DIALLATE FACT SHEET
Common Name;    Oiallate
Chemical  Name;  (S (2,3-0ichIoroaI IyI 1  diisoproyIthiocarbonate)
Structure:
H            CL        0      CH3
     OC                      I
CL           CH2 - S - C - N  CH
                              I
                              CH3
                                            CL
                                                C=C
 CL        0
 CH2 - S - C - N

els Isomer
        trans Isomer

Molecular Formula;  C.gH.,CL

Chem i s try;   Dial late, is the name applied to a thioarbamate
            herbicide known by the trade name Avadex.
            It  is a 'iquid which is formulated as an emulsi-
            ftable concentrate (4 pounds/gallon) and as a
            granular product (10$).   Its molecular weight
            Is 270.2.  Two Isomers, els and trans forms are
            known.

Character i st i cs;  Diallate is a volatile, amber colored liquid
                  with a melting point of 25-20C.  > l-t  is soluble
                  tn most organic solvents but only  slightly
                  so IubIe in water.

-------
    b).   Teratoqenic Effects






         A final  report concerning reproductive effects



         in 3 generations of rats is expected by the



         Agency sometime in 1980.  This study will be



         evaluated by the Agency and results included in



         Position Document 4.





Number of Products and Registrants





     Three registrants and eight products.






Product ion





     Confidential under section 7(c) and  10 of  FIFRA.






Principal Manufacturer





     Monsanto Co. is the sole  importer of  technical  dial late  in



     the U.S.
Uses
     Control of wild oats as  a  pre-emergence  herbicide  in



     sugar beets  (major.use), flax,  lentils,  dry  peas,



     alfalfa, barley, corn, potatoes,  and  soybeans.
Human Exposure
     Principal  human  exposure  is  by  dermal  exposure.   The



      inhalational  exposure  is  low  and  not  of  primary



     concern. The  principal  population  subject  to  these



     exposures  are applicators.   The risk  to  the general



     population  from  dietary exposure  is  low  and the  worst




     case  estimate is  at  10  .

-------
            Risk Criteria Met or Exceeded





I)   RPAR Triggers






    a)   Oncogen i c i ty






        Studies by- several  laboratories  indicate ingestion



        of dial I ate by different rod-ent  species results in a




        significant increase in numbers  of carcinomas and



        other tumors in experimental animals when compared



        w i th controIs.






    b)   Mutagen i c ? ty






        Studies employing the Ames test  for mutagenicity  in



        microbes indicated dial late induced gene mutations



        in SaImoneI I a typh ? mur i urn but not  in the lass



        sensitive bacterium Escher i ch i a  col i .   Induction of




        mitotic abberations were found  in  the yeast, Sac-



        charomyces cerev i s ? ae«  In a mammal ian ceI I  culture



        system, (mouse lymphoma) forward mutations were




        found at the TK locus.  Thus, evidence  is presented



        for in  vitro mutagenic  effects  by  dial late.






2)   Non Triggers:  Areas of Possible Adverse Effects






    a )   Neurotox i c i ty






        The Agency has determined that  a neurotoxic




        effect  may occur upon exposure  to  dial late.   It




        has calculated the annual  applicator exposure



        and determined that the effect  level is 600




        times greater than the  exposure  level.

-------
                DIALLATE
          Position Doc went 2/3
               April, 1980

Office of Pesticides and Toxic Substances
     Environmental Protection Agency
          401 'Mf Street, S.W.
         Washington, D.C. 20460

-------
ACKNOWLEDGEMENTS
EPA Prop ect Support Team

James E. Wilson, Jr.,  Senior PM, SPUD
Minnie R. Sochard, Ph.D.,  PM, SPRD
Karl 0. Bayer, OGC
Tom Edwards, "*"
Roger Gardiner,  BED
Janice Jensen, BED
Dave Johnson,  BED
Charles Lewis,  BFSO
Richard Stevens, BSD
Linda V. Zygadlo, BFSD

-------
                             TABLE OF CONTENTS

                                                            Page Number
I.   Background                                              1-1
     A.   Introduction                                       I-1
          1.   Organization of Position Document             1-1
          2.   Chemical and Physical Characteristics         1-1
          3.   Registered Uses                               1-1
          4.   Tolerances                                    1-2
     B.   Applicable Sections of FTFRA                       1-2
     C.   The "RPAR" Process                                 1-3
     D.   Regulatory History                                 1-3
     E.   Basis for the Rebuttable Presumption               1-4
II.  Analysis and Assessment of Risk                        II-1
     A.   Analysis of Rebuttal Arguments
            for Oncogenicity                                II-1
          1.   Rebuttal Pertaining to National
                 Cancer Institute (NCI) Rat
                 Study (Litton-Bionetics)                   II-1
          2.   Rebuttal Pertaining to Industrial
                 Bio-Test (IBT) Rat Study
                 (Sponsored by Monsanto Company)            II-5
          3.   Rebuttal Pertaining to NCI
                 House Study (Innes Study)                  11-11
     B.   Analysis of Data Submitted Since PD  1
            for Other Possible Adverse Effects              11-14
          1.   Mutagenicity                                 11-14
          2.   Neurotoxicity                                11-16
          3.   Reproductive Effects                         11-17
     C.   Exposure Analysis                                 11-18
          1.   Spray Applicator Exposure                    11-18
               a.   Spray Applicators                       11-18
               b.   Granular Applicators                    11-19
          2.   Dietary Exposure                             11-19
     D.   Risk Assessment                                   11-19
          1.   Oncogenic Effects                            11-19
          2.   Hutagenic Effects                            11-24
     E.   Risks Associated with Alternative  Chemicals       11-27
III. Benefit Analysis                                       III-1
     A.   Sugar Beets                                       III-2
     B.   Flax                                              III-7
     C.   Lentils                                           III-7
     D.   Peas                                              III-8
     E.   Minor Uses                                        III-9
IV.  Development of Regulatory Options                      IV-1
     A.   Introduction                                      IV-1
     B.   Salient Risk/Benefit Considerations               IV-1
          1.   Salient Risk Considerations                  IV-1
          2.   Salient Benefit Considerations               IV-2

-------
     C.   Risk/Benefit Analysis Options                       IV-3
          1.   Dietary Risk/Benefit Analysis                  17-3
          2.   Applicator Risk/Benefit Analysis               IV-4

     D.   Regulatory Options                                  IV-6
          1.   Option 1                                       IV-6
          2.   Option 2                                       IV-7
          3.   Option 3                                       IV-7
 V.  Proposed Regulatory Decision                             V-1
VI.  References

-------
                      DIALLATE:  POSITION DOCDMENT 2/3

I.   Background

     A*   Introduction

          1 .   Organization of Position Document:

               This Position Document contains five parts*  Part I is the
introductory  section.   Part II is  an evaluation of the risk of diallate.
It includes descriptions  of the  relevant data on toxicity, exposure, and
the Agency's  present  risk assessment.  Part III is a summary of the
economic benefits of  diallate.   Part IV describes the range of the
regulatory  options considered by the Agency.  Part V puts forward the
Agency's reconmended  option.

          2.   Ch«"1cal and Physical Characteristics

               Diallate ( S- ( 2 , 3-Dichloroally 1 ) diisopropy 1 thiocarbamat e ) is
a thiocarbamate  which is also known by the trade name AVADEX0.  Diallate
acts as a pre-emergence selective  herbicide.  It is an amber-colored liquid
which  is formulated  as an emulsifiable concentrate (4 pounds/ gallon) and as
a granular  (10%).—   Its molecular weight is 270.2.  Its structural
formula is  as follows:
            Diallate  "trans looner                   ds isoaer

            H           a           CH3     ci        a           CH3
                     ^                                 f,
           d       ^CH,- 3-G-«-CH
                               XCH'CH-                       NCH— CH
                                 \
           3*    Registered Pses

                Monsanto Agricultural Products Company  is the  sole producer
 of  technical-grade diallate.  As a registered pesticide, diallate is  cur-
 rently used to control wild oats in sugar beets, flax, barley,  corn  (grain
 and silage),  forage legumes (alfalfa, sweet, red and alsike clover),
     As a granular, diallate is registered for use on  sugar beets  only.
                                     1-1

-------
lentils, peas, potatoes, safflower,  and  soybeans.   In combination with
pebnlate or cycloate, it is used  also  to control other grasses and broad-
leaf weeds in sugar beets.  Diallate is  incorporated into the soil in the
fall before the freeze or in the  spring,  either before or after seeding,
bat before emergence.  Usually only  one  application is made by the user per
season.

               Eight federal registrations of products containing diallate
are held by three registrants.

          4.   Tolerances

               Tolerances for diallate in or on raw agricultural
commodities are listed in 40 CFR  180.277 as follows:  negligible residues
on alfalfa  (fresh and hay), barley (grain, forage, and straw), clover
(fresh and hay), field corn grain,  fodder and forage, flaxseed, lentils,
peas, pea forage and hay, potatoes,  safflower seed, soybeans, soybean
forage and hay, and sugar beet roots and tops at 0.05 part per million.

     B.   Applicable Sections of  FIFRA

          The Federal Insecticide, Fungicide and Rodenticide Act (7 U.S.C.
136 et seq.), as amended, confers authority on EPA to regulate pesticide
products.   Section 3 (a)  of  the Act requires all pesticide products to be
registered  by the Administrator  before they may be sold or distributed.
Before the Administrator may register  a pesticide, however, he must
determine that  its use will not  result in "unreasonable adverse effects on
the environment," defined in Section 2(bb) of FIFRA to mean "unreasonable
risk to  man  or  the environment,  taking into account the economic, social,
and environmental costs  and benefits of the use of any pesticide."  In
other words, any registration  decision must take into account both risk and
benefits from the pesticide's  uses.

          Section 6(b) of  FIFRA  authorizes the Administrator to issue a
notice of intent to  cancel  the registration of a pesticide or to change its
classification  if it appears to  him that the pesticide or its labeling
•does not comply with the provisions of  [FIFRA] or, when used in accordance
with widespread and  commonly recognized practice, generally causes
unreasonable adverse  effects on  the environment."  Thus, the Administrator
may cancel  the  registration of a pesticide whenever he determines that  it
no longer satisfies  the  statutory standard for registration; this standard
requires, among other things,  that the pesticide "perform its intended
function without unreasonable  adverse effects on the environment"   [FIFRA
3(c)(5)(C)].  Be may also  cancel the registration of a pesticide if its
labeling does not comply with  the misbranding provision of FIFRA, which
requires the labeling to contain certain language  "adequate to protect
health  and  the  environment* (FIFRA 2(q)).
                                    1-2

-------
     C«   The "KPAR* Process

          To implement its authorized  functions,  t*vt Agency has designed
the Rebuttable Presumption Against Registration (RPAR)  process, which
involves gathering data on the risks y«* benefits associated with the uses
of suspect pesticides. By allowing all interested parties to participate by
submitting information, the process enables EPA to make balanced decisions
concerning problem pesticides.

          If the presumptions of  risk  are not rebutted, the evidence
pertaining to benefits must be evaluated and considered together with the
evidence pertaining to risk.  Various  risk reduction methods and their
costs are analyzed.  The Agency then determines whether the pesticide may
be regulated so that a balance is achieved between risks and benefits.  If
the statutory balance cannot be reached for a use, the registrations for
that use must be cancelled.

     D.   Regulatory- History

          The first registration  for a product containing diallate was
granted to Monsanto on December 9,  1960 (EPA Registration Number 424-119).
Use of diallate on corn", forage legumes, lentils, peas, potatoes,
safflower, and sugar beets was approved.  Because no residues were detected
in or on these raw agricultural commodities the use of diallate was
accepted under the "no residue-zero  tolerance" concept.

