and threatened plants as noted under the Endan-
gered Species Act. The preliminary level (Tier 1) study evaluates
the effect of the maximum exposure level while the greenhouse/labora-
tory/small field plot (Tier 2) study evaluates the effects of differ-
ing exposure levels. Where a phytotoxic effect is noted in one or
more plants, further seed germination/seedling emergence studies may
be required. These studies are required by 40 CFR § 158.150 to sup-
port the registration of any pesticide intended for outdoor use under
the federal Insecticide, Fungicide and Rodenticide Act (FIFRA), as
amended.
Pesticides with outdoor use patterns that do not readily release
the pesticide to the environment do not have to be evaluated using
this phytotoxicity test. These use patterns include tree injection,
subsurface soil applications, recapture systems, wick applications,
and swimming pool uses. If any of these use patterns do readily
expose non-target plants to the pesticide, as through vapors, the
pesticide phytotoxicity potential may need to be evaluated.
C. Objective of Seed Germination/Seedling Emergence Tests
1. Tier 1 Test
The objective of the Tier 1 seed germination/seedling emergence
test is to determine if a pesticide exerts a detrimental effect to
plants during critical stages in their development. The test is
performed on species from a cross-section of the non-target terres-
trial plant population that have been historically used for this type
-------
-2-
of testing and, therefore, have known types of responses. This is
a maximum aose test designed to quickly evaluate the phytotoxic
effects of the pesticide at the one dose.
2, Tier 2 Test
The objective ot the Tier 2 seed germination/seedling emergence
test is to determine it a pesticide exerts a detrimental effect to
plants during critical stages in their development. The test is per-
formed on species trom a cross-section of the non-target terrestrial
plant population that have been historically used for this type of
testing and, therefore, have known types of responses. This is a
multiple dose test designed to evaluate the phytotoxic effects of
the pesticide over a wide range of anticipated pesticide quantities
as may be found in the environment.
II. INFOKMATIUN TO BE SUPPLIED
The registrant's report on preliminary seed germination/seed-
ling emergence studies should include all information necessary to
provide: 1) a complete and accurate description of the laboratory/
greenhouse treatments and procedures, 2) sampling data and pnytotox-
icity rating, 3) data on storage of the plant materials until analy-
sis, if so performed, 4) any chemical analysis of the plant material
as to chemical content, if so performed, 5) reporting of the data,
rating system and statistical analysis, and 6) quality control mea-
sures/precautions taken to ensure the fidelity of the operations.
A guideline of specific information that should be included in
the registrant's report on seed germination/seedling emergence
studies is provided in Appendix 1 of this document. The lists of
requested information and reviewer aids are derived from the Pesti-
cide Assessment Guidelines, Subaivision J: Hazard Evaluation of
Non-Target Plants, which is complemented by this Standard Evaluation
Procedure.
III. DATA INTEKPRfTATlON
The acceptability of the study results will depend upon whether
the test requirements/standards are followed. If a deviation is
made, a determination must be made as to whether the deviation has
changed the quality of the results in such a manner that the results
cannot be extrapolated to the natural environment. There should be
little or no deviation from the liberal standards prescribed in this
study.
The results of the pesticide phytotoxicity tests with respect
to the quantity of material applied to or near the seed are important.
The concentration of the chemical in the carrier is important in that
even slightly stronger concentrations than normally used can lead to
-------
-3-
stunting and necrosis. Subtoxic concentrations, on the other hand,
may cause unwanted rapid growth.
Plants can recover from certain types of injury with little or
no resulting effect on the esthetic or economic value of the plant(s)
tested or upon which an evaluation is made. Therefore, it is impor-
tant that a minimum of two weeks of observations be made after appli-
cation of the pesticide to evaluate seedling emergence. If seed ger-
mination is evaluated, the extent of germination (percentage of seed
showing root and shoot emergence) should be evaluated at least five
days after imbibition.
A decision point to proceed to the next higher test is a 25%
detrimental effect, i.e., a 25% change in the average germination
or plant growth or injury as compared to untreated controls. This
level is considered to be that point at which the plants will not
recover to their full esthetic value, economic value or reproductive
potential as in the case of the maintenance of the endangered or
threatened species.
IV. THE DATA EVALUATION PROCESS
Upon careful examination of the information/data supplied by
the registrant in his submission to the Agency, the reviewer shall
evaluate the data as follows.
A. Identify Data Gaps
Using Appendix 1 of this document as a guide, the reviewer
should then look tor data gaps - omissions in the information sup-
plied by the registrant in his report. These should be duly noted
in the reviewer's report, and a judgment made as to which are con-
sidered significant enough to adversely affect the review process.
. Those so identified should be communicated back to the registrant
by the Product Manager for corrective action.
B. Assess the Appropriateness and Adequacy of the Data
The data reviewer then considers the appropriateness, i.e., the
intended use pattern, and adequacy of the data/information that has
been supplied. Appendix 1 of this document is a useful guide to the
various parameters that need to be considered. Appendix 2 provides
specific questions that should be answered by the reviewer during
the study evaluation process. Statistical treatments of the data
should be independently verified and the quality control precautions
noted.
As an adjunct to these, the reviewer should draw upon the tech-.
nical guidance in the reviewer aids materials that are available.
(See also the recommended references in Subdivision J - Hazard Eval-
uation; Non-Target Plants.) A listing of additional source materials
is located in the References section of this document.
-------
-4-
In addition to the data gaps noted above, any perceived defici-
encies in the data/information supplied should also be identified.
A statement as to these deficiencies should be made in the reviewer's
report and corrective action to resolve them should be provided.
This information can be relayed to the registrant by the Product
Manager for appropriate action.
C. Report Preparation
The Agency reviewer prepares a standard review report following
the standard format for preparation of scientific reviews as provided
in Appendix 3 of this document. All important information provided
by the registrant including the methodology and results should be
summarized in order that future evaluations can be made. The
results may be expressed in the form of tables where specific
values are related. Figures (graphs) may be provided but are not
to be the sole source of the values needed for future evaluations.
D. Conclude if the Requested Action is Supportable
Lastly, the reviewer considers the results of the seed germi-
nation/seedling emergence studies and makes a judgment as to whether
they support the requested registration action of the data submitter.
If the data are not supportive, possible alternative action(s) that
may be taken by the registrant, such as label ;aodif ications, are sug-
gested. If deficiencies/omissions exist in th" submitted data, the
reviewer may have to defer judgment until such time as appropriate
corrective action has been rendered by the registrant.
-------
-5-
APPENDIX 1
INFORMATION REQUESTED OF THE REGISTRANT
The registrant's report on preliminary seed germination/seed-
ling emergence studies should include all information necessary to
provide: 1) a complete and accurate description of the laboratory/
greenhouse/small field plot treatments and procedures, 2) sampling
and phytotoxicity rating, 3) data on storage of the plant material
until analyzed, if so performed, 4) any chemical analysis of the
plant material as to chemical content, if so performed, 5} reporting
of the data, rating system and statistical analysis, and 6) quality
control measures/precautions taken to ensure the fidelity of the
operations.
Specifically, each laboratory/greenhouse/small field plot seed
germination/seedling emergence report should include the following
information.
I. General
0 Cooperator or researcher (name ^nd address), test location
(county and state; country, if outside of the U.S.A.), and date of
study;
0 Name (and signature), title, organization, address, and
telephone number of the person(s) resprr.sible for planning/supervising/
monitoring and, for the field plot stu' ies, applying the pesticide;
0 Trial identification number;
0 Quality assurance indicating: control measures/precautions
followed to ensure the fidelity of the phytotoxicity determinations;
record-keeping procedures and availability of logbooks; skill of
the laboratory personnel; equipment status of the laboratory or
greenhouse; degree of adherence to yr>r,d laboratory practices; and
degree of adherence to good agricultural practices in maintaining
healthly plants; and
0 Other information the registrant considers appropriate and
relevant to provide a complete and t-iorough description of the test
procedures and results.
II. Test Substance (Pesticide)
0 Identification of the test pesticide active ingredient (ai)
including chemical name, common name (ANSI, BSI, ISO, WSSA), and
Company developmental/experimental r. -me;
-------
-6-
0 Active ingredient percentage in the technical grade material
or in the manufacturing-use product, if the technical grade material
is unavailable for test purposes;
0 solvent used to dissolve and apply the pesticide if the pesti-
cide is insoluble in water or other intended carrier;
0 Dose rate(s) in terms of
or concentration as applied;
active ingredient per area of land
0 For Tier 1, dose rate(s) in terms of the maximum label rate,
or if the registrant has shown that the maximum quantity that will
be present in the non-target area is significantly less than the
maximum label rate, the dose equal to or no less than three times
that maximum environmental quantity;
0 For Tier 2, dose rate(s) in terms of less than the maximum
label rate, with dosages in a geometrical progression of no more
than two-fold and with subtoxic (< EC$Q level) and non-toxic (no-
observable-effect-level) concentrat ions;
0 Method of application including equipment type; and
0 Number of applications.
III. Plant Species
0 For Tier 1, identification of the six dicotyledoneae species
and four monocotyledoneae species with family identification. The
six dicots are to be of at least four different families and the
moncots of at least two families. Soybeans, corn, and a dicot root
crop like carrot are the required species. The proposed species
and families as originally provided in Subpart J of the proposed
guidelines [FR notice of 3 November 1980J are given below and are
acceptable for the laboratory/greenhouse seed germination/seedling
emergence test:
Family
Solanaceae
Cucurbitaceae
Compositae
Leguminosae
Cruciferae
Umbelli ferae
Gramineae
Gramineae
Species
Lycopersicon esculentum
Cucumis sativus
Lactuca sativa
Glycine max
(Innoculation
unnecessary)
with Rhizobium
Commo n
Tomato
Cucumber
Lettuce
Soybean
japonicum is
Brassica oleracea
Daucus carota
Avena sativa
Lolium perenne
Cabbage
Carrot
Oat
Perennial
Ryegrass
-------
-7-
Family Species Common
Gramineae Zea mays Corn
Amaryllidaceae Allium cepa Onion
Seeds of plants with a low or variable germination potential should
be avoided for the seed germination study.
e For Tier 2, identification of the plant species tested in-
cluding those phytotoxically affected in the Tier 1 test;
0 Identification of the cultivar(s) of the plant species or
assignment of an identification number to the cultivar used and
seed or plant source;
0 Identification of the number of replicates and the number
of plants per replicate per dose; and
0 Identification of the date of planting or imbibition, date
ot pesticide application, and date of phytotoxicity rating or
harvest and analysis.
IV. bite of the Test
0 bite description of the seed germination/seedling emergence
study such as the type of growth chamber, greenhouse, or field
(small field plots);
0 Location of the test site;
0 Climatological data during the test (records of applicable
conditions for the type of site, i.e., temperature and thermoperiod,
rainfall or watering regime, light regime - intensity and quality,
relative humidity, wind speed);
0 field lay-out (for small field plots), e.g., size and number
of control and experimental plots; number of plants per plot/unit
area;
0 Pot, plant or row density of seeds or plants;
0 Cultural practices such as cultivation and irrigation; and
0 Substrate characteristics (name/designation of soil type and
its physical and chemical properties, including pH and percent
organic matter).
V. Results
0 Reporting of percent germination/emergence, root length or
other growth parameters that may have been measured to ascertain
-------
-8-
toxic effects of
vat ions;
the pesticide upon the plants with dates of obser-
0 Phytotoxicity rating (including a description of
system) for each plant or population in the test; and
the rating
0 Statistical analysis of the results including an environ-
mental or effective concentration (EC) value. (Note, for Tier 1,
there will be only a percent effect level at a specific concentration
which is then compared to 25% of the growth [mass or rate] of the
control.)
VI. Evaluation
0 For Tier 1 studies, determination as to whether Tier 2
studies would be required due to phytotoxic effects noted in one or
more of the tested species.
0 For Tier 2 studies, determination as to whether'Tier 3 tests
(terrestrial field study) would be required due to phytotoxic effects
noted in one or more of the tested species.
-------
-9-
APPENDIX 2
SPECIFIC QUESTIONS FDR THE REVIEWER
The following questions are provided to aid the reviewer in
performing the standard evaluation procedure in a scientific manner
and in acquiring the necessary information to complete a standard
format for preparation of scientific reviews.
I. General
0 Was the name of the cooperator or researcher (name and
address), test location (county and state; country, if outside of
the U.S.A.), and date of study provided?
0 Was the name (and signature), title, organization, address,
and telephone number of the person(s) responsible for planning/super-
vising/monitoring and, for small field plot studies, applying the
pesticide provided?
0 Was the trial identification number provided?
0 Were quality assurance control measures/precautions indicated?
0 Was the Tier 1 seed germination/seedling emergence study done
a••• a separate study? If not, were the doses and plant species re-
quired by Tier 1 included in the Tier 2 study?
