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.
                                          =E£ED 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)
 £_

 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 -
    £2   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
	 £

   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

-------