          In  1954, the Food,  Drug,  and Cosmetic Act was amended to provide
for the establishment of tolerances  or exemptions from tolerances for
pesticide chemicals in or  on  food crops.  The amendments also provided for
the establishment of a tolerance  "at zero level" if data showed that no
detectable residues were present  in the treated crop at the time of
harvest.  In  1965 the National Research Council of the National Academy of
Sciences recommended that  the "no residue-zero tolerance" concept be
abandoned.  This recommendation was  based on the fact that this concept, as
applied in the registration and regulation of pesticides, had become
scientifically and administratively untenable.  Among the reasons given by
the council were that analytical  methodology had improved, and that small
levels of pesticide residues  had  become detectable (Pesticide Residues
Committee Report on "No Residue"  and "Zero Tolerance", NAS-NRC, June 1965).

          In  1966 the OSDA began  to  phase-out this concept (FR April 13-,
1966).  Registrants of products under  this concept were given  (through a
series of 1 year extensions)  until  1971 to convert the "no residue"
tolerances to finite tolerances.

          The Council also recommended that pesticides previously regulated
under the "no residue* concept be regulated on the basis of "negligible
residue" tolerances.  These tolerances could be established by supplying a
limited amount of data.  It was concluded that a negligible residue is the
amount which will produce  no  effects in test animals, which (effects) are
                                     1-3

-------
indistinguishable from control animals.  The  tolerance,  in many cases, may
reflect the sensitivity of the method and require  only two 90-day
(subacute) animal studies. Such was the  case  when  the tolerance for
negligible residues was established for  diallate.

          During the phase-out period, Monsanto  filed a  petition (PP
7F0607) requesting a tolerance of 0.3 ppm on  corn,  forage legumes,  lentils,
peas, potatoes, safflower, and sugar beets.   That  petition was subsequently
withdrawn and a second petition was filed (PP 9F832)  requesting a tolerance
for negligible residues of 0.05 ppm on the above crops.   Tolerances for
negligible residues of diallate were established in or on the above raw
agricultural commodities on July 28, 1971 (40 CFR  180.277).

          On May 31, 1977, the Agency issued  in  the Federal Register  (42
FR 27669) a notice of rebuttable presumption  against registration and
continued registration of pesticide products  containing  diallate.
Diallate:  Position Document  1, published together with  the RPAR notice,
explained the background and  supporting  data  for the presumption of risk
cited in the RPAR notice.

          Following the publication of the RPAR  notice,  Monsanto Company
requested and was given a 60-day extension of the  rebuttal period.   The
extension was granted to all  registrants and  interested  parties.

     K.   Basis for the Rebuttable Presumption

          The rebuttable presumption against  registration and continued
registration of pesticide products containing diallate was issued on the
basis of oncogenic effects in test animals as a  risk criterion [40 CFR
162.11  (a)(3)].  Specifically, the presumption was based on the following
three long-term feeding studies which indicated  that diallate is poten-
tially  oncogenic:  1) a National Cancer  Institute  (NCI)  Ulland et al.,
1973) rat study (Litton Bionetics), 2) an Industrial Bio-Test (IBT)
(Keplinger. 1976a) rat study  (sponsored  by Monsanto Company), and 3) an NCI
mouse study (Xnnes et al., 1969).—

          Data from the rat study conducted by Ulland and coworkers at
Litton  Bionetics were verified by the MCI and reviewed by the EPA
Carcinogen Assessment Group (CAG).  The  GAG concluded that the Litton
Bionetics study showed a statistically significant increase in malignant
tumors  at the highest dose in male rats  and in carcinomas at the higher
dose in female rats.

          Industrial Bio-Test concluded  from  its rat study "that the
neoplastic lesions noted in the test and controls  were considered normal
for rats of this age and strain."  In its evaluation, the CAG concluded
that "the Industrial Bio-Test study in rats showed a statistically
significant excess of mammary carcinomas in females."
                               1-4

-------
          The GAG also concluded  that  the  mouse study conducted by Innes
and coworkers showed an  increased incidence of liver-cell and pulmonary
tumors*  This study was  the basis for  the  Mrak Commission Report recommen-
dation that human exposure to  diallate be  minimized because there was
evidence of tumor induction in mice*

          Respondents were given  an opportunity to rebut the presumption
against diallate by showing  (1) that the Agency's initial determination of
risk was in error, or (2) that given current use patterns, exposure to
diallate is not likely to result  in any significant chronic adverse effects
[40 CFR 162.11(a)(4)].   Respondents were also invited to submit evidence on
behalf of the social, economic, and environmental benefits of the use of
the pesticide [40 CFR 162.11(a)(5)(iii)].
                                     1-5

-------
II.  Analysis and Assessment of Risk

     A.   Analysis of Rebuttal Arguments  -  for Oncogenicity

          The Agency has analyzed the  rebuttal-comments submitted to it in
response to to the presumption of oncogenicity—  and responded to these
comments in *•*»*« section.  From the analysis  of rebuttal comments, the
Agency has concluded that  the oncogenic presumption against diallate has
not been rebutted and that humans are  subj ect to the risk of developing
cancer from the use of diallate.

          1.   Rebuttal Pertaining to  national Cancer Institute (NCI) Rat
               Study (Litton Bionetics)

               a.   Errors in Tabulated Data

                    Monsanto Company ncted  errors in the data cited in the
EPA1s Carcinogen Assessment Group  (CAG)  (Albert, 1979) report and in
rebuttal presented a retabulation of the  raw  data (Monsanto Co.,
#1A[30000/15]).

                    The Agency  acknowledges Monsanto's rebuttal on this
point.  The CAG has re-evaluated the raw  data and has"" corrected its report
accordingly.   However, these corrections  do not change the Agency's
interpretations of the study.

                    The CAG's revised  tabulation of the data (shown in
Tables  II-1 and IX-2) differs somewhat from the data presented by Monsanto.
Most of these differences  are due  to  the  classification of gliomas and
leukemias as  sarcomas by  the CAG and  as  carcinomas by Monsanto.  These
differences in classification are  unimportant since the Agency bases its
regulatory decisions on  oncogenic  risks  which include all tumors  (Albert,
1979a).

               b.    Statistical Difference in the Malignant Tumors in Males

                    Monsanto Company contended that according to their
analysis  of the data, there  is  no  statistical difference in the number of
malignant tumors  in  the  diallate-treated male rats  (both dose groups
combined)  compared to pooled  controls (13/52 treated  vs. 11/64 controls,p *
.21).   Monsanto asserted  further that there is no statistical difference in
the number of malignant  tumors  or of  sarcomas in either treated group of
male rats (high or low  dose) as compared to control groups  (Monsanto Co.,
t1A[30000/1S]).

                     The  respondent acknowledged an  increased incidence of
carcinomas  in the high-dose male group as compared  to the controls, but
submitted that the incidence  in the low-dose male group was not different
from either  control group (Monsanto Co.,  #1A[330000/15]).
 2/   In addition to the above-mentioned rebuttals, the Agency  received
      16 responses pertaining to the benefits of diallate.  Comments
      submitted on benefits are addressed in Section III,  "Benefit  Analysis."

                                  II-l

-------
       Table II-1.  Nd Rat Study, Incidence of Malignant  Tumors  in
                    Hale Bats Ingesting Avadex  (Diallate)—
        Comparison of Low- and High-Dose Groups  to Pooled Controls

            NO. of Bats with    No. of

Pooled Control
Low Dose
High Dose
4/64
3/26
4/26
(6%) .
(12%)^
(15%)
7/64
1/26
4/26
(11%)
(4%)
(15%)
11/6,4
4/26
10/26
(17%)
(15%)f/ /
(38%)-
* Revised GAG tabulation.   (Albert,  1979a)
—. Gliomas and leukemias were  counted as  sarcomas.
G/
—' Rats with carcinomas did  not  have  sarcomas.
^. Corrected for survival.
—' Two of these rats with carcinomas  had  metastases (carcinomas of the
   prostate metastatic to lung and lymph  nodes;  islet cell carcinomas of
 . . the pancreas with metastases  to the heart).
—  The tumor incidence in the  high dose group compared to the pooled
   control group is statistically significant (p - .032) (Fisher Exact
  . Test).
** The total incidence of  10/26  includes  2 unclassified malignant
   tumors in the subcutaneous  tissue.
                                 II-2

-------
       Table II-2.  NCI Bat Study, Incidence of Malignant Tumors In
                    Female Bats Ingesting Avadex  (Diallate)—
      Comparison of Low- and High-Dose Groups  to  Pooled Controls

            No. of Bats withNo. of Rats,withTotal No.  of Rats
              Carcinomas          Sarcomas—        with Malignant TumorsS/
.
Pooled Control
Low Dose
High Dose
3/64 (5%}.y
2/26 (8% J-2'
5/26 (19%)-'
4/64 (6%)
2/26 (8%)
0/26 (0%)
7/64 (11%)
4/26 (15%)
5/26 (19%)
* Revised GAG tabulation.   (Albert,  1979a).
—, Gliomas and leukemias were  counted as  sarcomas.
—  Rats with carcinomas did  not  have  sarcomas.
d/  .
—  The two rats with carcinomas  of  the mammary  gland had aetastases to
  . the lungs.
-f The tumor incidence in the  high  dose group compared to the pooled
   control group  is statistically significant (p -  .042) (Fisher Exact Test)
                                 n-3

-------
                    The Agency rejects Monsanto's rebuttal  on this point.
The GAG analyzed twv incidence from each treatment  group  individually,
whereas Monsanto Company combined the data from the high- and low-dose
groups and in combining the data the significance found in  the high dose
group was masked*  The GAG re-evaluation of  the data  demonstrated a statis-
tically significant increase  in total malignant tumors in male rats of the
high-dose group as compared to pooled controls  (10/26 treated vs.11/64
controls; p - .032)   (Refer to Table 11-1}  (Albert,  1979a).

               c.   Statistical Significance of Carcinomas, Sarcomas,
                    and Total Malignant Tumors  in Females

                    From their analysis of the  data on female rats,
Monsanto Company  concluded  that there  is  no statistically significant
increase in carcinomas, sarcomas,  or total malignant tumors in either
treated group  (Monsanto Co.,  *1A[30000/15]).

                    The Agency rejects  this rebuttal attempt.  From the
CAG's  revised tabulation of the data,  it  is apparent that there is a
statistically  significant increase of  carcinomas in the female rats of the
high-dose  group as compared to pooled  controls (5/26 treated vs. 3/64
controls;  p *  .042).  (Albert, 1979a.)   (Refer to Table  II-2.)

               d.   Low number of Tumors  in High-Dose Male Rats

                     Monsanto Company asserted that the number of tumor-
bearing rats  in  each diallate-treated group is lower than  in either  control
 group, and emphasized that the high-dose group has the lowest number  of
 animals with tumors (Monsanto Co., #1A[30000/1S]).

                     The Agency rej ects this point of rebuttal.   The  number
 tabulated by the GAG for the controls was 11/64 (17%) as compared  with  10/
 26 (38%)  and 4/26 (15%) in the high-and low-dose male groups, respectively.
 The individual numbers of animals possessing tumors  is not relevant.  The
 appropriate comparison is the percentage of animals  which  have tumors.   The
 percentage of animals with tumors in the high-dose group is  statistically
 greater tfa*n the percentage of animals with tumors in the  control  group.
 (Albert, 1979a.)

                e.   No Apparent Effect of Diallate on the  Formation  of
                     Individual Tumor Types

                     Monsanto contended that an evaluation  of individual
 tumor types/sites is necessary to conclude  that a compound is  carcinogenic,
 and that there was no apparent effect of diallate on the formation of
 individual tumor types (Monsanto Co., #1A[30000/15]).