11. Test Chemical
0 Is the test chemical being used the technical grade, or if
not available, the manufacturing-use product with the highest
percentage of active ingredient?
0 Is the active ingredient percentage or degree of purity of
the chemical given?
0 If a solvent was used, was it used at concentrations that
are not phytotoxic and was a solvent control used?
0 Is the dose given in quantity per unit area (of plant or
land surface) or in tank concentration?
0 for Tier 1, was the dose equal to or greater than the maxi-
mum lab*; rate, or if the registrant has shown that the maximum
quantity that will be present in the non-target area is significantly
less than the maximum label rate, was the dose equal to or no less
than three times that maximum environmental quantity?
-------
-10-
0 For Tier 2, was the maximum dose less than the maximum label
rate?
0 For Tier 2, were the additional dosages of a geometric pro-
gression of no more than two-fold, e.g., 0.1, 0.2, 0.4, 0.8, 1.6 kg/ha?
0 For Tier 2, were a subtoxic (< EC$Q level) and a non-toxic
(no-observable-effeet-level) concentration evaluated?
III. Test Species
8 For Tier 1, were at least ten different species tested with
species names provided?
0 For Tier 1, were the ten species split between monocots and
dicots, four and six, respectively?
0 For Tier 1, were the ten species from six different families
and the family names provided?
0 For Tier 1, were two of the species tested soybeans and corn
and was the third species a dicot root crop?
0 For Tier 2, were at least those species that were phytotox-
ically affected in Tier 1 tested?
0 Where various cultivars could be used, such as in the case
of most agronomic and horticultural plants, were cultivar or varietal
navies provided?
0 Were seed and plant sources provided?
0 Were at least three replicates used with ten seeds per repli-
cate for each dose level?
0 Were some of the seeds pretested for germination and emer-
gence potential? seeds of plants with a low or variable potential
should be avoided.
0 Were endangered or threatened plant species not used?
IV. Test Procedures
0 Was the test site specified, i.e., greenhouse, growth cham-
ber, or small field plot?
0 Were the environmental conditions that prevailed during the
test (temperature and thermoperiod, light regime - intensity and
quality, rainfall or watering regime, relative humidity, wind)
provided as appropriate for the site?
-------
-11-
0 Were the environmental conditions that prevailed during the
test those most favorable and roost typical to the growth of. the
plants used? Were these conditions referenced?
0 was the test duration for seedling emergence at least two
weeks in length or for seed germination at least five days in length?
0 Were observations taken at least weekly for seedling emer-
gence and after the five days for seed germination?
0 Was the method of pesticide application including the type
of application equipment employed given?
V. Reporting
0 were the detrimental effects reported as severity of phyto-
toxicity (rating or percentage), percent germination or percent
emergence?
0 If a rating system was used, was an explanation provided?
0 Were abnormal changes in growth, development and/or morpho-
logy reported with comparisons to the controls or "normal" plants?
0 Though not required, were direct measurements of root
length or seedling length provided?
0 Were the results statistically analyzed? Note that care
should be taken in interpreting the statistical results where the
sample size is small.
VI. Evaluation
0 were the results tabulated to indicate a percentage effect
level (EC value) tor each species as compared to the untreated
control plants?
0 for Tier 1 studies, was a determination made as to whether
Tier 2 tests should be performed if any of the Tier 1 species were
detrimentally affected (greater than 25% detrimental effect on
growth)?
0 For Tier 2 studies, were 25 and 50 percent detrimental effect
levels determined for those plant species of Tier 1 that showed a
phytotoxic effect to the chemical?
0 For Tier 2 studies, was a determination made as to whether
Tier 3 tests (terrestrial field study) should be performed if any of
the Tier 2 species were detrimentally affected (greater than 25% de-
trimental effect on growth)?
-------
-12-
APPENDIX 3
SAMPLE STANDARD FORMAT FOR PREPARATION OF SCIENTIFIC REVIEWS
The following format shall be used in documenting the review
of the subdivision J - Hazard Evaluation; Non-Target Plants - Seed
Germination/Seedling Emergence Tier 1 and Tier 2 Studies.
Chemical: (Common Name)
Formulation: (Percent Active Ingredient)
study/Action: (Purpose of the Submission)
Study Identification:
Reviewer:
Approval:
Conclusions
(Subdivision J Test Title)
(Reference or Registrant Data Information with
Study Number)
(EPA Accession Number)
(Name and Address of Reviewer; Date of Review)
(Quality Control Reviewer)
(Summary and Conclusion of Tests)
AcceptaDility and Recommendations:
Background:
Discussion:
(Decide as to (1) the scientific validity of the study
and (2) compliance to the Subdivision J - Seed Germi-
nation Tier 1 and Tier 2 Studies.)
(Introductory Information and Directions for Use)
1. Study Identification
2. Materials and Methods
3. Reported Results
4. Reported Conclusions
5. Reviewer's Interpretation of Results and Conclusion
-------
-13-
REFERENCES
Bewley, J. D. 1983. Physiology and Biochemistry of Seeds in Relation
to Germination.
Khan, A. A. 1977. Physiology and Biochemistry of Seed Dormancy and
Germination.
Mayer, A. M. 1982. Germination of Seeds.
Truelove, B., ed. 1977. Research Methods in Weed Science. Southern
weed science Society. Auburn, AL: Auburn Printing, Inc.
U.S. Department of Agriculture. 1952. Manual for Testing Agricul-
tural and Vegetable Seeds. Agriculture Handbook No. 30.
Other scientific articles of seed germination may be found in
the following journals:
Agronomy Journal
Environmental Science and Technology
Journal of Environmental Quality
Soil Science and Plant Nutrition
Weed Science
-------
APPENDIX C
Protection of
Environment
40
PARTS 150 to 189
Revised as of July 1, 1990
CONTAINING
A CODIFICATION OF DOCUMENTS
OF GENERAL APPLICABILITY
AND FUTURE EFFECT
AS OF JULY 1, 1990
With Anci/laries
Published by
the Office of the Federal Register
National Archives and Records
Administration
as a Special Edition of
the Federal Register
-------
Port 160
40 CFR Ch. I (7-1.90 Edition)
Specific use patterns—listed
according to use site group
Military uses—not specified
Quarantine uses—not specified
DHHS/FDA uses-not specified
Fillers (air conditioning, air. and fur-
nace)
Biological specimens
Underground cables
Cuspidors, spittoons
Vomrtus
Human wastes
Air sanrtirers
Diapers
Laundry equipment (carls, chutes,
tables, etc.)
Oust control—products and equip-
ment (mops, etc.)
Dry cleaning
Carpets
Upholstery
Bathrooms, toHets bowls, and related
sites
Bathroom premises
Toilet bowls and urinals
Toilet tanks
Portable toilets, chemical toilets
Vehicular holding tanks
Bathroom air treatment
Diaper pails
Refuse and soild waste
Reluse and solid waste containers
Refuse and solid waste transporta-
tion and handling equipment
Garbage dumps
Household trash compactors
Garbage disposal units, food dispos-
als
Incinerators
14. Miscellaneous Indoor Uses
Surface Treatments
Hard nonporous surfaces (painted.
tile, plastic, metal, glass, etc.)
Hard porous surfaces (cement, plas-
ter)
Camping equipment and gear
Grooming instruments (brushes,
clippers, razors, etc.) %,
Laundry, cleaning, and dry cleaning
Corresponding
general use pattern
Indoor
PART 160—GOOD LABORATORY
PRACTICE STANDARDS
Subparl A—General Provisions .
Sec.
160.1 Scope.
160.3 Definitions.
160.10 Applicability to studies performed
under grants and contracts.
160.12 Statement of compliance or non-
compliance.
160.15 Inspection of a testing facility.
160.17 Effects of non-compliance.
Sec.
Svbperf B—Organization and Personnel
160.29 Personnel.
160.31 Testing facility management.
160.33 Study director.
160.35 Quality assurance unit.
Subparl C—Facilities
160.41 General.
160.43 Test system care facilities.
160.45 Test system supply facilities.
160.47 Facilities for handling test, control.
and reference substances.
160.49 Laboratory operation areas.
160.51 Specimen and data storage facilities.
Subparl D—Equipment
160.61 Equipment design.
160.63 Maintenance and
equipment.
calibration of
Subparl E—Testing Facilities Operation
160.81 Standard operating procedures.
160.83 Reagents and solutions.
160.90 Animal and other test system care.
Subparl F—Test, Control, and Reference
Substance*
160.105 Test, control, and reference sub-
stance characterization.
160.107 Test, control, and reference sub-
stance handling.
160.113 Mixtures of substances with carri-
ers.
Subparl 6—Protocol for and Conduct of a
Study
160.120 Protocol.
160.130 Conduct of a study.
160.135 Physical and chemical character-
ization studies.
Subparls H—I [Reserved]
Subparl J—Records and Reports
160.185 Reporting of study results.
160.190 Storage and retrieval of records
and data.
160.195 Retention of records.
AUTHORITY: 7 U.S.C. 136a, 136c, 136d. 136f,
136J, 136t, 136v, 136w: 21 U.S.C. 346a, 348,
371, Reorganization Plan No. 3 of 1970.
SOURCE: 54 FR 34067, Aug. 17, 1989. unless
otherwise noted.
142
-------
Environmental Protection Agency
§160.3
Subpart A—General Provisions
6160.1 Scope.
(a) This part prescribes good labora-
tory practices for conducting studies
that support or are intended to sup-
port applications for research or mar-
keting permits for pesticide products
regulated by the EPA. This part is in-
tended to assure the quality and integ-
rity of data submitted pursuant to sec-
tions 3. 4. 5, 8. 18 and 24(c) of the Fed-
eral Insecticide. Fungicide, and Roden-
ticide Act (FIFRA). as amended (7
U.S.C. 136a, 136c. 136f. 136q and
136v(c)) and sections 408 and 409 of
the Federal Food, Drug and Cosmetic
Act (FFDCA) (21 U.S.C. 346a. 348).
(b) This part applies to any study de-
scribed by paragraph (a) of this sec-
tion which any person conducts, initi-
ates, or supports on or after October
16. 1989.
§160.3 Definitions.
As used in this part the following
terms shall have the meanings speci-
fied:
Application for research or market-
ing permit includes:
(1) An application for registration.
amended registration, or reregistration
of a pesticide product under FIFRA
sections 3, 4 or 24(c).
(2) An application for an experimen-
tal use permit under FIFRA section 5.
(3) An application for an exemption
under FIFRA section 18.
(4) A petition or other request for es-
tablishment or modification of a toler-
ance, for an exemption for the need
for a tolerance, or for other clearance
under FFDCA section 408.
(5) A petition or other request for es-
tablishment or modification of a food
additive regulation or other clearance
by EPA under FFDCA section 409.
(6) A submission of data in response
to a notice issued by EPA under
FIFRA section 3(c)(2)(B).
(7) Any other application, petition,
or submission sent to EPA intended to
persuade EPA to grant, modify, or
leave unmodified a registration or
other approval required as a condition
of sale or distribution of a pesticide.
Batch means a specific quantity or
lot of a test, control, or reference sub-
stance that has been characterized ac-
cording to { 160.105(a).
Carrier means any material, includ-
ing but not limited to feed, water, soil,
nutrient media, with which the test
substance is combined for administra-
tion to a test system.
Control substance means any chemi-
cal substance or mixture, or any other
material other than a test substance,
feed, or water, that is administered to
the test system in the course of a
study for the purpose of establishing a
basis for comparison with the test sub-
stance for known chemical or biologi-
cal measurements.
EPA means the U.S. Environmental
Protection Agency.
Experimental start date means the
first date the test substance is applied
to the test system?
Experimental termination date
means the last date on which data are
collected directly from the study.
FDA means the U.S. Food and Drug
Administration.
FFDCA means the Federal Food,
Drug and Cosmetic Act, as amended
(21 U.S.C. 321 etseq).
FIFRA means the Federal Insecti-
cide, Fungicide and Rodenticide Act as
amended (7 U.S.C. 136 et seq).
Person includes an individual, part-
nership, corporation, association, sci-
entific or academic establishment, gov-
ernment agency, or organizational
unit thereof, and any other legal
entity.
Quality assurance unit means any
person or organizational element,
except the study director, designated
by testing facility management to per-
form the duties relating to quality as-
surance of the studies.
Raw data means any laboratory
worksheets, records. memoranda,
notes, or exact copies thereof, that are
the result of original observations and
activities of a study and are necessary
for the reconstruction and evaluation
of the report of that study. In the
event that exact transcripts of raw
data have been prepared (e.g., tapes
which have been transcribed verbatim,
dated, and verified accurate by signa-
ture), the exact copy or exact tran-
script may be substituted for the origi-
nal source as raw data. "Raw data"
may include photographs, microfilm
143
-------
§ 160.10
40 CFR Ch. I (7-1-90 Edition)
or microfiche copies, computer print-
outs, magnetic media. Including dictat-
ed observations, and recorded data
from automated instruments.