                     The Agency rejects this rebuttal attempt.  Although
 there was no statistically significant incidence of  individual tumor types,
 there was a statistically significant increase of total  malignant tumors in
                                 rr-4

-------
male rats of the high-dose group relative  to pooled controls.  Moreover,
there was a statistically significant  increase  of  carcinomas in female rats
of the high-dose group compared to pooled  controls.  The use of total tumor
data in evaluating carcinogenicity is  discussed in a recent Interagency
Regulatory Liaison Group (IKLG) document as follows:

                    "At the present timn there  is  considerable
                    uncertainty about  the  interpretation of carcinogenic
                    responses  in terms of  the total tumor yield in contrast
                    to the response  in terms of a statistically significant
                    increase of tumors in  specific target organs or
                    tissues. Traditionally, carcinogens have been
                    recognized in human and aT•*m*"* studies by a decisive
                    increase in tumors of  target organs.  However, it is
                    conceivable that a generalized increase in total tumor
                    yield, in  the  absence  of an excess incidence in one or
                    more target tissues', could occur, for example by a
                    promoting  effect that  generally increases the
                    spontaneous incidence  of tumors in test animals or by
                    the  action of  a multipotent carcinogen whose response
                    did  not  reach  statistical significance in any one organ
                    even at  the maximum tolerated dose.— "

           2.    Rebuttal  Pertaining to Industrial Bio-Test  (1ST) Rat
                Study  (Sponsored by Monsanto Co.)

                The  existing  registrations  and tolerances for diallate were
 established and supported  by data  contained in  studies conducted by
 Industrial Bio-Test Laboratories  (IBT) for Monsanto.

                In 1977,  subsequent to the publication of Position Document
 1,  the Agency,  in the Office of Pesticide Programs, established a
 Toxicology Data Audit Program (TDAP)  to assure  the  reliability and
 integrity of  data supplied to  the  Agency by pesticide manufacturers.  These
 data are the  integral component of the information  upon which pesticides
 are registered and  tolerances  are  established in the United  States.

                Industrial  Bio-Test Laboratories (IBT) was  one of the
 initial laboratories  audited jointly with the Food  and Drug  Administration.
 The IBT laboratory performed a large volume of  testing utilized by the
 Agency in its regulatory decision-making.   During an audit at the facility,
 numerous significant  departures from acceptable laboratory practice were
 noted.  As a result,  the Agency decided to reevaluate all pivotal IBT
 studies used in support of tolerances.  The workload of this evaluation
 program was shared with the  r*-n»M an Government in  cases where chemicals
 were registered on identical data bases.  The IBT  studies  for diallate  were
 evaluated by Canada and the  results are shown in Table II-2a.
 3/ IRLG (February 6, 1979), Scientific Bases for Identifying Potential
    Carcinogens and Estimating Their Risks.
                                    IX-5

-------
Table II-2a.  Diallata 1ST Studies
Type of Study
   1ST Ho.
Status
Remarks
Fish & Wildlife

8-Day LC   Mallard Duck

8-Day Dietary LC-Q Bob-
  white Quail

8-Day Dietary LC_0 Bob-
  white Quail

8-Day Dietary I»C-Q Bob-
  white Quail

4-Day Fish Tox

Fish & Wildlife

Fish & Wildlife

Fish & Wildlife

Acute Studies

4-week Pilot oral  study

Pilot feeding study/mice

Acute Inhalation

Acute cholinesterase
651-3026

J-6672


J-6673


651-3025
Valid

Valid


Valid


Valid
665-3027
A-6675
A-6674
A-6849
8531-9714
8532-9820
59-13-3
8530-9030
Valid
Invalid No raw data
Invalid No raw data
Invalid No raw data
Valid
Invalid Incomplete raw data
Invalid No raw data
Valid
                                    II-6

-------
Table II-2a. (continued)..  Diallate IBT Studies
Type of Study .
Subacute Studies
Subchronic oral
Subchronic oral
Subchronic oral
Subchronic oral
IBT Ho.

59-13A
59-13-2
59-13
59-13-1
Status

Invalid
Invalid
Invalid
Invalid
Remarks

No raw data
Ho raw data
Ho raw data
Ho raw data
Chronic Studies

Teratogenicity/rabbits       651-5254       Invalid



Mutagenicity/mice            622-5252       Invalid

Heurotoxicity/chicken        85 80-9119      Valid

Heurotoxicity/chicken        8580-10813     Invalid

Chronic feeding/rats         622-5250       Invalid
Deficiencies in
experimental
design

Ho raw data
Ho raw data

Incomplete raw
data
                                   II-7

-------
               TDAP Evaluation of 1ST Rat  Study

               The IBT study sunitted by Monsanto  was  evaluated by the
Office of Pesticide Program's Toxicology Data  Audit Program (TDAP).  The
analysis of the study indicates that the 2-year feeding study is invalid
because of the lack of histopathology data,  food consumption data, organ
weight data on surviving animals, and dietary  preparation data to show
unequivocally that doses were correctly prepared.   Therefore, no scientific
conclusions can be drawn from +h±*  study.
               •«   Errors in Tabulated Data

                    Monsanto observed  errors  in the data cited in the GAG
report and in their rebuttal submitted a  retabulation of the raw data.
Monsanto contended there are no  statistically significant differences in
the number of animals with tumors  in the  diallate-treated groups as
compared to the controls (Monsanto Co., #1A[30000/1S] ) .

                    The GAG has  re-evaluated  the raw data and has corrected
its report accordingly.  The revised GAG  tabulation (shown in Tables II-3
and 11-4} now agrees with the retabulation submitted by Monsanto Company.
The data indicate that there is  no statistically significant increase
either of total tumors (benign and malignant) or of tumors of any
anatomical site in any diallate-treated group of male rats as compared to
controls.

                    As Table II-5  indicates,  however, female rats treated
with  100 ppm diallate  (middle dose) showed a  statistically significant
increase of benign mammary tumors  (p » .021)  and hence of total mammary
gland tumors (Albert,  1979).  Therefore,  the  last portion of this rebuttal
is rej acted.

               b.   Ho Statistical Increase in Mammary Tumors

                    The incidence  of mammary  tumors in diallate-treated
female rats is shown in Table II-5.  Monsanto Company pointed out that
there was no statistically significant increase in the number of mammary
tumors in the treated female rats  of the  high- or low-dose groups as
compared to the controls.  The respondent argued that the admittedly
statistically significant increase in  mammary tumor among the middle-dose
groups represents a random event,  since there was no dose response.
Monsanto pointed out further that  there is no statistically significant
increase of mammary carcinomas.   (Monsanto Co., f 1A[30000/15] )

                    As Monsanto  noted, the female rats of the middle dose
group exhibited a statistically  significant increase of total mammary
tumors (p - .021), which was attributable to  the statistically significant
increase in benign mammary tumors  (p - .021). This increase of benign
mammary tumors in the middle-dose  group may indicate an adverse effect,
although the possibility that this response may be a random event cannot be
                                 11-8

-------
    Table II-3.  1ST Bat Study, Incidence of Benign and Malignant
                      Tumors in Male Rats Ingesting Diallate--
Dose Group
No. of Bats with
Benign Tumors
Mo. of Bats with
Malignant Tumor:
                                                  a/c/
Total No. of Bats
with Tumors
Control
50
100
200
1/50 (2%)
0/50 (0%)
2/49 (4%)
1/50 (2%)
4/50 (8%)
7/50 (14%)
4/49 (8%)
6/50 (12%)
5/50 (10%)
7/50 (14%)
6/49 (12%)
7/50 (14%)
    */ Revised GAG tabulation  (Albert,  1979a).
    — . Rats with endocrine tumors  are not  included.
    —  Rats with both benign and malignant tumors  were tabulated as having
       malignant tumors. •
                                    n-9

-------
         Table II-4.
               ZBT Rat Study, Incidence of Carcinomas Ȥd. /  /
               Sarcomas in Male Rats Ingesting Diallate-'-'-'
                MO. of Rats with
Dose Group
No. of
Sarcoma
                                       with     Total No. of Rats
                                             with Malignant Tumors^/
Control
50
100
200
0/50 (0%)
3/50 (6%)
1/49 (2%)
2/50 (4%)
4/50 (8%)
4/50 (8%) .
3/49 (6%)-/
4/50 (8%)
4/50 (8%)
7/50 (14%)
4/49 (8%)
6/50 (12%)
    -' Revised GAG tabulation  (Albert,  1979a).
    —  Table II-4 is an elaboration  of  Total Malignant Tumors column in
      . Table II-3.
    —' Rats with endocrine tumors  are not included.
    —  Rats with both benign and malignant tumors were tabulated as having
    e/
malignant tumors.
One rat had metastatic fibrosarcoma to the  lung and liver,  another
to the lung.  Mo primary sarcomas were found in the two  rats.
                                H-10

-------
unequivocally refuted (Albert, 1979a).  However, the high percentage of
test ««^"«i« with tumors in the high dose group strongly suggests that
there is an adverse effect.

               c.   Spontaneous Pituitary Adenomas

                    Monsanto Company pointed out that  there  is a high
spontaneous incidence of pituitary adenomas in the  1ST rat study.  The
respondent pointed out further that there is no apparent significant
increase in the incidence of this tumor in any treated group,  nor any
linear increase by dose observed  (Monsanto Co., f 1A[30000/15] ) .

                    The Agency agrees  with 'this conclusion  (Albert, 1979).

          3.   Rebuttal Pertaining to  the NCI Mouse Study   (Innes Study)

               a.   Study Invalid

                    Monsanto Company claimed that the  Innes  study is
invalid for the  following two reasons:
                     1)    Only one dose level, the "QiHtnvm tolerated dose
 (MID),  was  used in the  study.  The rebuttal contended that the MTD has been
 redefined,  thereby making the doses used in the study higher than what is
 now considered the MTD  and invalidating the experiment.

                     2)    The experimental design necessitated the dosing
 of litter-mates.  The rebuttal contended that biological and statistical
 significance cannot be  drawn from this inferior experimental design.
 Monsanto suggested that a bias of inherited tendencies (e.g., predisposi-
 tion to hepatoma formation) cannot be eliminated because the litter mates
 were not randomly distributed among the test groups  (Monsanto Co.,
 *1A[30000/15).

                          The Agency rejects this rebuttal argument.
 Concerning Monsanto 'S point  1) above, the Agency points out that the MTD is
 defined in the NCI Guidelines as "...the highest dose of the test agent
 given during the chronic study that can be predicted not to alter tlje.
 animal's normal longevity from effects other than carcinogenicity."-

                          Innes et al. reported that  the MTD was
 selected on the basis of a series of studies in which the maximal level
 given in a-single dose, in 6 daily and in  19 daily doses, resulted  in  zero
 mortality .-
 —'  NCI Guidelines for Carcinogen Bioassay in Small  Rodents,  NCI Tech.
     Report Ser. No. 1, Feb., 1976, p.  14. U.S. Dept. of  Health,  Education
     and Welfare, PHS, NIH, NCI-CG-TR-1.

 £/  Innes, J.R.M. et al (1969) Bioassay of Pesticides  and Industrial
     Chemicals for Tumorigenicity in Mice:  A Preliminary Note.   J.  Nat.
     Cancer Inst. Vol. 42, p. 1102.
                                  n-ii

-------
    Table II-5.  IBT Bat: Study, Incidence of Benign and Malignant  Tumors  of
                 the Mammary Gland in Female Rats  Ingesting Diallate^
Dose Group (ppm)
Ho. of Rats with
Benign Tumors-
No, of Rats with
Malignant Tumors
Total No. of Rats
with Tumors
Control
50
100
200
14/50 (28%)
19/49 (39%)
24/48 (50%)
15/46 (33%)
5/50 (10%)
4/49 (8%)
5/48 (10%)
10/46 (22%)
19/50 (38%)
23/49 (47%) ,
29/48 (60%)-
25/46 (54%)
    ~. Revised GAG tabulation  (Albert,  1979a).
    —  Rats with both benign and malignant  tumors  were counted as
      . malignant.
    —  The tumor incidence in  the  100 ppm dose  group compared to the
       control group is statistically significant  (p » .021)  (Fisher Exact Test).
                                     11-12

-------
                         A comparison of the survival data in the carcino-
genicity study indicates that the number of 18-month survivors in the
diallate-treated groups was similar to that of the vehicle (0.5% gelatin)
and untreated control groups for each respective species/sex.  Hence the
doses given in the study were tolerated by the treated animals and
therefore did not exceed the MTD.

                         In reference to Monsanto's point 2) above, Innes
administered diallate to mice (in 0.5% gelatin) by stomach tube from 7 days
to 4 weeks of age.  Thereafter, the mice were weaned and were given
diallate (without a vehicle) in the diet.  Since the study began prior to
weaning, each test group (e.g., diallate-treated, positive or negative
control group) was comprised of litter-mates instead of a random assortment
of litter-mates, which was precluded by the study design.

                         Darin? the MRAK Commission review of this study,
which recommended that human exposure to diallate be minimized,—  Mr.
Carrol Weil presented a dissenting opinion which included criticism of the
non-random allocation of animals.  Monsanto cited  Mr. Weil's criticism  in
its rebuttal.

                         In response to Mr. Weil's criticism on non-
randomization, the MRAK Commission reported that "...the data were
reexamined on a litter basis, in keeping with the Epstein-Mantel approach,
rather than on the single-animals basis employed in the Journal of the
National Cancer Institute  report.  All compounds which had been judged
positive for tumor ^nduction (significant at the 0.01 level, or stronger),
remained positive.*—   Thus, although non-randomization of litter-mates
is not the optimal experiment design, there is no evidence in this study
that a bias existed for a  gene-tic predisposition to tumor occurrence
(Albert, 1979a).

               b.   Sex Specificity—Increased Hepatomas in  Male Mice

                    Monsanto Company acknowledged the statistically
significant increase in hepatomas in male mice in the Innes  study, but
contended that the apparent sex-specificity is unusual  (Monsanto Co.,
f  1A[30000/15]).

                    The Agency rejects this rebuttal point.  The data  from
the Innes study indicate a statistically significant increase in hepatomas
in both strains of male mice when compared with either their respective
matched (vehicle) control, negative control, or pooled negative control
6/
—   Report of  the  Secretary's  Commission of Pesticides and Their
    Relationship to Environmental  Health,  Parts  I  and II.   U.S.  Dept.  of
    Health,  Education and Welfare, Washington, D.C.   (1969),  p.  470.

-'  Ibid p.  483.
                                  n-13

-------
groups; and a statistically significant incidence in female mice of strain
X when compared with the pooled negative control group only.  Contrary to
Monsanto'a assertion, hepatic tumors are known to occur more  frequently  in
male mice th*n in females.- (Albert, 1979a).

               c.   Mo Statistically Significant Increase  in  Pulmonary
                    Adenomas

                    Monsanto Company argued that there was no statistically
significant  increase in pulmonary  adenomas  in either  sex of either strain
of mouse, whereas the GAG  report  indicated  a small  statistically
significant  increase of lung adenomas  in both sexes of  strain X and in
males  of  strain  Y  (Monsanto Co.,  #1A[30000/15]).

                    The Agency rejects this point of  rebuttal.  Having re-
analyzed  the data,  the CAG finds  a borderline statistically significant
increase  of  pulmonary  adenomas in the  diallate-treated males  of strain X as
compared  with the matched (vehicle) control group     (p - .051) and with
the  pooled control groups (p • .041).   (See Table II-6)  (Albert, 1979a).

      B.   Analysis of  Data Submitted Since PD 1  for Other Possible
           Adverse Effects

           1. Mutaqenicity

                In PD 1 the Agency, based on  the two available  studies,
 stated that it had insufficient data to indicate diallate is mutagenic.
 The two studies were one, an Ames  test in  bacteria, and the  second, a
 dominant lethal study in mice  (Keplinger 1974).  On the basis  of this
 conclusion, the Agency requested  comments  and information on diallate's
 mutagenic potential.  Additional  studies were submitted and  evaluated,  and
 the Agency now  concludes  that  diallate meets the criteria stated  in 40  CFR
 162.11 for mutagenicity by multi-test evidence.  The additional studies
 which led to  the Agency's finding are discussed below.

                 In  response to  the Agency's request for additional
 information with regard to the possible mutagenic  properties of diallate,
 Rinkus and  Legator (1977) submitted a study entitled "Mutagenicity of
 Diallate."  This study, an Ames  test,  investigated the mutagenic  effects of
 diallate in five strains  of  Salmonella.  Diallate  exhibited  mutagenic
 activity in strains TA 1535 and TA 100 which exceeded the mutation
 frequency of controls by  factors  of approximately  20 and 12  respectively.
 This  activity was  only observed  in the presence of a microsomal activation
 system.   No activity  was  observed in  strains TA 1537,  TA 1538, and TA 98 in
 identical tests.
      Stewart, H. L. (1976).  Comparative aspects of certain cancers.
      Chpt. 10 in Cancer,  A Comprehensive Treatise, Vol. 4, P. F. Becker
      (ed.), Plenum Press, Mew York.

                                   n-14

-------
                         Table II-6.  NCI Mouse Study (Innes Study)>   Lung Tumors in Mice
                                           Ingesting Avadex (Diallate)
                                                 Dose
                                                 Group
                                        Strain X
                                                                                     Strain Y
                                        Male
Female
Male
Female
                          Pulmonary Adenoma   Matched Control
                          Pulmonary Carcinoma
                                        0/16*(0%)  0/16 (0%)2/18 (11%) 1/17(5%)
                                        0/16 (0%)  0/16 (0%)0/18 (0%)  0/17(0%)
                          Pulmonary Adenoma   Negative Control
                          Pulmonary Carcinoma
                                        2/17 (12%) 1/18 (5%)2/18 (11%) 0/17(0%)
                                        0/17 (0%)  0/18 (0%)0/18 (0%)  0/17(0%)
                          Pulmonary Adenoma   Pooled Control
                          Pulmonary Carcinoma
                                        5/79**(6%) 3/87 (3%)10/90(11%) 3/82(4%)
                                        0/79 (0%)  0/87 (0%) 0/90 (0%) 0/82(0%)
                          Pulmonary Adenoma     560 ppm
                          Pulmonary Carcinoma
                                        4/16 (25%) 2/16 -"
Ul
* p » .051 (Fisher Exact Test) for the Incidence of pulmonary adenomas in
the treated group compared with the matched control.

** p m .041 (Fisher Exact Test) for the incidence of pulmonary adenomas in
the treated group compared with the pooled control.

-------
               In PD 1 an unevaluated  study by  Sikka and Florczyk (1978)
was mentioned.  The study investigated the ability of diallate to induce
nutations in four strains of Salmonella  typhimurium (TA 100, TA 1535, TA
98, and TA 1538) with and without a  rat-liver microsomal activation system.
The study has now been evaluated and found to show activity at the 1 ug per
plate level in the TA 100 and  TA 1535  strains indicating base-pair substi-
tutions with metabolic activation.   Diallate  did not cause nutation in
strains TA 98 and TA 1538  (frameshift  mutants).  These results confirm the
findings of Rinkus and Legator.

               Litton Bionetics, Inc.  (LSI)  (Brusick, 1977b) investigated
the effects of diallate  in  the L5178Y  mouse  lymphoma cell.  The study
concluded that "The test compound,  CP  15336,  induced forward mutation at
the TK  locus  in L5178Y mouse lymphoma  cells  in the presence of an uninduced
mouse liver S-9 metabolic .activation system.*  No dose-related effects were
observed in the absence  of  metabolic activation.

               Studies by SRI International  (Simon,  1978) for EPA show that
diallate does not induce gene mutation in Z^ coli  (WP2).  The bioassay was
designed to monitor induced genetic alteration in E. coli at the tryptophan
locus.  However,  tests  in ]S. coli are not as sensitive  as tests in
Salmonella and,  therefore,"positive findings may not be manifested  through
experiments  in this organism  (Sandhu, 1978).

                Diallate's potential to  cause primary DNA damage was studied
in two strains of Saccfaaromyces cerevisiae.  SRI  International  (Simon,
 1978)  employed strain D, to measure induced  mitotic recombination  and LBI
 (Brusick,  1977a)  used strain  D. to measure gene  conversion.  While  posi-
 tive results were reported for mitotic  recombinations  in the  SRI  study,  the
 iai study results were  negative.  However, this mixed  finding does  not
 detract from the finding of induction of mitotic  aberrations,  a supportive
 finding for mutagenesis.

                Monsanto submitted a dominant lethal study in mice in 1975
 (Xeplinger,  1974).  This study  done by  1ST was reported by Monsanto as  a
 negative study.  The TDAP program reviewed this study and found it to be
 invalid because of the  lack of  raw  data.

           2.   Neurotoxicity

                The Agency  concluded in  the PD  1 that diallate is neurotoxic
 at the dose  levels tested.  This conclusion  was based on an IBT study in
 chickens  (Keplinger, 1976b).  That  study did not provide a dose-response
 relationship nor a no observable effect level  which is needed to determine
 a margin of  safety.

                In  the diallate   PD  1, published on May 31, 1977,
 registrants  were given  180 days to  complete  appropriate neurotoxicity
 studies and  submit the  results  to  the Agency or face possible cancellation
 under  the provisions of FIFRA Section 6(b)(1)— .  Monsanto submitted an
  —   In 1978 FIFRA was amended to provide for suspension under Section
      3(c)(2)(B).
                                     11-16

-------
1ST study in which diallate was administered  to  chickens—'  at dose
levels ranging from 0.01 to 0.32 gmAg which  were  administered twice daily
for 3 consecutive days (Phillips,  1977). Twenty  days  following the initial
dose, alX surviving birds were again given the same dose  regimen.   Controls
were dosed with 0.32 got/kg corn oil and positive controls received 500 mg/
kg TOCP on day 0.

               All positive controls exhibited lesions typically associated
with delayed neurotoxicity (Phillips,  1977).   No such lesions were found in
the negative controls.

               Two test  birds,  one in  the  0.04 gmAg and one in the 0.16
gmAg group showed focal lesions of axonal degeneration and secondary
demyelination  in  the  sciatic  nerve.  While these lesions were described as
morphologically indistinguishable  from those observed in the positive
control  birds, the affected birds  showed no clinical signs which could be
characterized  as  delayed neurotoxicity prior to sacrifice.  Ho dose
response relationship was established nor were there any reasons given for
the  absence of lesions at the highest dose.