Reference substance means any
chemical substance or mixture, or ana-
lytical standard, or material other
than a test substance, feed, or water.
that is administered to or used In ana-
lyzing the test system In the course of
a study for the purposes of establish-
ing a basis for comparison with the
test substance for known chemical or
biological measurements.
Specimens means any material de-
rived from a test system for examina-
tion or analysis.
Sponsor means:
(DA person who Initiates and sup-
ports, by provision of financial or
other resources, a study;
(2) A person who submits a study to
the EPA in support of an application
for a research or marketing permit; or
(3) A testing facility, if it both initi-
ates and actually conducts the study.
Study means any experiment at one
or more test sites, in which a test sub-
stance is studied in a test system
under laboratory conditions or In the
environment to determine or help pre-
dict its effects, metabolism, product
performance (efficacy studies only as
required by 40 CFR 158.640), environ-
mental and chemical fate, persistence
and residue, or other characteristics in
humans, other living organisms, or
media. The term "study" does not In-
clude basic exploratory studies carried
out to determine whether a test sub-
stance or a test method has any poten-
tial utility.
Study completion date means the
date the final report is signed by the
study director.
Study director means the individual
responsible for the overall conduct of
a study.
Study initiation date means the
date the protocol Is signed by the
study director.
Test substance means a substance or
mixture administered or added to a
test system In a study, which sub-
stance or mixture:
(1) Is the subject of an application
for a research or marketing permit
supported by the study, or is the con-
templated subject of such an applica-
tion; or
(2) Is an Ingredient, Impurity, degra-
dation product, metabolite, or radioac-
tive isotope of a substance described
by paragraph (1) of this definition, or
some other substance related to a sub-
stance described by that paragraph,
which Is used In the study to assist In
characterizing the toxlcity, metabo-
lism, or other characteristics of a sub-
stance described by that paragraph.
Test system means any animal,
plant, microorganism, chemical or
physical matrix, including but not lim-
ited to soil or water, or subparts there-
of, to which the test, control, or refer-
ence substance is administered or
added for study. "Test system" also in-
cludes appropriate groups or compo-
nents of the system not treated with
the test, control, or reference sub-
stance.
Testing facility means a person who
actually conducts a study, I.e., actually
uses the test substance in a test
system. "Testing facility" encompasses
only those operational units that are
being or have been used to conduct
studies.
Vehicle means any agent which fa-
cilitates the mixture, dispersion, or so-
lubilization of a test substance with a
carrier.
§ 160.10 Applicability to studies performed
under grants and contracts.
When a sponsor or other person uti-
lizes the services of a consulting labo-
ratory, contractor, or grantee to per-
form all or a part of a study to which
this part applies, it shall notify the
consulting laboratory, contractor, or
grantee that the service is, or is part
of, a study that must be conducted in
compliance with the provisions of this
part.
§ 160.12 Statement of compliance or non-
compliance.
Any person who submits to EPA an
application for a research or market-
ing permit and who, In connection
with the application, submits data
from a study to which this part ap-
plies shall include in the application a
true and correct statement, signed by
the applicant, the sponsor, and the
144
-------
Environmental Protection Agency
§160.29
study director, of one of the following
types:
(a) A statement that the study was
conducted in accordance with this
part; or
(b) A statement describing in detail
all differences between the practices
used in the study and those required
by this part; or
(c) A statement that the person was
not a sponsor of the study, did not
conduct the study, and does not know
whether the study was conducted in
accordance with this part.
§ 160.15 Inspection of a testing facility.
(a) A testing facility shall permit an
authorized employee or duly designat-
ed representative of EPA or FDA, at
reasonable times and in a reasonable
manner, to inspect the facility and to
inspect (and in the case of records also
to copy) all records and specimens re-
quired to be maintained regarding
studies to which this part applies. The
records inspection and copying re-
quirements should not apply to qual-
ity assurance unit records of findings
and problems, or to actions recom-
mended and taken, except that EPA
may seek production of these records
in litigation or formal adjudicatory
hearings.
(b) EPA will not consider reliable for
purposes of supporting an application
for a research or marketing permit
any data developed by a testing facili-•
ty or sponsor that refuses to permit in-
spection in accordance with this part.
The determination that a study will
not be considered in support of an ap-
plication for a research or marketing
permit does not, however, relieve the
applicant for such a permit of any ob-
ligation under any applicable statute
or regulation to submit the results of
the study to EPA.
§ 160.17 Effects of non-compliance.
(a) EPA may refuse to consider reli-
able for purposes of supporting an ap-
plication for a research or marketing
permit any data from a study which
was not conducted in accordance with
this part.
(b) Submission of a statement re-
quired by S 160.12 which is false may
form the basis for cancellation, sus-
pension, or modification of the re-
search or marketing permit, or denial
or disapproval of an application for
such a permit, under FIFRA section 3,
5. 6. 18, or 24 or FFDCA section 406 or
409, or for criminal prosecution under
18 U.S.C. 2 or 1001 or FIFRA section
14, or for imposition of civil penalties
under FIFRA section 14.
Subpart B—Organization and
Personnel
§ 160.29 Personnel.
(a) Each individual engaged in the
conduct of or responsible for the su-
pervision of a study shall have educa-
tion, training, and experience, or com-
bination thereof, to enable that indi-
vidual to perform t\\e assigned func-
tions.
(b) Each testing facility shall main-
tain a current summary of training
and experience and job description for
each individual engaged in or supervis-
ing the conduct of a study.
(c) There shall be a sufficient
number of personnel for the timely
and proper conduct of the study ac-
cording to the protocol.
(d) Personnel shall take necessary
personal sanitation and health precau-
tions designed to avoid contamination
of test, control, and reference sub-
stances and test systems.
(e) Personnel engaged in a study
shall wear clothing appropriate for
the duties they perform. Such cloth-
ing shall be changed as often as neces-
sary to prevent microbiological, radio-
logical, or chemical contamination of
test systems and test, control, and ref-
erence substances.
(f) Any individual found at any time
to have an illness that may adversely
affect the quality and Integrity of the
study shall be excluded from direct
contact with test systems, and test,
control, and reference substances, and
any other operation or function that
may adversely affect the study until
the condition is corrected. All person-
nel shall be instructed to report to
their immediate supervisors any
health or medical conditions that may
reasonably be considered to have an
adverse effect on a study.
145
-------
§ 160.31
40 CFR Ch. I (7-1-90 Edition)
§ 160.31 Testing facility management
For each study, testing facility man-
agement shall:
(a) Designate a study director as de-
scribed in 1160.33 before the study is
Initiated.
(b) Replace the study director
promptly If it becomes necessary to do
so during the conduct of a study.
(c) Assure that there is a quality as-
surance unit as described in | 160.35.
(d) Assure that test, control, and ref-
erence substances or mixtures have
been appropriately tested for Identity,
strength, purity, stability, and uni-
formity, as applicable.
(e) Assure that personnel, resources,
facilities, equipment, materials and
methodologies are available as sched-
uled.
(f) Assure that personnel clearly un-
derstand the functions they are to per-
form.
(g) Assure that any deviations from
these regulations reported by the
quality assurance unit are communi-
cated to the study director and correc-
tive actions are taken and document-
ed.
§ 160.33 Study director.
For each study, a scientist or other
professional of appropriate education,
training, and experience, or combina-
tion thereof, shall be identified as the
study director. The study director has
overall responsibility for the technical
conduct of the study, as well as for the
interpretation, analysis, documenta-
tion, and reporting of results, and rep-
resents the singTe point of study con-
trol. The study director shall assure
that:
(a) The protocol, including any
change, Is approved as provided by
I 160.120 and Is followed.
(b) All experimental data, including
observations of unanticipated re-
sponses of the test system are accu-
rately recorded and verified.
(c) Unforseen circumstances that
may affect the quality and integrity of
the study are noted when they occur,
and corrective action is taken and doc-
umented.
(d) Test systems are as specified in
the protocol.
(e) All applicable good laboratory
practice regulations are followed.
(f) All raw data, documentation, pro-
tocols, specimens, and final reports are
transferred to the archives during or
at the close of the study.
§ 160.35 Quality assurance unit.
(a) A testing facility shall have a
quality assurance unit which shall be
responsible for monitoring each study
to assure management that the facili-
ties, equipment, personnel, methods,
practices, records, and controls are In
conformance with the regulations In
this part. For any given study, the
quality assurance unit shall be entire-
ly separate from and independent of
the personnel engaged In the direction
and conduct of that study. The quality
assurance unit shall conduct inspec-
tions and maintain records appropri-
ate to the study.
(b) The quality assurance unit shall:
(1) Maintain a copy of a master
schedule sheet of all studies conducted
at the testing facility indexed by test
substance, and containing the test
system, nature of study, date study
was initiated, current status of each
study, identity of the sponsor, and
name of the study director.
(2) Maintain copies of all protocols
pertaining to all studies for which the
unit is responsible.
(3) Inspect each study at intervals
adequate to ensure the integrity of the
study and maintain written and prop-
erly signed records of each periodic in-
spection showing the date of the in-
spection, the study inspected, the
phase or segment of the study inspect-
ed, the person performing the Inspec-
tion, findings and problems, action
recommended and taken to resolve ex-
isting problems, and any scheduled
date for reinspectlon. Any problems
which are likely to affect study integ-
rity found during the course of an in-
spection shall be brought to the atten-
tion of the study director and manage-
ment immediately.
(4) Periodically submit to manage-
ment and the study director written
status reports on each study, noting
any problems and the corrective ac-
tions taken.
(5) Determine that no deviations
from approved protocols or standard
operating procedures were made wlth-
146
-------
Environmental Protection Agency
§160.43
out proper authorization and docu-
mentation.
(6) Review the final study report to
assure that such report accurately de-
scribes the methods and standard op-
erating procedures, and that the re-
ported results accurately reflect the
raw data of the study.
(7) Prepare and sign a statement to
be included with the final study report
which shall specify the dates inspec-
tions were made and findings reported
to management and to the study direc-
tor.
(c) The responsibilities and proce-
dures applicable to the quality assur-
ance unit, the records maintained by
the quality assurance unit, and the
method of indexing such records shall
be in writing and shall be maintained.
These items including inspection
dates, the study inspected, the phase
or segment of the study inspected, and
the name of the individual performing
the inspection shall be made available
for inspection to authorized employees
or duly designated representatives of
EPA or FDA.
(d) An authorized employee or a
duly designated representative of EPA
or FDA shall have access to the writ-
ten procedures established for the in-
spection and may request testing facil-
ity management to certify that inspec-
tions are being implemented, per-
formed, documented, and followed up
in accordance with this paragraph.
Subport C—Facilities
§160.41 General.
Each testing facility shall be of suit-
able size and construction to facilitate
the proper conduct of studies. Testing
facilities which are not located within
an indoor controlled environment
shall be of suitable location to facili-
tate the proper conduct of studies.
Testing facilities shall be designed so
that there is a degree of separation
that will prevent any function or activ-
ity from having an adverse effect on
the study.
§ 160.43 Test system care facilities.
(a) A testing facility shall have a suf-
ficient number of animal rooms or
other test system areas, as needed, to
ensure: proper separation of species or
test systems, isolation of individual
projects, quarantine or isolation of
animals or other test systems, and rou-
tine or specialized housing of animals
or other test systems.
(1) In tests with plants or aquatic
animals, proper separation of species
can be accomplished within a room or
area by housing them separately in
different chambers or aquaria. Separa-
tion of species is unnecessary where
the protocol specifies the simultane-
ous exposure of two or more species in
the same chamber, aquarium, or hous-
ing unit.
(2) Aquatic toxiclty tests for individ-
ual projects shall be isolated to the
extent necessary to prevent cross-con-
tamination of different chemicals used
in different tests. «
(b) A testing facility shall have a
number of animal rooms or other test
system areas separate from those de-
scribed in paragraph (a) of this section
to ensure isolation of studies being
done with test systems or test, control,
and reference substances known to be
biohazardous, including volatile sub-
stances, aerosols, radioactive materi-
als, and infectious agents.
(c) Separate areas shall be provided.
as appropriate, for the diagnosis.
treatment, and control of laboratory
test system diseases. These areas shall
provide effective isolation for the
housing of test systems either known
or suspected of being diseased, or of
being carriers of disease, from other
test systems.
(d) Facilities shall have proper provi-
sions for collection and disposal of
contaminated water, soil, or other
spent materials. When animals are
housed, facilities shall exist for the
collection and disposal of all animal
waste and refuse or for safe sanitary
storage of waste before removal from
the testing facility. Disposal facilities
shall be so provided and operated as to
minimize vermin infestation, odors,
disease hazards, and environmental
contamination.