               Both  of these  IBT studies were reviewed by the TDAP.  The
original study, discussed in  PD1,  was validated and therefore does demon-
strate  that diallate given at 312 mgAg causes neurotoxic effects.   How-
ever, as stated  in TO 1, this study does not show any dose  response  with
which to determine ultimate safety.  The second IBT  study on neurotoxicity
was  declared invalid.  TDAP found that raw data were totally missing on
this experiment.

                The Agency1s concerns on neurotoxic effects  of diallate have
 not been addressed by the registrant.  While attempting  to  satisfy the
 needs of EPA, Monsanto has failed to take adequate precautions  to insure
validity of their data.  Therefore, although the  Agency  does not  have
 enough data to quantify the potential neurotoxic  risks of diallate,  it has
 calculated the annual applicator exposure.   The Agency has  determined that
 the effect level is 600 times greater than the  exposure  level.

           3.   Reproductive Effects

                Prior to the RPAR review the  Agency requested Monsanto to
 perform a 3-generation  reproduction study in rats.   However, no time limit
 was imposed on the registrant.  At the time  of  PD 1  only the data in the
 F° (parental) generation had been  submitted. The final  report is
 expected in 1980.  This study will be evaluated by the Agency and its
 results will be  included  in PD 4.
 —   The  study used 5  birds per dose group with dose given at 0.01,
     0.02,  0.04,  0.08,  0.16,  and 0.32 gmAg.
                                   H-17

-------
     C.   Exposure Analysis

          The Agency analyzes, exposure to a pesticide as part of the
overall risk assessment.  The complete exposure analysis includes
incremental exposures to various populations depending on  the route of
exposure (e.g. dietary exposure to the general population,  occupational
exposure to applicators, drift to farm families, etc.).  In compiling the
analysis, the Agency takes into account the use patterns and methods  of
application so all populations likely to receive exposure  are included  in
the analysis.

          1.   Spray Applicator Exposure

               a.   Spray Applicators

                    Information concerning  spray applicator exposure  was
provided by the Environmental Fate Branch of the Agency's  Hazard Evaluation
Division (BED) (Selim,  1978).  Diallate  is  applied primarily as an emulsi-
fiable concentrate with ground equipment by boom sprayer.   Spray applica-
tors are exposed to diallate by  1) dermal exposure during  the loading of
the sprayer *i\* during  the application process, and 2)  inhalation of  the
volatilized compound.

                    Exposure from diallate  has not been measured, conse-
quently the Agency used published data on triallate to prepare exposure
analysis for  diallate  (the mode  of application and chemical-physical
properties of triallate are  similar  to diallate).   To further check these
results, the  Agency validated  the assumptions  used for diallate by an
exposure analysis which was  published for paraquat.  The pattern of
exposure to paraquat  is similar  for  the  applicator working with diallate.
Therefore, the Agency felt  justified in  utilizing the paraquat data for
extrapolation to  diallate.   Both of  these  extrapolations produced very
similar  results.

                    The estimated dosages  were reported in mg/hr, then
converted  to  mg/kg/year. The  conversion from dermal exposure expressed in
mgAg/year to equivalent lifetime dietary exposure expressed in ppm in the
diet is  as follows:
           X »  60 kg (worker dermal exposure in mg/kg/yr) x 40 yr
                365 d/yr x2?0 yr x 1.5 kg/d
             *  6.26x10   x (worker dermal exposure in mg/kg/yr)
             m  lifetime dietary exposure in ppm
 where X is ppm in diet, 60 is average body weight in kg, and 1.5 kg is the
 average daily dietary intake.  A 40-year working history and a 70-year
 lifetime is assumed for the applicators.  This value (X) will be used to
 calculate lifetime probability in the Risk Assessment Section (II.0.1).
                                 11-18

-------
                    Table II-7 presents data on the absorbed dose in mg/kg/
year and parts per tiHii-jrm for spray applicators exposed  to  diallate.
Rubber gloves and coveralls were considered as protective clothing when
calculating the exposure reduction levels.

                    The annual exposure to applicators  of diallate is 0.516
agAg/y* (.0323 ppm) and 0.018 mgAg/yr (0.0011 ppm)  from dermal and
inhalation exposure respectively (Selim,  1978).  These  estimates are based
on 10% absorption from both routes (Gardner,  1980).   Based on the assump-
tion that dermal exposure would be reduced by a factor  of 4  if protective
clothing is used, the dermal exposure  level can be reduced to 0.13 mg/kg/yr
(8.1 x 10   ppm) (Selim, 1978).

               b.   Granular Applicators

                    An applicator's exposure  from  granular diallate
applications would be lower than an applicator's exposure using the
emulsifiable concentrate formulations.  With  the granular formulation there
is no chance of exposure from spray drift or  from  spray splash as there is
with the emulsifiable concentrate  formulations.  The  granules do not
adhere to the  skin as the emulsifiable concentrates would.  Additionally,
because diallate is applied during the late  fall or early spring in the
northwestern states, it is likely  that most  diallate  applicators would be
wearing clothing such as long sleeved shirts  and trousers to protect
themselves  from the cold as well as protect  themselves  against the
diallate.   However, for the brief  period  of  loading the  (pre-mixed)
granular diallate formulations  there  is a potential dermal and inhalation
exposure hazard to the applicator  due to  dust from the granules.  To
mitigate this  potential hazard,  the Agency  suggests the use of rubber gloves
and cloth face masks for applicators  during the loading process.

          2.   Dietary Exposure

               The human population encounters direct dietary exposure to
diallate residues through  consumption of  the following foods:  barley,
lentils, peas, soybeans,  sugar  beets, corn (grain), and  flax  (seed).
                        worst-case exposure was developed from tolerances
 established for residues of diallate in foods.  The FDA, in its Market
 Basket Survey,  has not analyzed raw agricultural commodities specifically
 for diallate.   It is assumed that residues are present in all individual
 raw agricultural commodities to the extent permitted by the tolerances and
 that the commodities are uniformly distributed throughout the country.
 Table II -8 presents the dietary exposure of the entire U.S. population to
 diallate.

                A second set of estimates were developed which were based on
 available information concerning the percentage of crops treated and were
 provided by the Agency's Benefits and Field Studies Division (Lewis, 1978).
 Table II -8 presents the exposure estimates when the percentage of crop
 treated is considered.
                                   11-19

-------
     D.   Risk Assessment

          1.   Oncogenie Effects

               The cancer risk assessment of diallate is based on  the
principles and procedures outlined in the EPA interim cancer  risk  assessment
guidelines (41 FR 21402, May 25, 1976).  These  guidelines  specify  that a
substance will be considered a "presumptive cancer risk when  it causes a
statistically significant excess incidence of benign or malignant  tumors in
humans or animals." Current and  anticipated exposure levels  are appro-
priate considerations for establishing cancer risk estimates.  These
estimates may be derived from a variety of risk extrapolation models such
as the log-probit and linear non-threshold models.
                                   11-20

-------
      Table II-7   Diallate Dermal and Inhalation Exposure to Spray
                  Applicators and Equivalent Lifetime Dietary Exposure
Estimated
Route of Dose
Exposure mq/hr
DEHMAL
Absence of
protective cloth-
ing (Assuming 2.58
10% dermal ...
absorption)—
With protec-
tive clothing
(Assuming 10% .65
dermal .
absorption )— —
INHALATION
Without mask .09

With mask
(10% of .009
exposure
without
mask)
Duration
of
Exposure



6-12
hr/yr.



6-12
hr/yr.

6-12
hr/yr.

6-12
hr/yr.


Dose
in
mg/kq/yr



.258
.516



0.065
.13

.009
.018

.0009
.0018


Dose in
ppm


-
2 x 10
3 x 10*


_3
4 x 10"
8 x 10"

6 x 10__
1 x 10
-5
6 x 10 .
1 x 10


1/   10% absorption is assumed in absence of data on diallate  (Gardner,
     1980).
2/   Assuming that protective clothing provides a four  fold reduction  in
""    exposure. (Selim, 1973)
                                  11-21

-------
     Table II-8.  Annual U.S. Population Dietary Exposure to Diallate, Based
                    On Tolerance Levels and Percent of Crop Treated
     Exposure Based on
     100% of Crop Treated
     with Residues at
     Tolerance Levels
Source
                 Percent of Crop
                 Treated with
                    Diallate
Exposure Based on Actual
Percent of Crop Treated
with Residues at
Tolerance Levels
       pom
Barley
Lentils
Peas
Potatoes
Safflower*
Soybeans
Sugar beets
Corn, grain
Flax, seed
0.000013
0.00002
0.00035
0.00271
0.000013
0.00046
0.00001
0.0005
0.000013
0.10
38
10
0.46
100
0.20
14.3
0.009
3
0.00000
0.00001
0.00005
0.00002
0.00002
0.00000
0.00012
0.00000
0.00000







**
**
Total
0.004089
       0.00022
     *     100% is assumed  in  the absence  of  data .

     **    Mot actually  "0", remaining significant figures truncated.
                                  11-22

-------
               In accordance with these principles,  the  EPA Cancer
Assessment Group (GAG) (Albert,  1979b) developed  risk  estimates based on
several different models and a range of exposure  estimates.  After
reviewing the data sources- and the preliminary  risk  estimates,  the GAG
concurred in recommendations that the final quantitative risk estimates be
based on data from the MCI mouse study (Innes study) using the  one-hit
model.  The CAG used the Innes data because animals  in the Innes study were
fed the compound beginning at a  younger age *>»««  were  animals in the. NCI
(rat.) or IBT studies, and therefore provided the  most  sensitive animal upon
which to base the conservative analysis.

               To develop a risk estimate, CAG  evaluated the animal test
data and the human exposure data using several  different models.  They
selected the one-hit model because it provided  the most  conservative
estimate of potential risk to humans.

               As explained above, the animal bioassay data used for the
quantitative risk assessment were based on the  Innes oral feeding study in
mice.  In this study one treated group of mice  were  fed  560 ppm diallate in
the diet.  A statistically significant higher incidence  of hepatomas in
males of both strains X and Y was observed, as  compared  to matched controls
(see Table II-9).

               The proportion of hepatomas observed  in Strain X males was
used to calculate the slope parameter for the one-hit  model, adjusting for
background tumor incidence.  Therefore, using the proportion of hepatomas
in the matched control 97oup and the treated group,  the  one-hit slope
parameter is as follows:—

                   B - - In  [(1-Pt)/M-Pc)] /y
                              B  - - In  [{1-  13/16)/(1-0/16)J/570
                                 » 2.989. x 10

From the slope, the estimate lifetime probability can  be estimated from
the following equation.

                           P » BX -  (2.989 x  10~3)X
                   (where  X is the ppm in the diet from  actual exposure
                   and from equivalent dermal exposure as calculated in
                   Section II.C.1.)
 —        Where:  B *  slope  coefficient of the one-hit model
                  P »  (Pt-Pc)d-Pc)
                 PC *  Incidence  of hepatomas in control animals
                 Pt *  Incidence  of hepatomas in test animals
                  7 *  Test animal exposure (ppm)
                  x *  Potential  human  exposure (ppm)
                                 11-23

-------
Table II-9.  NCI House Study (Innes Study):  Liver Tumors (Hepatomas) in
             Mice Ingesting Diallata
Dose Group
Matched Control (vehicle)
negative Control
Pooled Control
560 ppm
Strain
X
Male Female
0/16*
1/17*
8/79*
13/16
0/16
0/18
0/87**
2/16
Strain
Male
1/18*
3/18**
5/90*
10/18
Y
Female
0/17
0/17
1/82
1/15
*  Statistically significant when p is £ .01.
** Statistically significant when p is £ .05.
                                  11-24

-------
               The exposure estimates for dietary  and applicator exposure
were factored into the above equation.  Risks from these  exposures.are
shown in Tables II-10 and 11-11.  A dietary worst-case risk of 10
occurs at tolerance levels assuming that 100% of the  crops  are treated.   It
is not likely that 100% of the crops will be treated  with diallate.   There-
fore using the projected percentages of treated crops from  Section  III,
results in a reasonable worse case risk of 10

               Risk to applicators without protective clothing is estimated
at 10~ , a relatively high risk.  With protective  clothing, risk to
applicators is improved to 10    (Table 11-11).