(e) Facilities shall have provisions to
regulate environmental conditions
(e.g., temperature, humidity, photo-
period) as specified in the protocol.
(f) For marine test organisms, an
adequate supply of clean sea water or
artificial sea water (prepared from
147
-------
§ 160.45
40 CFR Ch. I (7.1-90 Edition)
deionized or distilled water and sea
salt mixture) shall be available. The
ranges of composition shall be as spec
ified in the protocol.
(g) For freshwater organisms, an
adequate supply of clean water of the
appropriate hardness, pH, and temper-
ature, and which is free of contami-
nants capable of interfering with the
study, shall be available as specified in
the protocol.
(h) For plants, an adequate supply
of soil of the appropriate composition.
as specified in the protocol, shall be
available as needed.
§ 160.45 Test system supply facilities.
(a) There shall be storage areas, as
needed, for feed, nutrients, soils, bed-
ding, supplies, and equipment. Storage
areas for feed nutrients, soils, and bed-
ding shall be separated from areas
where the test systems are located and
shall be protected against infestation
or contamination. Perishable supplies
shall be preserved by appropriate
means.
(b) When appropriate, plant supply
facilities shall be provided. As speci-
fied in the protocol, these include:
(1) Facilities for holding, culturing,
and maintaining algae and aquatic
plants.
(2) Facilities for plant growth, in-
cluding, but not limited to green-
houses, growth chambers, light banks,
and fields.
(c) When appropriate, facilities for
aquatic animal tests shall be provided.
These include, but are not limited to,
aquaria, holding tanks, ponds, and an-
cillary equipment, as specified in the
protocol.
§ 160.47 Facilities for handling test, con-
trol, and reference substances.
(a) As necessary to prevent contami-
nation or mlxups, there shall be sepa-
rate areas for:
(1) Receipt and storage of the test,
control, and reference substances.
(2) Mixing of the test, control, and
reference substances with a carrier,
e.g., feed.
(3) Storage of the test, control, and
reference substance mixtures.
(b) Storage areas for test, control,
and/or reference substance and for
test, control, and/or reference mix-
tures shall be separate from areas
housing the test systems and shall be
adequate to preserve the identity.
strength, purity, and stability of the
substances and mixtures.
§ 160.49 Laboratory operation areas.
Separate laboratory space and other
space shall be provided, as needed, for
the performance of the routine and
specialized procedures required by
studies.
§ 160.51 Specimen and data storage facili-
ties.
Space shall be provided for archives,
limited to access by authorized person-
nel only, for the storage and retrieval
of all raw data and specimens from
completed studies.
Subpart D—Equipment
§ 160.61 Equipment design.
Equipment used in the generation,
measurement, or assessment of data
and equipment used for facility envi-
ronmental control shall be of appro-
priate design and adequate capacity to
function according to the protocol and
shall be suitably located for operation,
inspecticn, cleaning, and maintenance.
§160.63 Maintenance and calibration of
equipment.
(a) Equipment shall be adequately
inspected, cleaned, and maintained.
Equipment used for the generation,
measurement, or assessment of data
shall be adequately tested, calibrated,
and/or standardized.
(b) The written standard operating
procedures required under
§ 160.81(b)(ll) shall set forth in suffi-
cient detail the methods, materials,
and schedules to be used in the rou-
tine inspection, cleaning, maintenance,
testing, calibration, and/ or standardi-
zation of equipment, and shall specify,
when appropriate, remedial action to
be taken in the event of failure or mal-
function of equipment. The written
standard operating procedures shall
designate the person responsible for
the performance of each operation.
(c) Written records shall be main-
tained of all inspection, maintenance,
testing, calibrating, and/or standardiz-
148
-------
Environmental Protection Agency
§160.90
ing operations. These records, contain-
ing the dates of the operations, shall
describe whether the maintenance op-
erations were routine and followed the
written standard operating proce-
dures. Written records shall be kept of
nonroutine repairs performed on
equipment as a result of failure and
malfunction. Such records shall docu-
ment the nature of the defect, how
and when the defect was discovered,
and any remedial action taken in re-
sponse to the defect.
Subpari E—Testing Facilities
Operation
§ 160.81 Standard operating procedures.
(a) A testing facility shall have
standard operating procedures in writ-
ing setting forth study methods that
management is satisfied are adequate
to insure the quality and integrity of
the data generated in the course of a
study. All deviations in a study from
standard operating procedures shall be
authorized by the study director and
shall be documented in the raw data.
Significant changes in established
standard operating procedures shall be
properly authorized in writing by
management.
(b) Standard operating procedures
shall be established for, but not limit-
ed to, the following:
(1) Test system area preparation.
(2) Test system care.
(3) Receipt, identification, storage,
handling, mixing, and method of sam-
pling of the test, control, and refer-
ence substances.
(4) Test system observations.
(5) Laboratory or other tests.
(6) Handling of test systems found
moribund or dead during study.
(7) Necropsy of test systems or post-
mortem examination of test systems.
(8) Collection and identification of
specimens.
(9) Histopathology.
(10) Data handling, storage and re-
trieval.
(11) Maintenance and calibration of
equipment.
(12) Transfer, proper placement, and
identification of test systems.
(c) Each laboratory or other study
area shall have immediately available
manuals and standard operating pro-
cedures relative to the laboratory or
field procedures being performed.
Published literature may be used as a
supplement to standard operating pro-
cedures.
(d) A historical file of standard oper-
ating procedures, and all revisions
thereof, including the dates of such re-
visions, shall be maintained.
§ 160.83 Reagents and solutions.
All reagents and solutions in the lab-
oratory areas shall be labeled to indi-
cate identity, liter or concentration,
storage requirements, and expiration
date. Deteriorated or outdated rea-
gents and solutions shall not be used.
§ 160.90 Animal and other test system
care. »
(a) There shall be standard operat-
ing procedures for the housing, feed-
ing, handling, and care of animals and
other test systems.
(b) All newly received test systems
from outside sources shall be isolated
and their health status or appropriate-
ness for the study shall be evaluated.
This evaluation shall be in accordance
with acceptable veterinary medical
practice or scientific methods.
(c) At the initiation of a study, test
systems shall be free of any disease or
condition that might interfere with
the purpose or conduct of th? study. If
during the course of the study, the
test systems contract such a disease or
condition, the diseased test systems
should be isolated, if necessary. These
test systems may be treated for disease
or signs of disease provided that such
treatment does not Interfere with the
study. The diagnosis, authorization of
treatment, description of treatment,
and each date of treatment shall be
documented and shall be retained.
(d) Warm-blooded animals, adult
reptiles, and adult terrestrial amphib-
ians used in laboratory procedures
that require manipulations and obser-
vations over an extended period of
time or in studies that require these
test systems to be removed from and
returned to their test system-housing
units for any reason (e.g.. cage clean-
ing, treatment, etc.), shall receive ap-
propriate identification (e.g., tattoo,
color code, ear tag, ear punch, etc.).
149
-------
§ 160.105
40 CFR Ch. I (7-1-90 Edition)
All Information needed to specifically
identify each test system within the
test system-housing unit shall appear
on the outside of that unit. Suckling
mammals and Juvenile birds are ex-
cluded from the requirement of indi-
vidual identification unless otherwise
specified in the protocol.
(e) Except as specified In paragraph
(e)(l) of this section, test systems of
different species shall be housed In
separate rooms when necessary. Test
systems of the same species, but used
in different studies, should not ordi-
narily be housed in the same room
when Inadvertent exposure to test,
control, or reference substances or test
system mlxup could affect the out-
come of either study. If such mixed
housing Is necessary, adequate differ-
entiation by space and Identification
shall be made.
(1) Plants, invertebrate animals,
aquatic vertebrate animals, and orga-
nisms that may be used In multlspe-
cies tests need not be housed in sepa-
rate rooms, provided that they are
adequately segregated to avoid mlxup
and cross contamination.
(2) [Reserved]
(f) Cages, racks, pens, enclosures.
aquaria, holding tanks, ponds, growth
chambers, and other holding, rearing
and breeding areas, and accessory
equipment, shall be cleaned and sani-
tized at appropriate intervals.
(g) Feed, soil, and water used for the
test systems shall be analyzed periodi-
cally to ensure tfcat contaminants
known to be capable of interfering
with the study and reasonably expect-
ed to be present in such feed, soil, or
water are not present at levels above
those specified in the protocol. Docu-
mentation of such analyses shall be
maintained as raw data.
(h) Bedding used in animal cages or
pens shall not interfere with the pur-
pose or conduct of the study and shall
be changed as often as necessary to
keep the animals dry and clean.
(I) If any pest control materials are
used, the use shall be documented.
Cleaning and pest control materials
that interfere with the study shall not
be used.
(j) All plant and animal test systems
shall be acclimatized to the environ-
mental conditions of the test, prior to
their use in a study.
Subpert F—Test, Control, and
Reference Substance*
§ 160.105 Test, control, and reference sub-
stance characterization.
(a) The Identity, strength, purity,
and composition, or other characteris-
tics which will appropriately define
the test, control, or reference sub-
stance shall be determined for each
batch and shall be documented before
its use in a study. Methods of synthe-
sis, fabrication, or derivation of the
test, control, or reference substance
shall be documented by the sponsor or
the testing facility, and the location of
such documentation shall be specified.
(b) When relevant to the conduct of
the study the solubility of each test,
control, or reference substance shall
be determined by the testing facility
or the sponsor before the experimen-
tal start date. The stability of the test,
control, or reference substance shall
be determined before the experimen-
tal start date or concomltantly accord-
ing to written standard operating pro-
cedures, which provide for periodic
analysis of each batch.
(c) Each storage container for a test,
control, or reference substance shall
be labeled by name, chemical abstracts
service number (CAS) or code number,
batch number, expiration date, if any,
and, where appropriate, storage condi-
tions necessary to maintain the Identi-
ty, strength, purity, and composition
of the test, control, or reference sub-
stance. Storage containers shall be as-
signed to a particular test substance
for the duration of the study.
(d) For studies of more than 4 weeks
experimental duration, reserve sam-
ples from each batch of test, control,
and reference substances shall be re-
tained for the period of time provided
by 1160.195.
(e) The stability of test, control, and
reference substances under storage
conditions at the test site shall be
known for all studies.
150
<7 :
-------
Environmental Protection Agency
§ 160.120
§ 160.107 Test, control, and reference sub-
stance handling.
Procedures shall be established for a
system for the handling of the test,
control, and reference substances to
ensure that:
(a) There is proper storage.
(b) Distribution is made in a manner
designed to preclude the possibility of
contamination, deterioration. or
damage.
(c) Proper identification is main-
tained throughout the distribution
process.
(d) The receipt and distribution of
each batch is documented. Such docu-
mentation shall include the date and
quantity of each batch distributed or
returned.
§ 160.113 Mixtures of substances with car-
riers.
(a) For each test, control, or refer-
ence substance that is mixed with a
carrier, tests by appropriate analytical
methods shall be conducted:
(1) To determine the uniformity of
the mixture and to determine, periodi-
cally, the concentration of the test.
control, or reference substance in the
mixture.
(2) When relevant to the conduct of
the study, to determine the solubility
of each test, control, or reference sub-
stance in the mixture by the testing
facility or the sponsor before the ex-
perimental start date.
(3) To determine the stability of the
test, control, or reference substance in
the mixture before the experimental
start date or concomitantly according
to written standard operating proce-
dures, which provide for periodic anal-
ysis of each batch.
(b) Where any of the components of
the test, control, or reference sub-
stance carrier mixture has an expira-
tion date, that date shall be clearly
shown on the container. If more than
one component has an expiration date,
the earliest date shall be shown.
(c) If a vehicle is used to facilitate
the mixing of a test substance with a
carrier, assurance shall be provided
that the vehicle does not interfere
with the integrity of the test.
Subport G—Protocol for and Conduct
of a Study
6 160.120 Protocol.
(a) Each study shall have an ap-
proved written protocol that clearly
indicates the objectives and all meth-
ods for the conduct of the study. The
protocol shall contain but shall not
necessarily be limited to the following
information:
(DA descriptive title and statement
of the purpose of the study.
(2) Identification of the test, control,
and reference substance by name,
chemical abstracts service (CAS)
number or code number.
(3) The name and address of the
sponsor and the name-end address of
the testing facility at which the study
is being conducted.
(4) The proposed experimental start
and termination dates.
(5) Justification for selection of the
test system.
(6) Where applicable, the number,
body weight range, sex, source of
supply, species, strain, substrain, and
age of the test system.
(7) The procedure for identification
of the test system.
(8) A description of the experimen-
tal design, including methods for the
control of bias.
(9) Where applicable, a description
and/or identification of the diet used
in the study as well as solvents, emul-
sifiers and/or other materials used to
solubilize or suspend the test, control,
or reference substances before mixing
with the carrier. The description shall
include specifications for acceptable
levels of contaminants that are reason-
ably expected to be present in the die-
tary materials and are known to be ca-
pable of interfering with the purpose
or conduct of the study if present at
levels greater than established by the
specifications.