               Aggregated, these  risks imply that  1 in 10,000 applicators
might have a lifetime risk of developing a diallate induced tumor taking
into account both occupational and dietary exposure.   It  is estimated that
there are 2380 diallate applicators  (Selim, 1978). The general population
would have a lesser risk  (10-7),  but this is based on current usage and
should any increase in usage occur, the dietary risks would increase.

          2.   Mutagenic Effects

               While adequate evidence exists  to establish  the mutagenicity
of diallate in in vitro systems,  no  quantitative assessment of risk can be
made because of  (1) insufficient data from-mammalian  test systems(Mauer,
1978), and (2) no generally applicable method  has  been developed to
quantify mutagenic risk.——   Recent  studies have established £ strong
correlation between a chemical's  carcinogenic  potential and its ability to
induce mitotic recombination  (Sandhu,  1978).
—        The  Agency has  not yet developed a standard procedure for
           defining mutagenic risk in quantitative terms.  At the present
           time, much attention is being focused on developing a battery of
           test systems  and other data that are predictive of mutagenic risk
           in humans.  Until such time as more quantitative methods and
           procedures for  risk estimation are developed for each mutagenic
           endpoint of concern, the Agency will evaluate each mutagenic
           chemical on a case-by-case basis,  taking into account all
           available  test  data.  The approach taken by the Agency will of
           necessity  be  conservative in order to assure that man and the
           environment are protected from the risk of "unreasonable adverse
           effects" through the action of mutagenic agents.  The evolving
           nature  of  methodology in the field of mutagenicity testing
           dictates that the Agency will revise its risk estimation
           procedure  for future chemicals under evaluation as superior risk
           predictive models and other relevant information become
           available.  As  well, the Agency will revise its risk estimates
           for  chemicals which have previously been subjected to risk
           assessments if  additional more relevant test data and other
           predictive information are developed.
                                   11-25

-------
Table 11-10.  Lifetime Spray Applicator Cancer Risk  from Dermal and
                         Exposure to Diallate
                    Equivalent Lifetime
                    Dietary Dose Assuming
                    12 Hour *!«««« i Exposure        Lifetime Probability of
Route (ppm)
DERMAL
Without protective 3 x 10~
clothing*
With protective
clothing (10% _3
absorption)* 8 x 10~
INHALATION
Without mask 1 x 10
With mask ( 10%
of exposure _4
without mask)* 1 x 10
Cancer Due to Diallate

1 x 10~4
2 x 10~5
3 x 10"6
3 x 10~?
 *  10%  absorption is assumed in the absence of data on diallate (Gardner
   1980).
                                    11-26

-------
   Table 11-11.   Cancer Risk to D.S. Population from Dietary

                      Exposure to Diallate
Source
Lifetime Probability
of Cancer Based on
100% of Crop Treated
Residues at Tolerance
Levels
                                             Lifetime Probability
                                             of Cancer Based on
                                             Estimated Percent of Crop
                                             Treated with Residues
                                             at Tolerance Levels
Barley
Lentils

Peas
Potatoes
Safflower
-
Soybeans

Flax seed
Beet Sugar

Corn
Total
4
6

1
8
4

1

4
1

1
1
x 10~8
x 10~8
-6
x 10
X10-6
x 10~8
-6
x 10
-6
.9 x 10
-6
x 10
-6
x 10 *
X10-5
4 x ID'11
2 x 10~8
-7
1 x 10
-8
4 x 10
4 x 10~8
-9
3 x 10

Negligible
2 x 10~7

Negligible
4 x 10~?
                               11-27

-------
     E.   Risks Associated with Alternative Chemicals

          Several chemicals have been proposed as  alternatives should
diallate become unavailable.  In non-irrigated areas,  eptam would be the
major substitute with barban and dalapon used to a lesser extent.  In
irrigated areas cycloate and barban are the major  substitutes.  Eptam and
barban are substitutes used on flax.  Triallate, propham and barban are
alternative chemicals for use on lentils and peas.

          The data bases for the alternative chemicals are not complete.  A
complete list indicating the studies which the Agency has on hand appears
in Table 11-12.   Based on the studies reviewed  for all alternatives other
<-h»ti triallate, no unreasonable adverse effects  were found associated with
the proposed alternatives.  Because of the lack  of chronic studies no quali-
tative ranking of alternatives can be made with  regard to their relative
toxicities.

          The battery of toxicological tests performed on triallate include
mutagenic, chronic feeding, teratogenic,  and neurotoxic studies. Triallate
was found to exhibit mutagenic activity  in the Ames test in bacteria  with
metabolic activation.   It was also found to be positive in yeast when
tested for mitotic recombination.   Negative  findings were reported to gene
conversion in yeast, mouse  lymphoma,  and dominant lethal assays.  When
triallate is compared to  diallate in tests which are positive for both,
diallate  is at  least 3  times more active.

          Negative  findings for triallate were reported in a  chronic
feeding  study  in rats.   This  study was evaluated by TDAP and  found to be
valid only as  an oncogenic screen test because of deficiences in the
experimental  design.

          The  teratogenic and neurotoxic studies  on triallate were
performed by IBT.  These studies were evaluated by TDAP and found to be
invalid.
                                      11-28

-------
                      Table 11-12
TOXICOLOGICAL DATA SUMMARY FOR DIALLATE ALTERNATIVES (CHROMIC DATA)
CHEMICAL ONCOGENICITY

Trial late **Rat






Barban N/A
Cycloate N/A
Dalapon N/A

i EPTC*
NJ
i /\ 1 n_ 4_ _ __ \

MUTAGENICITY
Gene Conversion (-)
Mitotic Recom-
bination ( + )
Bacteria 4+ (with
activation)
2- (no activation)
Dominant Lethal (-)
Mouse Lymphoma (-)
N/A
N/A
N/A



CHRONIC REPRODUCTIVE
FEEDING TBRATOGENICITY EFFECTS


"Rat (-) Rabbit (Invalid) N/A





Dog (-) N/A N/A
Rat (-) N/A N/A
Rat (-) Dog (-) N/A
Rat (-)


NEUROTOXICITY


Chicken (2 Inval
Studies)




N/A
N/A
N/A



(+) Positive Study
(-) Negative Study
* EPTC has negligible residue tolerances, therefore no chronic data have been
  submitted.
** This IBT study listed under "Oncogenicity" and "Chronic Feeding," was
   reviewed by TDAP.  While the study was considered invalid as a chronic
   feeding study, it was found valid as an oncogenic screening study.

-------
III. Benefit Analysis

     As a pre-emergence selective herbicide,  diallate is used as a soil
treatment on field crops for control of wild  oats.   Because Monsanto is the
sole producer of diallate, production  and marketing data are confidential.
However, it is estimated that approximately 390,000 pounds active
ingredient of diallate were applied annually  to 319,000  acres of sugar
beets, flax, lentils, and peas, the major use sites of diallate between
1976 and 1978.  Small amounts of diallate are also  applied to potatoes,
soybeans, barley, corn, and forage legumes.—

     The Agency has received 16 submissions from registrants and interested
parties pertaining to the benefits of  diallate, in  particular as it is used
on sugar beets, lentils, and peas.  The Agency considered this information
in analyzing the benefits of diallate.

     For sugar beets, flax, lentils, and peas, the  estimates of acreage
treated, the identification of biologically viable  alternatives, and the
use data were based primarily on Assessment of the  Meed for Diallate in
Agriculture, USDA/State Assessment Team  on Diallate (September, 1977, and
1979 modifications).  The economic analysis based on these biological data
was prepared by Development Planning  and Research Associates, Inc. in March
of  1979.

     However,  lack of published  data  on  yield changes limited certain
aspects of  the analysis.  Expert opinion was  used in place of these data.
Alternate weed control  strategies  also lacked firm data, necessitating the
use of  expert opinion to  generate  impacts  of  alternate  control programs
(USDA/State Assessment  Team,  1977).

     The  alternatives to  diallate  were selected on the  basis of cost,
efficacy, and  availability.  Partial  budgeting was employed to assess the
economic  impacts  of  diallate  cancellation.—   The partial budgeting
methodology allowed  the change  in  the cost of weed control to be measured,
together  with  the effect  on  gross  returns  associated with substituting
alternative weed  control  practices while all  other inputs were held
constant.   The economic analysis  also assumed that in some instances the
cancellation of diallate  would  cause  growers  to shift production to
alternative crops.   In  these  cases,  net  returns to the  producer associated
with  the  production  of  alternative crops were evaluated.
J/   The USDA/State Assessment Team on Diallate (August 31, 1977) esti-
~~    mated that approximately 10,000 acres or less of each of these  crops
      were treated with diallate annually, accounting for 0.5% or  less of
      the total crop planting in each case.  The Monsanto submission  filed
      September 9, 1977, did not address the benefits of diallate  use on
      any of these minor crops.

 2/   The partial budgeting methodology allows the measurement of  the
      change in the cost of pest control and the effect on returns associated
      with substituting alternative chemical and non-chemical pest control
      practices into the budget with all other inputs held constant.
                                     III-l

-------
     If the major uses of diallate  on  sugar  beets,  flax,  peas,  and lentils
were cancelled, varied effects on producers  would result.  Because an
acceptable substitute herbicide  is  available,  peas and lentils  would
actually return more inmff" to producers  if  diallate were cancelled.  The
positive economic impact is based on increased yields which are due to a
decrease in phytotoxicity and better wild oat  control in some instances,
see further discussion in Section III.C.   Sugar beet producers  would suffer
an estimated adverse impact of $4.0 million.  Flax producers are expected
to experience an estimated $0.4  million economic loss.  The net adverse
impact upon all affected user groups is approximately $3.2 million
annually.—   These  aggregate economic  impacts  are summarized in Table
IXI-1.  The following subsections  (A through E) briefly explain the
economic impacts involved should the major and minor uses of diallate be
cancelled.

     A.   Sugar Beets

          Sugar beet production  subject to wild oat infestation is  located
in the non-irrigated acreage of  Minnesota and North Dakota, and irrigated
acreage  in Montana, Wyoming,  Idaho, Utah, Washington, Oregon, and
California.  As an  average  for  the  period 1976-1978, see Tables III-2 and
III-3, these nine states planted 995,000 acres of sugar beets annually  or
72.6% of the U.S. total  acreage.  Of the nine-state total, 418,000  acres,
or 42%,  were in Minnesota and North Dakota;  577,000 acres, or 58%,  were in
the  seven western  states  subject to wild oat infestation.

          Diallate  is  a major herbicide used to control wild oats in sugar
beets.   It  is  most  widely used in non-irrigated acres.  As an average
during the  1976-1978  period,  diallate was estimated to have been  applied to
 185,175  acres  of  non-irrigated beets, and to 35,800 acres of irrigated
beets.   In  total,  diallate  was  applied to 220,975 acres, or 22% of  the
total sugar beet  acreage subject to wild oat infestation.