(10) The route of administration and
the reason for its choice.
(11) Each dosage level, expressed in
milligrams per kilogram of body or
test system weight or other appropri-
ate units, of the test, control, or refer-
ence substance to be administered and
the method and frequency of adminis-
tration.
151
4U-145 O-90 6
-------
§ 160.130
40 CFR Ch. I (7.1-90 Edition)
(12) The type and frequency of tests.
analyses, and measurements to be
made.
(13) The records to be maintained.
(14) The date of approval of the pro-
tocol by the sponsor and the dated sig-
nature of the study director.
(15) A statement of the proposed
statistical method to be used.
(b) All changes in or revisions of an
approved protocol and the reasons
therefore shall be documented, signed
by the study director, dated, and main-
tained with the protocol.
§ 160.130 Conduct of a study.
(a) The study shall be conducted In
accordance with the protocol.
(b) The test systems shall be moni-
tored in conformity with the protocol.
(c) Specimens shall be identified by
test system, study, nature, and date of
collection. This Information shall be
located on the specimen container or
shall accompany the specimen In a
manner that precludes error in the re-
cording and storage of data.
(d) In animal studies where histo-
pathology is required, records of gross
findings for a specimen from postmor-
tem observations shall be available to
a pathologist when examining that
specimen histopathologically.
(e) All data generated during the
conduct of a study, except those that
are generated by automated data col-
lection systems, shall be recorded di-
rectly, promptlytand legibly in ink. All
data entries shall be dated on the day
of entry and signed or initialed by the
person entering the data. Any change
in entries shall be made so as not to
obscure the original entry, shall Indi-
cate the reason for such change, and
shall be dated and signed or identified
at the time of the change. In automat-
ed data collection systems, the Individ-
ual responsible for direct data input
shall be identified at the time of data
Input. Any change in automated data
entries shall be made so as not to ob-
scure the original entry, shall indicate
the reason for change, shall be dated.
and the responsible individual shall be
Identified.
6160.135 Physical and chemical charac-
terization studies.
(a) All provisions of the GLP stand-
ards shall apply to physical and chem-
ical characterization studies designed
to determine stability, solubility, octa-
nol water partition coefficient, volatili-
ty, and persistence (such as biodegra-
dation, photodegradation, and chemi-
cal degradation studies) of test, con-
trol, or reference substances.
(b) The following GLP standards
shall not apply to studies, other than
those designated in paragraph (a) of
this section, designed to determine
physical and chemical characteristics
of a test, control, or reference sub-
stance:
I 160.31 (c). (d). and (g)
I 160.35 (b) and (c)
{ 160.43
i 160.45
I 160.47
{ 160.49
5 160.8Kb) (1). (2). (6) through (9). and (12)
{ 160.90
{ 160.105 (a) through (d)
§ 160.113
{ 160.120O) (5) through (12). and (15)
I 160.185(a) (5) through (8). (10), (12), and
(14)
{ 160.195 (c) and (d)
Subparts H—I [Reserved]
Subpart J—Records and Reports
§ 160.185 Reporting of study results.
(a) A final report shall be prepared
for each study and shall include, but
not necessarily be limited to, the fol-
lowing:
(1) Name and address of the facility
performing the study and the dates on
which the study was initiated and was
completed, terminated, or discontin-
ued.
(2) Objectives and procedures stated
in the approved protocol, including
any changes in the original protocol.
(3) Statistical methods employed for
analyzing the data.
(4) The test, control, and reference
substances identified by name, chemi-
cal abstracts service (CAS) number or
code number, strength, purity, and
composition, or other appropriate
characteristics.
152
-------
Environmental Protection Agency
§ 160.195
(5) Stability and. when relevant to
the conduct of the study the solubility
of the test, control, and reference sub-
stances under the conditions of admin-
istration.
(6) A description of the methods
used.
(7) A description of the test system
used. Where applicable, the final
report shall include the number of
animals used, sex, body weight range,
source of supply, species, strain and
substrain, age, and procedure used for
identification.
(8) A description of the dosage,
dosage regimen, route of administra-
tion, and duration.
(9) A description of all circumstances
that may have affected the quality or
integrity of the data.
(10) The name of the study director,
the names of other scientists or pro-
fessionals and the names of all super-
visory personnel, involved in the
study.
(IDA description of the transforma-
tions, calculations, or operations per-
formed on the data, a summary and
analysis of the data, and a statement
of the conclusions drawn from the
analysis.
(12) The signed and dated reports of
each of the individual scientists or
other professionals involved in the
study, including each person who. at
the request or direction of the testing
facility or sponsor, conducted an anal-
ysis or evaluation of data or specimens
from the study after data generation
was completed.
(13) The locations where all speci-
mens, rav. data, and the final report
are to be s: ured.
(14) The statement prepared and
signed by the quality assurance unit as
described in § 160.35(b)(7).
(b) The final report shall be signed
and dated by the study director.
(c) Corrections or additions to a final
report shall be in the form of an
amendment by the study director. The
amendment shall clearly identify that
part of the final report that IF being
added to or corrected and the reasons
for the correction or addition, and
shall be signed and dated by the
person responsible. Modification of a
final report to comply with the sub-
mission requirements of EPA does not
constitute a correction, addition, or
amendment to a final report.
(d) A copy of the final report and of
any amendment to it shall be main-
tained by the sponsor and the test fa-
cility.
§ 160.190 Storage and retrieval of records
and data.
(a) All raw data, documentation,
records, protocols, specimens, and
final reports generated as a result of a
study shall be retained. Specimens ob-
tained from mutagenicity tests, speci-
mens of soil, water, and plants, and
wet specimens of blood, urine, feces,
and biological fluids, do not need to be
retained after quality assurance verifi-
cation. Correspondence and other doc-
uments relating te interpretation and
evaluation of data, other than those
documents contained in the final
report, also shall be retained.
(b) There shall be archives for order-
ly storage and expedient retrieval of
all raw data, documentation, protocols,
specimens, and interim and final re-
ports. Conditions of storage shall mini-
mize deterioration of the documents
or specimens in accordance with the
requirements for the time period of
their retention and the nature of the
documents of specimens. A testing fa-
cility may contract with commercial
archives to provide a repository for all
material to be retained. Raw data and
specimens may be retained elsewhere
provided that the archives have specif-
ic reference to those other locations.
(c) An individual shall be identified
as responsible for the archives.
(d) Only authorized personnel shall
enter the archives.
(e) Material retained or referred to
in the archives shall be indexed to
permit expedient retrieval.
0 160.195 Keltntion of record*.
(a) Record retention requirements
set forth in this section do not super-
sede the record retention require-
ments of any other regulations in this
subchapter.
(b) Except as provided in paragraph
(c) of this section, documentation
records, raw data, and specimens per-
taining to a study and required to be
retained by this part shall be retained
153
-------
Part 162
40 CFR Ch. I (7.1-90 Edition)
in the archive(s) for whichever of the
following periods is longest:
(1) In the case of any study used to
support an application for a research
or marketing permit approved by EPA,
the period during which the sponsor
holds any research or marketing
permit to which the study Is pertinent.
(2) A period of at least 5 years fol-
lowing the date on which the results
of the study are submitted to the EPA
in support of an application for a re-
search or marketing permit.
(3) In other situations (e.g.. where
the study does not result in the sub-
mission of the study in support of an
application for a research or market-
ing permit), a period of at least 2 years
following the date on which the study
is completed, terminated, or discontin-
ued.
(c) Wet specimens, samples of test,
control, or reference substances, and
specially prepared material which are
relatively fragile and differ markedly
in stability and quality during storage,
shall be retained only as long as the
quality of the preparation affords
evaluation. Specimens obtained from
mutagenicity tests, specimens of soil,
water, and plants, and wet specimens
of blood, urine, feces, and biological
fluids, do not need to be retained after
quality assurance verification. In no
case shall retention be required for
longer periods than those set forth in
paragraph (b) of this section.
(d) The master schedule sheet,
copies of protocols, and records of
quality assurance inspections, as re-
quired by § 160.35(c) shall be main-
tained by the quality assurance unit as
an easily accessible system of records
for the period of time specified in
paragraph (b) of this section.
(e) Summaries of training and expe-
rience and job descriptions required to
be maintained by § 160.29(b) may be
retained along with all other testing
facility employment records for the
length of time specified In paragraph
(b) of this section.
(f) Records and reports of the main-
tenance and calibration and inspection
of equipment, as required by 5 160.63
(b) and (c), shall be retained for the
length of time specified in paragraph
(b) of this section.
(g) If a facility conducting testing or
an archive contracting facility goes
out of business, all raw data, documen-
tation, and other material specified in
this section shall be transferred to the
archives of the sponsor of the study.
The EPA shall be notified in writing
of such a transfer.
(h) Specimens, samples, or other
non-documentary materials need not
be retained after EPA has notified in
writing the sponsor or testing facility
holding the materials that retention is
no longer required by EPA. Such noti-
fication normally will be furnished
upon request after EPA or FDA has
completed an audit of the particular
study to which the materials relate
and EPA has concluded that the study
was conducted in accordance with this
part.
(i) Records required by this part
may be retained either as original
records or as true copies such as pho-
tocopies, microfilm, microfiche, or
other accurate reproductions of the
original records.
154
-------
APPENDIX D
LIST OF PARTICIPANTS
Dr. Karl H. Arne
Pesticide Specialist
U.S. E.P.A.
Region 10
1200 Sixth Avenue
Seattle, WA 98101
Dr. Jerry Barker
Range Ecologist
ManTech Environmental Tech., Inc.
EPA, ERL-C
200 S.W. 35th Street
Corvallis, OR 97333
Dr. Frank Benenati
Test Rules Development
US-EPA
401 M St. S.W.
Washington, DC 20460
Dr. Richard A. Brown
Ecology and Soil Science Sec.
ICI Agrochemicals
Jealott's Hill Research Station
Bracknell, Berkshire, RG 12 6EY, UK
Dr. Robert H. Callihan
Extension Weed Specialist
Department of Plant, Soil
and Entomological Sciences
University of Idaho
Moscow, Idaho 83843
Dr. Victor Canez
PAN-AG Laboratories
32380 Avenue 10
Madera, CA 93638
Dr. Joseph J. Dulka
E.I. du Pont de Nemours & Co.
Agricultural Products Dept.
Wilmington, DE 19880-0402
Dr. Charlotte Eberlein
R & E Center
P.O. Box AA
University of Idaho
Aberdeen, Idaho 83210
Dr. Frank A. Einhellig
Chair, Department of Biology
University of South Dakota
Vermillion, SD 57069
Dr. Eric Feutz
ABC Laboratories
P.O. Box 1097
Columbia, MO 65203
Dr. John Fletcher
Department of Botany
and Microbiology
770 Van Vleet Oval, Room 136
University of Oklahoma
Norman, OK 73019-0245
Dr. K.E. Freemark
Canadian Wildlife Service
Environment Canada
Ottawa, Canada, ON K1A OH3
Mr. Joseph W. Gorsuch
Health and Environment Laboratories
Eastman Kodak Company
Building 306, UP
Rochester, New York 14652-3617
Joseph C. Greene
Western Region Hazardous
Substance Research Center
Department of Civil Engineering
Oregon State University
Corvallis, OR 97331
222
-------
Dr. Hans Harms
Instut Fu Pflanzenernahrung und
Bodenkunde
Bundesforschungsanstalt fur
Landwirtschaft
Bundesallee 50, D-3300
Braunschweig-Volenrode (FAL)
Federal Republic of Germany
Dr. James Hoberg
Springborne Laboratories
790 Main St.
Wareham, MA 02571
Dr. Bill Hogsett
Plant Physiologist
EPA, ERL-C
Corvallis, OR 97333
Dr. Robert Hoist
Office of Pesticide Programs
EACH 7507C RM 700
US-EPA
Washington, DC 20460
Dr. Alan Hosmer
Ciba-Geigy
Agricultural Division
P.O. Box 18300
Greensboro, NC 27419
Dr. Charles Lewis
Office of Pesticide Programs
H7507C CM-2
US-EPA
Washington, DC 20460
Dr.Greg Linder
ERL-C/NSI
200 SW 35th St.
Con/all is, OR 97333
Dr. Michael Marsh
Pesticides Branch, Region 10
Environmental Protection Agency
1200 6th Ave.
Seattle, WA 98101
Dr. James Nellessen
Department of Botany and
Microbiology
University of Oklahoma
770 Van Vleet Oval, Room 136
Norman, OK 73019
Dr. Robert Parker
Extension Weed Scientist
Irrigated Agricultural Research
and Extension Center
Box 30
Prosser, WA 99350-0030
Mr. Richard Petrie
Office of Pesticide Programs
H7507C RM 815J
US-EPA
Washington, DC
Thomas Pfleeger
Plant Ecologist
EPA, ERL-C
Corvallis, OR 97333
Hilman Ratsch
Plant Pathologist
EPA, ERL-C
Corvallis, OR 97333
Dr. George Taylor, Jr.
Biological Sci. Center
Desert Research Inst.
P.O. Box 60220
Reno, NV 89506-0220
Dr. Philip Westra
112 Weed Science Lab
Colorado State University
Ft. Collins, CO 80523
223
-------
APPENDIX E
AGENDA
FOR
TERRESTRIAL PLANT TESTING WORKSHOP
Thursday, Nov.
8:10 - 8:15
8:15 - 8:30
29
Session I.
8:30 - 9:00
9:00 - 9:30
9:30 - 10:00
10:00 - 10:15
Session II.
10:15 - 10:45
10:45- 11:15
11:15 -11:45
11:45- 12:30
12:30 - 1:30
Session III.
1:30 - 2:00
2:00 - 2:30
2:30 - 3:00
3:00 - 3:30
3:30 - 4:00
4:00 - 5:00
Welcome to Corvallis Lab (Hogsett, ERL-C)
Opening Comments (Fletcher, Univ. of Okla.)
• Purpose of Workshop
• Introduction of Participants
• Topics
• Organization
• Products
Regulatory Policy and Needs
Mission of OPP and the role played by plant testing (Hoist, EPA, Washington, DC)
Adequacy of plant test data (Lewis and/or Petrie, EPA, Washington, DC)
Discussion
Break
Ecological and Taxonomic Considerations
Natural biotic and abiotic factors which influence plant growth and changes in
biodiversity. (Taylor, Desert Res. Inst.)
Assessment of published literature concerning pesticide influence on nontarget
plants. (Fletcher, Univ. of Okla.)
Innovative ways to use computerized data to estimate chemical damage to nontarget
vegetation. (Hogsett, ERL-C)
Discussion
Lunch
Laboratory Tests
Difficulties in performing existing tier 1 and 2 tests in Subdivision J. Guidelines
(Gorsuch, Kodak)
Development of nontarget plant test methods (Brown, ICI, Great Britain)
Break
Tissue culture tests (Harms, FAL, Germany)
Life cycle testing (Ratsch, ERL-C)
Discussion
224
-------
Friday, Nov. 30
8:30 - 9:00
9:00 - 9:30
9:30 -10:00
10:00-10:30
10:30-11:00
11:00-12:00
12:00 -1:30
Session V.
1:30 - 3:00
Dose-response of five sensitive crops (peas, lentils, alfalfa, sugar beets, and
potatoes) to sulfometuron (Callihan, Univ. of Idaho)
Effects of Oust, Harmony, Extra and Assert on potatoes (Westra, Colorado State
Univ.)
Break
Nontarget plant testing at Washington State Expt. Station (Parker, WA State)
Natural population testing (Pfleeger, ERL-C)
Discussion
Lunch
Formulation of Recommendations
Two working groups will be formed. One (the ecology group) will discuss
protection of natural plant communities, and the other (the agricultural group) will
address the protection of nontarget agricultural crops. Both groups will focus
attention on: (1) revision of existing lab tests, (2) field testing (design and
implementation), and (3) research needs. Each group will compile summary
comments and recommendations.
Break
All participants will be reassembled into one body, and a spokesperson from each
group will present a summary of their group's concerns, comments and compiled
recommendations. An open discussion will be conducted to compare the views and
recommendations put forth by the ecology and agriculture groups.
Drafting of Formal Recommendations
Selected authors will use the recommendations and discussion notes generated as
a result of Session V to draft a list of "working recommendations" pertaining to:
1) revision of existing lab tests
2) field testing (design and implementation)
3) research needs
Saturday, Dec. 1
Session VII. Presentation, Discussion, and Revision of "Working Recommendations"
9:00 - 10:30 Authors of the "Working Recommendations" will present their draft to the entire
group for discussion and, if necessary, revision.
3:00 -3:15
3:15 - 5:00
Session VI.
7:00 - 8:00
225
-------
APPENDIX F
Summary of key Subdivision J issues and aspects of this workshop which address these
issues.
Paper Delivered
Issue
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
Correct test species?
Number of methods used
for each test?
Life-cycle testing?
Test chemical, its makeup?
Validity of estimated
environmental concentrations?
Endangered species?
Tier III testing?
Standardization of tests?
Statistical analyses?
Test measurements?
Environmental test conditions?
Soil type?
Geographic use of chemical?
Author
Fletcher
Brown
.
Ratsch
-
-
Fletcher
Hogsett
Callihan
Westra
Parker
Pfleeger
Gorsuch
Gorsuch
Gorsuch
Gorsuch
Gorsuch
Hogsett
Page
27
55
_
77
-
-
27
36
88
95
102
105
55
55
55
55
55
36
Recommendation
II-7
IV-6
_
IV-2
II-7
-
IV-7
III
IV-3
I
II-5
I-l-f
II-8
II-4
-
226
-------
APPENDIX G
The following discussion was transcribed from rather faulty tapes and edited
when necessary. Despite difficulty in doing this, we feel that the text
included below is an accurate synopsis of key thoughts, opinions, concerns,
and ideas put forth by participants during the discussion period.
FRIDAY AFTERNOON COMMENTS ON SUMMARY PRESENTATIONS
At the conclusion of the formal presentations, the participants were requested
to review the information which had been presented at the workshop and prepare
two lists of major points and issues. One list dealt with the effectiveness
of Subdivision J testing to protect natural plant communities, and the other
to protect nontarget agricultural crops. These lists were presented to the
members of the workshop by Frank Benenati and Frank Einhellig, and the
participants responded with the following questions and comments.
Callihan: "I didn't think that we had meant to expand the diversity of the
species list into endangered species, but just to expand it to
increase more diversity as you said."
Einhellig: "I thought I heard if we expand it into more diversity we might be
able to pick things that may better mimic endangered species, not
that we would test endangered species. Pick a genus or a family
that would better represent a possible target that might be an
endangered species."
Petri: "Are there any species that have been tested like lily or cacti or
something like that?"
Callihan: "That data might enable us to pick up something. But we should
not aim for endangered species specifically. If a herbicide is to
be used in an area like that, where there are endangered species,
then we need some special treatment for registration purposes."
Westra: "I have not done this kind of testing, but in our group there was
a desire on the part of some people to have more precise and
helpful, clear cut guidelines for current tiers 1 and 2 testing
activities."
Greene: "My position was that we don't have technical support documents on
these tests. We have not performed interlaboratory tests on
assays. Although tests have been performed in different places,
we don't know the variability between laboratories with same
tests. In some cases, we're jumping ahead of where we should be
with guidelines. Perhaps some are written loosely for a
particular reason, but we should have a basic set of tightly
structured guidelines with modifications around that. If you want
to go to different soils, you need a standard set of assays so
that you can compare sensitivity. Then based on chemical and
physical data available you can select modifications in the
environmental conditions for the assays."
227
-------
Fletcher:
Hoist:
Greene:
Fletcher:
Hoist:
Greene:
Gorsuch:
"Actually, when you recognize that tests have been used
extensively for the last 4 years or so, there is no test better
than the actual user. Persons in this room have actually been
engaged in conducting these tests (tier 1 and 2). Comments made
at this meeting are the things that need to be corrected. I don't
see a need for a complete overhaul or long or lengthy
deliberation. We need to address those things that we think are a
problem. There are working tests out there that are being used
and are effective in gaining information."
"We need to fix things that we can fix now, based on our
experience with these tests over the last 10 years. Most things
can be worked on rapidly; in a year or two; subjects like species,
soil, etc. can be worked out. Joe is saying do some round robin
tests to make a critical evaluation of some parts of the test
(those not well defined that we need to get a better handle on).
There are two parts of the question. If we wait for round robin
tests with everything, it might take 10 years before certain
decisions are made. We need to make certain decisions in the
agency and need to know what things can be corrected. Then make
those corrections and put them out for comment and allow for some
time and testing and serious evaluation. It may take 3-5 years
(as was the case with mesocosms) to get it through."
"I was thinking of it, also from the perspective
harmonize tests between OECD, ASTM, and OSI. At
have to go to the literature and examine chemical
tested (similar chemicals) to see if results are
and if not, then someone must make a decision on
best. I have seen some scientific decisions made
have a responsibility to do some testing or find
been done."
of attempts to
some point people
s that have been
representative,
whose approach is
at the desk, we
if testing has
"In some cases there is a need for round-robin tests. In some
cases there is no excuse for inconsistencies that exist, and there
is no need for additional testing. The government agencies are
dragging their feet. I don't see why certain differences between
FDA and EPA are in existence. There is no need for round robin
testing, but need for action to clarify some of these things."
"I think we have enough experience to get to some of the points.
Some parts we don't have enough experience; we need to build up
experience in those areas."
"My point is that EPA shouldn't have the right to say , we are
right and FDA is wrong. We are not in a position to do that. We
need data to convince OECD or FDA that our test is better. The
data should be in the literature."
"It may not be in the literature, but submitted to an agency and
available from the agency, but not published."
228
-------
Greene: "Then the agency has the responsibility to put it in the
literature."
Hoist: "There are certain things that need to be put out, some things we
can't. It is difficult for a regulatory agency to release our
information, because industry needs to take credit for what they
worked on. Is it possible for chemical companies to come around
on some of their stuff? You do put out some information at
professional societies, meetings, and published papers."
Dulka: "It is a long torturous process."
Hoist: "But it's after the registration process."
Brown: "There has been talk about taking this information and finding a
forum for distribution of information, like ASTM?"
Gorsuch: "ASTM is a good forum; it may not have wide distribution, but is
peer reviewed and does get distributed."
Brown: "I would like to talk about this from the industrial point of
view. We think that somewhat of a hopeful position can be gained
perhaps with the use of mesocosms. With this we have an
opportunity to try to bring some sort of framework from both sides
of the debate so all sides will benefit. We have to move quite
fast to do that; the world will not wait for us to sort out what
it is we should measure."
Fletcher: "In my association with terrestrial microcosms, there is always
some question of how well microcosms reflect the field. Some of
you have experience, have you found them to be beneficial?"
Dulka: "What Richard [Brown] was alluding to was our experience with the
mesocosm process with avian and aquatic tests. Typically we're
doing it by starting out with tier testing, laboratory tests, life
cycle tests for early life stages, and acute and chronic tests.
We generate that information, then move to the field program. In
the aquatic area we start off with farm pond studies, then on to
better controlled and more highly defined mesocosm studies,
starting to look at community response. The farm pond situation
is kind of what we are talking about here. Going out with
monitoring correlates with talking about replicated fields and
looking at field edge. Finding mass replicated fields with the
same plant population is limited. These cost hundreds of
thousands of dollars. Then we move on to the aquatic mesocosm
studies running 2-3 million dollars.
Concerning aquatic mesocosm studies, looking at community level
effects, initially the protocols were designed as an integrated
approach looking at the highest trophic levels for particular
organisms in that system and extrapolated down to other organisms
in the system. The issue that keeps coming up is if you produce
an effect at lower trophic levels, what does that mean? No one
has a clear definition. It's impossible to generate dose response
229
-------
curves for different species in the system because the study was
not designed for it. A key element is what is the purpose of the
study and what is the endpoint you are trying to achieve in the
study. Once we define that, then design statistics for it. To
design an experiment as to what to address and try to do that in
such a short period of time is a major job. It is analogous to
what Richard said about putting up goalposts. We need to clearly
define what we are trying to accomplish and not take the approach
of playing on a field without any ends to the field or idea of the
game. We are now in the situation with aquatic mesocosms where we
have a lot of data and still not sure we are answering questions.
What are the key elements we need to address and what is the
proper study design to use?"
Fletcher: "Are you talking about the design of the field study or of
mesocosms?"
Dulka: "Maybe design of a field study, or intermittent tier study to
examine the key elements of a lower tier study. In aquatic
systems, we moved from a microcosm study to mesocosm, identify
same effects in the mesocosm study, at least help you determine if
a full blown mesocosm study is needed or to focus effort for key
things to look for when going to the field test."
Fletcher: "When you talk about microcosms and mesocosms in terrestrial
systems, you're talking about open top chambers or self-contained
chambers with light and humidity that are very expensive studies.
There is always a question of how well chambers duplicate the
environment. I think it is a mistake for the agency to use
expensive growth chambers. When you talk about microcosms and
mesocosms that's what is involved. I think a field study will get
the answers quicker than a long elaborate microcosm-mesocosm
research effort."
Brown: "We are not shackled to mesocosms, but what we want is agreement
as to what precisely we are trying to measure to protect the
environment. As far as crops are concerned, we have good ideas,
but for natural species, there are questions and problems to be
worked out."