           Annual  use  of diallate to control wild oats in sugar beets, as an
average for the period 1976-1978,  was estimated  to have  been 231,470 pounds
 (active ingredient basis)  in non-irrigated  areas and  44,750 pounds  in
 irrigated areas.   Total estimated use of diallate on  sugar beets  is thus
 276,220  pounds.

           Two formulations of diallate are  used  on  sugar beets.   Granular
diallate is applied to approximately  15% of the  treated  acreage while the
 emulsifiable concentrate is applied to approximately  85%  (Lewis,  1979).
The degree control of wild oats provided by each of these  formulations
 appears to be the same for the fall application.  There  is  some decrease in
 control  of wild oats when the granular is substituted for  the emulsifiable
 concentrate in the spring application.
 3/   This reflects an estimated  1.2 million  net  increase in revenues when
      triallate is substituted for diallate where it  is  also registered for
      these uses.
                                       IXX-2

-------
Table III-1. Annual Economic Impact of Cancellation of Dial late
on the Maj or Use Sites -




H
H
H
A
Site
Sugar
beets
Flax
Lentils
Peas
Maj or use
total
Extent
of Use
( thousand
pounds )
276.2
38.4
43.3
30.4
388.3
Units
Treated
( thousand
acres)
221.0
30.7
43.3
24.3
319.3
Percent of Aggregate Economic Impacts
Total Units User Market & Consumer
(million (million dollars)-'
dollars)
16.7 4.0 loss none
3.0 .4 loss none
38.0 .7 gain none
10.5 .5 gain none


Total Market Value
(million dollars)
500
59
34
23.6

-   Sourcei  Economic Analysis of Effects of Restricting Use of
                 Diallate on Sugar Beets, Flax, Lentils, and Dry Peas.
                 Development Planning & Research Associates, Inc.
                 March, 1979.

-^  1978 price levels

-------
                Table Hl-2.  Eatlaiated Annual aggregate MM coata Cot eliminating diallata on  irrigated  augar  beat*,  1978-1978
H
H




Coati

Aaroe* Material Application Yield ioai** Other ooata


With dlallata
Without dlallata
Cycloate
Prophaa
Dalapon
Barban
Paraquat
Untreated
Hand weeding a/
Added cultivation to/
Delayed aaedlng
Shift to other crops
Cub-total w/o dlallata
Mat change w/o dlallata
• USDA/Stata aeaaaeaont Study,
•* Utea per acre Iron USDA/Stata
a/ Charge at (20 per acre, baaed


33.800

8.200
300
1.800
1,800
1,800
3,600
8,200
10,700
900
6,400
43,700

percentage of

(I)
223,940

192.336
3.960
8.262
20,318
36,000
	
»__
— _
-._
	
361,076
33.536
original acreage
Aasaeamer.ti Study, 1977. Sugar
on interview
information from

W
125,658

28.782
1,053
2.862
2,862
2,862
	
»_.



38,421
(87,237)
eatiMted

W III
—

167,132
43,660
42,660
122,860
133,100
663,350
164,000
129.418 27,713
42,660
1.632,000
1,343,840 1,843,713
1,343,840 1,843,713
to be treated with diallata.

Total coate

(1)
331.198

388.430
47,673
33,784
146.040
171.962
663,330
164.000
137.111
42.660
1,652,000
9,489.040
3,117,852

baeta valued at >23.70/ton.
augar boat
proceaaora.

b/ North Dakota Crop and Livestock Reporting Service. 1977.

-------
Tabla III-J.  EatiiMted annual aggregate  uaac Impacta for alienating dlallata on non-lxvlgata* aug«r








a
n
H
1
**t












HSU

jlarbiotda Kit act
a. with diallata
b. diallata nubutltutaa
1 . <>|>ton
Cultural Uffttut
a. with diallata
1 . uaud
2. cultivate OK)
b. without diallata
I . 61!
•• Total ooat Qtanga in poat (fl " ill <«t (f) + 116,292 1.166.608 649.964 1,816,572 1.932.064 345.960 133.520 1.104,060 154,659 152.419 42,218 + 321,847 4,466.420 2.401.145 546,266 1,430,009 4.700,267 2,503,107 551.075 65,052* 1,430,009 --- 224.344 * 430,126 4,710,060 1,917.672 2.793,108. 5,140,906 2,051.739 2,900,177 101.070 yi.ild of 2 Toii/aoru at $2O.2il j.ur ton (aoucau of yield loua uatlMta, USUA/Stata Auuaatinant Toon, 1977)
-------
          Should diallate use  on  sugar beets be cancelled,  procedures will
substitute an integrated chemical and cultural strategy for wild oat
control.  In non-irrigated  areas, eptam is a major substitute for diallate,
with barban and dalapon also used for wild oat control.  In irrigated
areas, cycloate and barban  are the major substitute herbicides.  Sugar beet
producers would also  need to increase the amount of mechanical and hand
labor used to cultivate their  beets.  Some producers may experience yield
losses with alternative weed control strategies (including use of
alternative herbicides and  delayed seeding), while other producers may
shift to alternate crops.

          Economic impacts  would  result from changes in herbicide costs,
increased mechanical  and hand  cultivation costs, additional reseeding
costs, decreased yields, and  shifts to other crops.  Should diallate be
cancelled, sugar beet producers may experience estimated economic losses
potentially as high  as $4.0 million (Tables III-2 and III-3).  Of that $4.0
million impact, approximately  $3.1 million would result from adverse
effects in the irrigated  sugar beet areas of the Western U.S., and $0.9
million would be attributed to ramifications in the non-irrigated Red River
Valley area of Horth Dakota and Minnesota.  In the Red River Valley most of
the  adverse impact would  derive from possible increased hand labor costs
and  possible  increased mechanical cultivation costs.   In irrigated areas,
losses would  be nearly equally divided between yield losses and  lost
revenue resulting  from changes in crop production  from sugar beets to other
crops.

           The effect of diallate cancellation on a typical Red River Valley
farm with 185 acres  of sugar beets would  be  increased  costs of $870
annually.   The average cost increase  per  acre of  sugar beets  is  $4.69.
This, of  course,  assumes  the typical  farm would make all herbicide,
cultural,  and labor adjustments  in  the same  proportions as the entire
Valley  area.   Changes in net yield  are not anticipated.  The  average
producer  could expect his net returns to  land, management, and labor  to
decrease  by only 2.1%, from $41,102 to $40,232.

           The effect of diallate  cancellation on  a typical irrigated sugar
beet farm would be much more  severe,  with an average  impact  of $87.65  per
 acre of sugar beets.  On a farm  which ordinarily  treated  100  acres  of sugar
beets with diallate, the adverse  annual  effects would  amount  to  $8,765,
most of which would be due to reductions  in net  returns  (over variable
 cost).   For a typical farm, crop returns  over variable cost  would be
 reduced nearly 18%.  This  loss,  however,  is  to the individual farmer.
 There is  little overall yield loss  since  diallate is  not  used extensively
 in irrigated plantings.  In a switch to the granular  formulation of
 diallate alone for the control of wild  oats  in sugar  beets,  the  increased
 cost to growers in the short  run will be approximately $6-$7 per acre
 treatment, given currentprices.

           While the  cancellation of diallate may pose significant problems
 to the local grower, this  effect will not seriously reduce total U.S.
                                    III-6

-------
production because the  sugar from sugar beets represents only a small
percentage of total U.S.  sugar  consumed.  Over 50% of the U.S. sugar is now
imported due to the favorable tariff for imports.  A decline in profit
because of sugar imports  is  already reducing sugar beet acreages.

     B.   Flax

          United States flax production is concentrated in the states of
Minnesota, North Dakota,  and South Dakota.  This area produced 98% of U.S.
flax on an average of 1,025,000 acres between 1976 and 1978.  Approximately
3% of the total flax acreage (30,750 acres)  was estimated to have been
treated with diallate during this period - an estimated 38,440 pounds
(active ingredient basis)  of diallate annually.  The emulsifiable concen-
trate formulation is the  only form of diallate currently registered for use
on flax; however, granular diallate could be used on flax if it were
registered, but at slightly  higher costs.  Herbicide use on flax is very
limited because of the  extreme  phytotoxic reaction of flax to any herbi-
cide, including diallate. Diallate is, however, used in preference to
other herbicides.

          Eptam and barban are  the most common herbicides which can be
substituted for diallate  to  control wild oats in flax; however, both of
these chemicals have characteristics which limit their use on flax.  Eptam
is phytotoxic in flax,  and barban can only be applied when the wild oats
are at the two leaf stage (2-4  days).  If weather is bad for that period
the effectiveness of the  chemical is lost.  A cultural method of wild oat
control is delayed seeding;  however, this non-chemical method of weed
control reduces flax yields  by  about 33%.

          Cancelling the  use of diallate on flax is anticipated to result
in annual losses of approximately $0.4 million to flax producers.  The
economic impacts result from a  combination of changes in herbicides costs,
shifts in production from flax  to alternative crops, and yield losses
resulting from delayed  seeding.  The average loss in returns per acre of
flax treated with diallate substitutes would be $13.59.  This loss
represents 25% of the expected  returns to land, labor, and management with
diallate available.

          Since only 3% of the  total flax acreage is estimated to be
treated with diallate,  the cancellation of diallate would not have a sig-
nificant effect on total  flaxseed supplies or established marketing systems
or patterns.  Moreover, some shift in production away from flax is already
occurring because of decreasing demand for flaxseed and flax straw.  (USDA,
Crop Production Annual  Summary, 1978.)

     C.   Lentils

          Conmercial lentil  production is located almost entirely in
Washington and Idaho.   During the 1976-1978 period, an average of 114,000
acres of lentils were planted annually in these two states.  Both states
                                    III-7

-------
are subject to wild oat infestations,  and diallate is a major herbicide
used to control wild oats  in  lentil  plantings.   It is estimated that 43,320
acres or 38% of the total  lentil acreage  were treated with diallate.  With
a normal use rate of one pound  (active ingredient basis)  of diallate
applied per acre, this would  require 43,320  pounds of diallate.

          Use of triallate provides  an excellent substitute for diallate
use on lentils.  Triallate, although increasing the overall costs of wild
oat control, offers increased yields which more than offset the increase in
costs.  The positive economic impact is based on increased yields resulting
from triallate being less  phytotoxic and  in  some cases better control of
the pest.  Triallate would pick up an  estimated 93% of the diallate usage
if diallate were cancelled.   Therefore, cancellation of diallate and a
shift to triallate would have a positive  impact to growers by increasing
returns by $512,000 annually, with an  increase  in production of 1.6%.

          The growers reluctance to  switch to triallate may be 1) unfamil-
iarity with triallate, 2)  inconvenience of stocking additional chemicals
when diallate can treat all crops, and 3)  marketing preferences.

          Other herbicides which are expected to offset the impact of
diallate's cancellation on lentils include propham and barban.  However,
these two herbicides are not  as useful as triallate because of altered
efficacy characteristics (e.g. critical timing  for effectiveness and rapid
biodegradation in soil).   Therefore, these herbicides are not considered as
complete alternatives for  diallate.

     D.   Peas

          Dry peas (including Austrian winter peas and wrinkled seed peas)
are produced primarily in  Washington and  Idaho.  During the 1976-1978
period, an average of 231,000 acres  of dry peas were produced in these two
states.  Wild oats are a major pest  in dry pea  production.  Approximately
11% of the total acreage is treated  with  diallate for wild oat control,
using 30,375 pounds (active ingredient basis) of diallate.