Dulka: "Another analogy to draw is the pyrethroid study we're going
through now. We ran the studies and now want to reduce risk by
reducing the exposure; to do that we need to establish hazard from
the toxicity data to know what kind of buffer to put in place. To
do that a mesocosm study may not be the best approach to answer
that question. There are other issues of multi-species and stages
of growth and development that must coincide with when the product
is actually used. The timing part and the likelihood of when the
product moves off target has not been discussed. The endpoints
are not clearly defined as to what constitutes unreasonable risk.
Testing is not an easy task, when concerned about ecosystem
effects which laboratory studies do not necessarily address."
230
-------
Fletcher: "I'm not denying the fact that you make progress in small steps.
I don't know how many meetings have been held, but coming away
from this meeting there should be some small step of progress
toward tier 3 testing."
Hoist: "It helps to go through the thought process; how do you (industry)
go from screening tests, to laboratory tests, to small plot field
tests, to large OUT tests. What is the basis for progressing?
How do you go from step to step? When do you go to the field?"
Dulka: "For efficacy tests, find out what the spectrum of control is and
what the breaking point is for that species. In a row crop test,
it is more of a noxious plant weed control community test within a
row crop."
Hoist: "Turn that around and try to answer the mirror side of that, find
out the other side of the question."
Dulka: "To put it another way, we are testing for EC 80 or better and
what your testing for is EC 25 and less."
Benenati: "How much would it take to redesign what you are doing now to make
tests more sensitive."
Dulka: "This is when it gets into GLP (Good Laboratory Practices) issues.
Our groups intent and design is to bring about new products that
meet new market needs (economic and environmental). It takes
years and depends on the product and the market."
Hoist: "On the other side of the coin, when you do one set of testing,
you also do the other simultaneously. I'm talking about testing
in general for phytotoxicity. When you do efficacy testing you
are also doing nontarget area toxicity at the same time."
Brown: "Efficacy tests gather very large amounts of low quality data.
The whole thing is designed for enormous throughput. We make
decisions from data collected all over the world. Because of GLP
requirements associated with the collection of environmental data,
any attempts to combine this with efficacy studies would force us
to gather a small amount of high quality data, and we would not
get to the point of evaluating chemicals quickly as we do in our
current efficacy testing programs."
Dulka: "The concept statistically is that we're willing to accept a false
positive in screening tests. From a regulatory view, that would
be unacceptable. The possibility of getting a false negative is
not as likely."
Hoist: "I'm looking at the process with more focus being placed on going
from tier to tier."
Dulka: "If we know we are controlling giant foxtail, millet, bindweed,
and other species in wheat, and at some percentage application we
have efficacy. Is it necessary to go to the field to prove that?
231
-------
Those are ecologically important key species as far as insects and
birds are concerned. It is a question of placement; it goes back
to some of the other things which we discussed before about
application technology and concerns for environmental risk."
Hoist: "Sometimes that extra little bit of information such as controlled
weeds and degree of control goes a long way in helping us know the
diversity of species control or species effects with respect to
nontarget plants. How good the GLP is that goes a long with that
is maybe something that we have to do in-house and get the
information."
Dulka: "I guess we were submitting it [data], but whether or not we can
design a study to get that information from both the efficacy
standpoint and the lower effects is questionable. In a field
study using plots, typically an investigator will set up a number
of controls and treatments in a random design with 3-4 replicates.
There might be a great deal of variation in the stand in both the
weeds and the target crop. It is difficult in comparing the
controls, to determine what constitutes an effect. I'm not sure
in that senario whether it is possible to come up with EC 25 or EC
20."
Freemark: "Is there a greater chance of doing that using greenhouse data
rather than a screening test?"
Hoist: "Its the amount of information; more than just the 10 species."
Dulka: "There is data generated typically by industry to see if activity
is based upon similarities between genus and family and other weed
species. This information is not necessarily generated from
replicated tests, and may work one year but not work the following
year."
Eberlein: "I want to bring up a point from yesterday. It concerns the whole
question of drift. The root of the problem is the 60% application
rate, we shouldn't forget that. If we improve efforts in
application efficiency then we may not have to worry about
developing testing."
Dulka: "Just remember that the 60% application rate came from a forestry
study with release 30 to 50 feet above the canopy."
Hoist: "With herbicides versus insecticides, herbicides have the better
efficiency they use a bigger droplet."
Benenati: "As part of the industry study, are they going to be looking at
that and other parameters?
Hoist: "Yes."
Eberlein: "Is anyone in industry looking at new, novel application
techniques that will greatly improve efficiency?"
232
-------
Hoist: "Industry is coming to the agency on the 18th and 19th, and at
that time we will be looking at what they plan to do. I do not
know what they will do, we are not in the party in respect to the
information, presentations and recommendations."
Eberlein: "The tests we've been talking about require millions of dollars
being spent, maybe we would be better off with a whole new
application system."
Brown: "We certainly have the techniques (electric ion sprays). At the
moment, the market requires special formulations and special spray
rigs. We must also register new formulations."
Dulka: "You're talking about current technology and the cost of modifying
current pieces of equipment. The spray group task force is
attempting to address the question of reducing off target spray
drift. Information will be coupled with the FSCBG spray drift
model used by the Forest Service for the last 5-6 years. The
overall goal is to generate enough significant information on a
significant number of different types of equipment (air blast,
ground ,air, helicopter,etc.) and to use physical and chemical
properties of spray tank mix, coupled with wind tunnel tests, and
actual field experiments to predict drift potential."
Callihan: "What is the nature of this Spray Group Task Force?
Dulka: "It includes representatives from industry and academia
(International)."
Hoist: "It started as part of NACA, but is now strictly the companies
themselves."
Callihan: "Any interaction with Western Region Coordinating Committee of
Improvement of Spray Technology?"
Dulka: "We have interaction with many individuals, applicator's
association, investigators at Kansas State, Davis CA, New mexico,
etc."
Westra: "When you look at interactions in plant communities (neighborhood,
interspecific, intraspecific, and competitive effects), with
regard to herbicide stress effects, how important are herbicide
effects in relation to competition and neighborhood effect in a
mixed community system?
Callihan: "According to the talk by George Taylor, biotic effects far
outweighed herbicide stress on the community."
233
-------
SATURDAY MORNING COMMENTS ON TENTATIVE
RECOMMENDATIONS
Friday evening a list of "tentative recommendations" was compiled by four
participants; Frank Benenati, Frank Einhellig, Joe Gorsuch, and Hilman Ratsch.
On Saturday morning the tentative recommendations were presented to all
members of the workshop and the following questions, comments, and requested
changes were made.
1) Drop tier 1 seed germination tests except for those cases where there is
reason to believe germination is a more sensitive indicator of effects.
Freemark: "So what does that leave in tier one then?"
Einhellig: "You go ahead and draw conclusions a little further away (further
into the plant life cycle). That is where you are going to find
greater sensitivity, and where you'll find more things."
Hoberg: "We could probably drop the germination test altogether; we're
covering that with the emergence test."
Einhellig: "I think that there might be some cases where people might want to
do a germination test as a presentation if they really did find
something. You're right, you're covered. This basically says
drop it, but leaves some openings."
2) Develop criteria for acceptability of emergence and germination (if conducted)
response in a particular soil medium.
Gorsuch: "Develop criteria to establish when emergence is actually taking
place. Criteria for germination is that you have at least 5mm
growth, that is not always consistent with other agencies. There
was not a criteria established for emergence. They are asking for
a definition of emergence. When does it take place? Also, define
the length of the test. Whether 10, 14, or 21 days, establish a
length of time where either plant dies or recovers."
Fletcher: "You're talking about a cut-off time for emergence and a cut-off
time for seedling growth."
234
-------
Gorsuch: "Yes. The idea was, vegetative vigor would follow immediately
after and continue on much like the OECD test. You have emergence
data, first 7-10 days, then continue on."
3) Analytical determination of the chemical in use will only be required to
conclude a negative finding and terminate the test. (All current analytical
procedures will be required in tier 2.)
Freemark: "How does that work for vegetative vigor? What about foliar
applied products?"
Gorsuch: "No, currently for tier 1, you must analyze soil and/or the stock
solution. For tier 1, since it is basically a screening test, you
should be allowed to use nominal concentrations, prepared without
verification by analytical analysis."
Freemark: "Except when you have no effects."
Gorsuch: "EPA/TSCA guidelines require that if you don't see any effect at
one concentration, you do a chemical concentration determination."
Benenanti: "Why do tier 1 testing for range finding data?"
Dulka: "So what are we analyzing?"
Gorsuch: "The test solution. The FIFRA guidelines also require that you
test the soil when you get no effect."
Dulka: "We should be analyzing the plant parts. If you make the test
solution up, and it is used within a short period of time, the
chances of it degrading are small."
Gorsuch: "The interpretation of GLP is analyze the soil."
Dulka: "That doesn't make sense!"
Gorsuch: "I didn't write the guideline."
Dulka: "I don't think we should set this so rigidly. There are a number
of ways of demonstrating how the material was applied to
demonstrate that it met the requirements."
Canez: "In field studies, EPA has not required tank mix samples. You can
do mass balances to calculate how much you are getting on that
plant. If you calibration is set and your analytical balance is
calibrated, all your documentation is there, why is analysis of
the test solution required?"
Einhellig: "Rick Petri is not here to tell us."
Dulka:
"It is not in the guidelines."
235
-------
Gorsuch: "We were cited for noncompliance."
Fletcher: "Is there an objection to what is written?"
Dulka: "It is not clear what the consensus is of analytical
determinations."
Benenanti: "We were trying to reduce the burden of analytical requirements."
Brown: "We need to characterize the test solutions just to cover out
backsides."
Einhellig: "We need clarity on this issue. Somehow this must be rewritten."
Gorsuch: "Talk to Bob Hoist."
Dulka: "The concern is where the testing should take place."
Freemark: "What does this phrase 'in lieu of GLP analytical determination'
mean?"
Benenati: "Let me think about it for a moment."
4) Identify and characterize the nature of the soil required for testing procedures
(perhaps start with OECD guidelines).
Einhellig: "Whatever soil you run tests in, it is not well defined at least
by FIFRA guidelines. But it has to be free of other chemicals and
you should have a better characterization."
Dulka: "Is that to identify a standard soil or characterize the matrix
that you are using?"
Benenati: "We want to maximize the soil characteristics and minimize the
organic content."
Brown: "We've optimized our soil, but we use different compost for
different species."
Benenati: "Each species will have a characteristic soil."
Brown: "It's all defined. It would be nice to have standardized
characteristics."
Dulka: "What about pasteurization?"
Einhellig: "Our group didn't discuss that."
236
-------
Canez: "Non-pasteurized soil eliminates microbial decomposition. You
eliminate microbial interaction and any fungal pathogen
interaction, it's a worse case scenario; but it is standardized."
Einhellig: "We are not trying to rewrite the guidelines; we just want to draw
attention."
Dulka: "If you don't use pasteurization and you are using non-fungicide
treated seed, you may not see your seed come up for other reasons.
You may pick that up in your control, you may not."
Fletcher: "There is a 25% effect buffer."
Dulka: "It also depends on the season."
Benenati: "We'll put that down as an additional issue; pasteurization as a
consideration."
Fletcher: "The main task is to identify those areas that are hazy or fuzzy
and the aspect of soil is hazy/fuzzy. These considerations should
include pasteurization and organic content."
5) Provide better guidelines on experimental design and interpretation of
statistical analysis procedures on a species by species basis (add reference
to Appendix).
6) Evaluate and expand the current recommended list of test species with the
objective of enhancing the use of more diversity. The intent would not be to
require more species to be tested, but to include representative genera and
families that might be extrapolated to woody species and/or endangered
species, where appropriate.
Freemark: "How about formulated versus active only testing?"
Einhellig: "We didn't discuss this. But it should be something to consider."
Benenati :
Canez:
Dulka:
Brown:
Freemark:
"At this stage of product development, we usually don't have
studies."
"For reregistration, we do."
"For reregistration, using the generic formula is too costly to
use in the field."
"When we do these studies, it's far down the track. The things
have been in field studies for years. You just need to deliver
the technical material in a reasonable way."
"Formulation should be used.
worst case."
It's an environmental test of the
237
-------
Dulka: "Different formulations may be used in the final reports."
Freemark: "But, if you are going to use a surfactant, you should test with
the surfactant, if you are going to use some sort of sticker to
keep it on the leaves, you should use a sticker."
Benenati: "If you use both, you should test both."
Fletcher: "This is tier 1...Should we include the suggestion that a generic
formulation should be used in tier l...Do we need a hand ballot?"
Feutz: "You have to make the formulation the end product. You're trying
to evaluate the technical grade, but you are increasing the mode
of action when you use a surfactant or sticking agent."
Dulka: "The test formula may not be the final product. If you don't have
the final formulation worked out, you still have to apply the
product with some sort of surfactant or sticking agent to maximize
the test. This is typical. It may not be your final formulation
that goes out to sales."
Einhellig: "Does the group want a suggestion to identify the make-up of the
final chemical in Tier 1?"
Dulka: "It has to be as flexible as possible."
Fletcher: "The point needs to be clarified."
Freemark: "The question becomes whether you want to recommend a formulative
or generic end-use product testing be required or whether the
whole issue be considered."
Brown: "We've over examined the question. We have products that look
like black vinyl, that must be emulsified in some vehicle to be
applied. But that may not be the final formulation."
Einhellig: "So we are going to make the point that we look at it, but don't
over-regulate the situation."
Fletcher: "I move that we have a point 7 stating chemical testing procedures
should be reviewed."
Freemark: "I think it should be more specific."
Dulka: "Another option is in the standard for reregistration to specify
technical grade active ingredient (TGAI) or technical end-use
produce (TEP)."
Brown: "The effects between different species outweigh the effects of
different formulations."
238
-------
Einhellig: "Let's just make a motion for review since we cannot make the
determination here."
Freemark: "Why not? Testing of TEP should be considered."
Fletcher: "I personally think that review is all we can recommend. This
needs to be considered by experts."
Feutz: "For technical versus end-use, technical was chosen in SEP."
Brown: "What do you do with black waxy solids?"
Callihan: "You test them with what you are going to use them in."
Brown: "That could be your development formulation or a "Mickey Mouse"
formulation."
Freemark: "My concern is that the surfactants, spreaders and stickers are
not being included in the testing. Yet, in the end-use situation,
they are specifically there to enhance the target uptake.
Environmentally, this represents a worst case scenario when they
say it has 'no effect,' but may have an effect when these vehicles
are used."
Dulka: "We normally use a standard surfactant."
Fletcher: "OK, we can identify this item as a concern, but we cannot make
recommendations."
Einhellig: "Let's vote to include a new point (7) that the guidelines be
reviewed.
(MAJORITY VOTE YES.)
7) Range finding for tier two concentration determinations should be used to
determine if foliar or soil application results in more sensitive responses. The
most sensitive exposure route should be used in tier two tests. If equally
sensitive, tier two should use the most relevant route of exposure.
(VOTED TO BE REMOVED FROM THE RECOMMENDATIONS SINCE IT WAS
NOW A MOOT POINT.)
(8) Optimize test conditions in tier two: i.e., temperature, photoperiod, light,
humidity, CO2.
Canez: "Optimize or standardize? Optimize means using corn in the
midwest, standardize means using growth chambers."
Benenati: "There you might want to add supplemental lighting."
239
-------
Fletcher: "The term optimize does not belong. If you optimize growth, you
will elevate C02."
Benenati: "What we are talking about is standardize."
Dulka: "In a greenhouse scenario, how close are you going to get to
standardized conditions?"
Brown: "Last summer, we had a very warm summer."
Dulka: "In the real world, plants grow under all kinds of conditions."
Benenati: "When we added item (2), the idea was acceptance criteria that the
plant had to grow a certain amount. It didn't really matter what
conditions the plant grew in, more that it met certain criteria
towards performance criteria. Our direction was to try to get
away from criteria that were analytically based, and get
performance based criteria."
Gorsuch: "All guidelines have some criteria, OECD..."
Fletcher: "This is a blanket document...We need to give rough ballpark
figures."
Callihan: "Why are we telling them how to grow plants?"
Fletcher: "Because we are telling them how to do the tests? I feel we need
some statement providing upfront guidance, setting minimum
standards."
Hoberg: "2000 footcandles intensity were recommended by FDA, but we were
allowed 2000 footcandles quality."
Eihellig: "How about 'Guidelines for minimum test conditions.'"
Fletcher: "Some freedom for individual species should be considered."
Einhellig: "They need to be acceptable guidelines so that people can produce
credible work."
Hoberg: "Guidance in environmental conditions and appropriate species
involved."
Dulka: "Purpose of the test is to test phytotoxicity. Rigid conditions
will not work."
Fletcher: "Minimum conditions are what need to be suggested."
Dulka: "You need credibility of tests and good results so you optimize
conditions."
Fletcher: Physical minimum conditions."
240
-------
Dulka: "We need wordsmithing so we don't get burned and you still have
confidence."
Pfleeger: "How about using a standard plant like wheat...and conditions are
optimized to grow wheat to certain standards?"
Fletcher: "I would like to know what the agency wanted."
Westra: "Controlling light intensity is the most important factor.
Suggest minimal lighting levels. Give minimum requirements."
Freemark: "Review test conditions."
Einhellig: "The point is to provide minimum guidelines."
Benenati: "I was looking at Rainbow trout data and all were using 'standard
conditions,' but the weights varied. The object is to demonstrate
optimum growth, and to demonstrate that your lab is properly
nurturing these organisms, that they are viable, and there is
enough difference between your controls and your test organisms to
show good response."
Arne: "Guidelines are not law. If you go away from guidelines and still
get the right information, that's OK."
Benenati: "You have to meet criteria."
Freemark: "It should be reviewed."
Dulka: "You should only have to define the conditions of the test that
you need."
Freemark: "The difficulty is if you document what you did and then they
don't accept the data."
Dulka: "Not for FIFRA."
Gorsuch: "GLP standards don't specify."
Einhellig: "Why don't we vote to (a) wipe it out, (b) give some kind of
statement to provide minimum guidelines, (c) refer it for review."
(REFER FOR REVIEW WAS SELECTED.)
II. Harmonize differences in test procedures between different regulatory
authorities or governing bodies (OECD, EEC, FIFRA, TSCA, FDA, CERCLA)
and work toward harmony with inter- national communities testing
requirements. Because of these inconsistencies, testing costs for laboratories
maintaining two or more programs are increased.
241
-------
1) Establish what the inconsistencies are between agencies, e.g.,
a) EC-25 for effect under FIFRA, compared to 1% tolerance for
FDA.
b) Number of species required, number of plants per test, and
number of replicates per test.
c) Nutrient addition (FDA) compared to no nutrient addition required
by FIFA.
d) Photoperiod requirements under FDA, but not specified in some.
e) Watering regiments that should be optimized.
f) Endpoints that are required: FIFRA does not require shoot
heights, root length, and shoot and root weights, whereas FDA
does.
2)
...ETC.
Call for joint efforts to arrive at a consensus on the testing procedures.
Einhellig: "I'm not sure who this is addressed to, but this is a working
group to recommend. Perhaps this goes to the head of the working
group. I'm not sure who is the recipient of that call."
Fletcher: "My understanding is that EPA in Washington, DC is concerned about
these differences. "
Einhellig: "So we should give examples of these differences?"
Freemark: "Different agencies have different mandates. For example, the
difference between TSCA and FIFRA in recommendations for using
test substrates."
Fletcher: "We are not trying to clarify these at the meeting. We are just
trying to point out that there are differences and that an effort
should be made to harmonize these."
Freemark: "Don't forget Canada when you harmonize."
Eihellig: "Let us move on to point four and come back to point three."
(ALL AGREE)
IV. Research is needed to improve the efficiency and in some cases, the validity
of testing protocols. Special case needs include: (in no priorital order)
1) Establish the feasibility of using tissue culture methods as options for
tier 1 and 2 work. Tier 1 might include multiple exposure
concentrations, but testing would be comparable to range finding in tier
2 without GLP/analytical determinations. A special focus should be to
242
-------
use tissue culture as a surrogate for tests on non-target woody species
of concern.
2) Develop efficient life cycle bioassays, both for representative dicot and
monocot species. This should include methods of chemical
applications in these bioassays.
3) Research the possibility and procedures for using mesocosms and field
studies to evaluate chemical effects on plant communities.
a) Agroecosystem models versus natural community models.
b) What parameters should be indicators of effects?
c) When should such studies be required?
d) Evaluate the feasibility of using soil core and terrestrial
microcosm chambers and other "off the shelf" technologies.
4) Study the possibilities of using remote sensing procedures and other
technologies to monitor chemical effects in field tests, including
possible effects on nontarget plants and plant communities.
Eberlein: "This could be a useful screen for use before you go on to
mesocosm testing."
Fletcher: "This seems like a good technology to use for monitoring."
Brown: "How do you define a problem?"
Eberlein: "The computer can identify differences between water stress or
imidazole. For EUPs, you could fly over before and after to check
for evidence of drift."
Freemark: "For point (3), we want to use these technologies to develop
better exposure assessments, spacial characteristics, modeling,
and hazard and risk assessment."
Dulka: "Add 'understanding' to (3a)."
Brown: "I want a description of what is the biological significant
effect."
Einhellig: "(b) does that."
Dulka: "Define what constitutes an adverse ecological effect."
Benenati: "You don't have to go back to basics. You have to define how you
go from a small scale system to a large scale system, that is
extrapolation. We are going to build upon 'off-the-shelf
technology. We need to go on from here."
Brown: "What is the answer then?"
243
-------
Benenati: "When we start showing effects on mineral cycling or the carbon
cycle or sulfate. Those are all logically important parameters
and now you're going to take these results in the limited sense
that you're doing a small scale mesocosm or microcosm work and
you're going to say, 'OK, let's go into a natural community and
try to limit that."1
Brown: "What concerns me is when you go into a field study, what are the
first things you want to do? You have to say, 'What is it we are
trying to measure?' We want to identify the variability of the
species. We have to determine what kind of effect and what
percentage of effect we are looking for."
Einhellig: "Could we solve this by saying what parameters should be
indicators of effects and descriptors of what unacceptable effects
will be?"
Dulka: "We are faced with all these multimillion dollar studies and we
are unable to determine what is the endpoint of the study, what
constitutes an effect. Carbon cycling and nutrient cycling are
not going to give you indications on the community."
Benenati: "We have to pick whether it's cycling of a nutrient or competition
or what ever, and then when you go out in the field, focus your
field studies on that aspect."
Dulka: "You have to identify the hazard. What is the endpoint of the
study? What are you looking for?"
Benenati: "You pick the parameters to study in relation to habitat and most
sensible variables."
Freemark: "You're talking about design of the studies. Much of the basic
ecological work has not yet been done."
Fletcher: "You have to describe the dynamics of the system first, measure
variance from that normal situation, and then someone decides what
is the acceptable amount of variation. Research is required."
Dulka: "You have to have endpoints to know where your research is
supposed to go."
Fletcher: "You have to have research to define endpoints."
Freemark: "What kind of research? Mesocosm?"
Fletcher: "The kind we are discussing."
Freemark: It's a shame to bury this in mesocosms. This same need is in
field tests, too."
244
-------
Brown:
"It is the number one research need."
Einhellig: "All this discussion leads to the fact that we need research in
this area."
Freemark: "Let's not leave this buried in mesocosms."
Fletcher: "Why not make it 'Procedures for using mesocosm studies and field
studies?'"
(ALL AGREE)
5) Research and a database on culture techniques are needed for plant
species identified for tier 3 evaluation. These include forest and
understory species and wetland species.
Fletcher: "We don't know if we have adequate data to identify the
differences between sensitivities of key families."
Einhellig: "Let's add that. Anything else?"
Unknown: "I want to add under Number 1: non-target organisms and
'endangered species.'"
Eihnellig: "Any objections to upping the priority of tissue cultures?"
(NONE.)
6) There needs to be research done to examine extrapolation of
toxicological data from inter- and intra-surrugotts.
Einhellig: "Let's move on to part III."
III. Design and implementation of field experiments should be clarified.
A. Current tier 3 requirements appear to be more efficiently accomplished if
divided into two phases, or considered as two separate tiers.
1. Develop protocols or consensus methodology for small plot tests with
regard to critical species, soil type, and other field variables.
(NO ONE HAS A PROBLEM WITH THIS)
Brown: "Our concern is that we want to know what we have to look for
before we get started. If we have some endpoints formulated...it
seems to be a case by case basis."
Pfleeger: "Everyone wants the definition of endpoint specified. Different
offices have different definitions. Superfund definition may
differ from your definition."
245
-------
Benenati: "That is part of the harmonization effort, but they may differ."
2. Preliminary field tests are carried out to identify sensitive variables
noted above.
B. The nature of more extensive tests, conceived as tier 4, can only be
estimated after doing part A. This includes the regional conditions needed for
the studies, species and species assemblages to be included in the tests,
and range of treatment levels expected.
1. Establish the minimum information necessary for conducting a valid risk
assessment.
2. Research needs to be conducted to determine under what conditions
test data are adequate without tier 4 information.
In the summary, neither the government nor private sector alone has the resources
or expertise to accomplish the objectives set forth in these recommendations and
goals. A group effort will be required and this must include mechanisms for the
sharing of data and improved communications between all involved.
Einhellig: "Is there anything else?"
Brown: "We need another meeting."
(VOTED TO BE INCLUDED IN THE SUMMARY)
246
------- |