          Triallate is considered the  primary substitute for diallate.
Triallate is already used  on  45% of  the dry  peas produced and offers
growers a larger return on the land  through  increased yields (see
discussion in III.C.), so  that the total  return to the grower increases  by
nearly $730,000 annually.

          Propham and barban  are also  registered for use in dry peas.
However, these two chemicals  have limitations which preclude their total
effectiveness as alternatives.  Upon diallate's cancellation,  the usage  of
both propham and barban would increase above the current 2% (propham) and
6% (barban) usage on dry peas.

          The cancellation of diallate is anticipated to result in U.S.  dry
pea production increases of less than  0.01%.  Therefore no economic impacts
would occur at either the  market or  consumer levels.
                                   III-8

-------
     5.   Minor Uses

          Diallate, in addition to the  foregoing major uses,  is also
registered for use on soybeans, corn, barley,  potatoes, safflower,  and
alfalfa.  The percentage of use on each of  the above crops is quite small;
thus, use of diallate has been determined to be minor on these crops.   Use
of diallate on these crops is basically limited to North Dakota, Minnesota,
Montana, and Idaho.

          The estimated use of diallate on  potatoes is 0.5% of the total
potato acreage.  For all other minor crops  except safflower and alfalfa,
the total treated acreage is thought to be  0.1% or less.  Data are not
available to show the total amount of diallate used on safflower and
alfalfa acreage.

          If diallate becomes  unavailable,  barban would be used to control
wild oats on soybeans, and eptam would be the chemical of choice in corn
and potatoes.  Triallate  is  registered for use on barley.  Monsanto has
expressed a desire  to register triallate for all crops for which diallate
is now registered  (Spurrier,  1979).

          The  cancellation of diallate on the crops discussed  in this
section is not expected to have any economic impact upon  the grower or
market prices.   Only in the case of potatoes does the  use approach 0.5% of
the  total acreage.   In other cases, diallate treated acreage amounts to
0.1%  or less  of  the total acreage.
                                     III-9

-------
IV.  Development of Regulatory Options

     A.   Introduction

          The risks and benefits  associated with the uses of diallate have
been identified in Sections  II and  III.   As explained in Section I, FIFRA
mandates that the Agency  achieve  a  balance between the competing considera-
tions of risks and benefits.  In  order to carry out that mandate, the
Agency has developed various regulatory  options and has evaluated each
option for its impact on  both sides of the risk/benefit equation.

          This section of Position  Document 2/3 briefly summarizes the
salient risks and benefits associated with the uses of diallate and
describes the process by  which the  Agency developed potential courses of
action.

     B.   Salient Risk/Benefit Considerations

          In performing  a risk/benefit analysis of the uses of diallate,
the Agency  identified several salient factors, on both sides of the risk/
benefit equation, which  became  determining considerations in the
development of  regulatory options.   These considerations are reviewed below.

           1.   Salient Risk Considerations

                As detailed in Section II of this document the original  risk
criteria  cited in the RPAR notice as the basis for the Agency's presumption
against diallate stands unrebutted.  The principal risk  associated with the
use of diallate is oncogenicity.  This risk manifests  itself in  the general
population through dietary exposures at very  low levels  and to pesticide
applicators through dermal and inhalation  exposures  when applying diallate
as an emulsifiable concentrate.

                It is estimated that there  are approximately 24004pesticide
applicators currently at  risk.   This risk  is  estimated to be 10    and is
 of primary concern to the Agency (Table  1^-10).  The dietary risk to  the
 general population is estimated  to be 10    based on  tolerance  levels
 adjusted to reflect the percent  of crop  treated  (Table 11-11).   The  Agency
 considers the dietary risks  of diallate  to be low  and not of primary
 concern when compared to  the benefits associated with its use.

                Since the  original  RPAR  notice was  published the  Agency
 received additional evidence to  support the conclusion that diallate is a
 mutagen.   Although quantitative estimates of the  mutagenic risk to
 applicators are not possible at  this time,  any risk reduction  procedures
 proposed to reduce the oncogenic risks  of diallate will concomitantly
 reduce mutagenic risks.

                There is  also evidence that diallate causes  neurotoxic
 effects.  As in the case of mutagenic risks quantitative estimates of risk
                                      IV-1

-------
are not presently possible.   However,  based on current exposure estimates
there is a 600-fold  span between  the  observed effect level and the exposure
level.

          2.   Salient  Benefit  Considerations

               The benefits  of  diallate  were assessed in terms of the
economic impacts which  would result if its uses were cancelled and users
were thereby forced  to  employ available  alternatives.  As detailed in
Section III, the economic  impacts associated with the cancellation of
diallate total just  over $3  million (Table III-1).

               Sutjar Beets
               The  total  annual market value of sugar beets is $500
million.  Should diallate become unavailable, growers are expected to
experience an  annual  loss of  $4 million.   More than 60% of the diallate
used in this country  is applied to  sugar  beets.  Presently the emulsifiable
concentrate formulation is applied  to 85% of the treated acreage while the
granular formulation  is applied to  the remaining 15%.  The degree control
of wild oats provided by  each of these formulations appears to be the same.

               Several alternative  chemicals were identified in Section III
of this document.   Specifically mentioned were, 1)  cycloate for control in
irrigated areas, and  2) eptam for control in non-irrigated areas, and 3)
barban which can be used  in both areas.  However, due to certain limita-
tions (see Section  III.A) none of these chemicals provide adequate
protection against  wild oats.

               In changing to granular diallate alone for control of wild
oats in sugar  beets,  the  increased  cost to growers in the short run will be
approximately  $6-$7 per acre-treatment given current prices.

               Flax

               Approximately 3% of  the total flax acreage is treated with
diallate.  If  all forms of diallate should become unavailable for use in
the control of wild oats  in flax, growers are expected to experience a
$400,000 annual loss.  The emulsifiable concentrate is the only formulation
presently registered  for  this use.

               Eptam  and  barban are the most commonly used alternatives for
diallate.  Both of  these  chemicals  have limitations which reduce their
desirability and effectiveness in the control of wild oats (see Section
III.B).  Granular dialiate is anticipated to provide effective control of
wild oats in flax;  however, this formulation is not currently registered
for this use.
                                     IV-2

-------
               Lentils

               Thirty-eight percent  of  the  total lentil acreage is treated
with diallate.  Lentil production  is basically limited to two western
states, Idaho and Washington.   It  is estimated that more than 43,000 pounds
of diallate are applied to lentil  acreage annually.  The emulsifiable
concentrate is the only diallate formulation presently registered for this
use.

               The major  alternate chemical used to control wild oats in
lentils is triallate.  Triallate has proven to be an effective alternate
and provides control of wild  oats  equal to  or greater than that provided by
diallate.  Triallate is also  less  phytotoxic to lentils than diallate.
Propham and barban are alternative chemicals but do not provide acceptable
control of wild oats.

               Peas

               Dry peas,  like lentils,  are  primarily grown in Idaho and
Washington.  Currently, approximately  11%  of the dry pea acreage is treated
with diallate.  Only the  emulsifiable  concentrate formulation of diallate
is registered for use on  peas for  the  control of wild oats.

               Triallate  is  the priory alternate chemical for diallate on
peas.  Presently, triallate  is used  on 45%  of the dry pea acreage.
Triallate  is  less phytotoxic to peas than diallate.

               Minor Uses (Alfalfa,  Barley, Corn, Potatoes, Safflower,
               and Soybeans)

               The total  percent of  minor crop acreage treated annually
with diallate ranges from <0.1% to 0.5%.  More specifically, it is
estimated  that 0.5% of  the potato  acreage is treated, whereas the
percentage of treated acreage for  all  other crops is 0.1% or less.  No
economic impacts are expected if diallate is cancelled.  Only the
emulsifiable  concentrate  formulation is registered for the minor uses.

               Barban and eptam are  considered as possible substitute
chemicals.  Triallate is  now registered for use on barley.

     C.    Risk/Benefit  Analysis

           1.   Dietary  Risk/Benefit  Analysis

               As indicated  in Section II.D.1. the dietary risk from
diallate is estimated  as  10   .  This estimate is based on assuming
residues exist on treated crops at the tolerance level.  This is a worst
case assumption and to  date,  residues  have not been found on any crops at
the level  of  detection  (.02 ppm).   More than 60% of the dietary risk  is
attributable  to the use  of diallate  on sugar beets.  The dietary risk is
                                     IV-3

-------
considered to be  low and  the  benefits  of diallate use are moderate for
sugar beets and flax, and low for  lentils,  peas,  and minor crops.  In sugar
beets the average cost  increase  per  acre will range from $4.69 to $87.65.
For flax the loss  in returns  per acre  would be about $14.  For all other
uses economic benefits  would  accrue  due to  minor  production increases from
either more effective control of wild  oats  or decreased phytotoxic effects
of alternate chemicals.   Therefore,  the Agency concludes that the benefits
(low to moderate)  outweigh the dietary risk (low).  In view of this,
regulatory action on the  basis of  dietary risk alone is not warranted.  As
indicated above the principal risk of  diallate is to pesticide applicators
through dermal and inhalation exposure when applying diallate as an
emulsifiable concentrate.

          2.   Applicator Risk/Benefit Analysis

               The applicator risk/benefit  matrix for diallate, expressed
in qualitative terms, is  shown in  Table IV-1.  For all presently registered
use patterns the  applicator risks  are  high  for the emulsifiable concentrate
formulations while the  risks  associated with the granular formulation are
no greater than the general populations risk from dietary exposure.  The
benefits of diallate use  (either formulation) on sugar beets and flax are
moderate and all  other  uses are  low.

               Applicators of the  emulsifiable concentrate formulation may
be exposed both dermally  and via inhalation as the result of  splashing,
vaporization, or  accidental spills.   Likewise during application, exposure
may occur both dermally and via  inhalation.  Table 11-10 identifies the
dermal  risk to applicators of the  emulsifiable concentrate formulation as
10   .

               One potential risk  reduction measure is to require applica-
tors using emulsifiable concentrate  diallate to wear protective clothing.
Protective clothing  in  this instance is defined as rubber gloves and
coveralls.  These items would reduce the exposure level by a factor_pf 4
(see Section II.C.1.).   This reduction would result in a risk of 10
which the Agency  still  considers unreasonable when compared to the low and
moderate benefits and,  therefore,  unacceptable.  Therefore, a protective
clothing requirement will not be considered as a viable  risk reduction
measure in this  analysis.

               Thus  the risk/benefit picture for the emulsifiable concen-
trate formulation is  essentially the same for all uses.  The applicator
risk is high and  the  benefits are, at best, moderate.  Since the risks out-
weigh the benefits in  every case,  the Agency must consider regulatory
options for reducing the risks associated with the emulsifiable  concentrate
formulation, in  particular, the  high applicator risks.

                From  this analysis  the Agency concludes that the high  risk
involved in the  use  of  the emulsifiable concentrate is unacceptable when
considered  against the  low to moderate benefits of all emulsifiable
                                      IV-4

-------
Table IV-1.  Applicator Risks versus Economic  Benefits  of Diallate
Uses
Sugar Beets
Flax
Lentils
Dry ?eas
Minor Uses
Barley,
Potatoes ,
Saf flower,
Soybeans ,
Corn, Alfalfa
Applicator Risk
Smulsifiable Granular Economic Benefits (c)
Concentrate
High (a) Low Moderate
High (a) {b) Moderate
High (a) (b) Low
High (a) {b) Low
High (a) (b) Low
 (a) High  Risk _>  10~4
 (b) Although not  registered for these uses the applicator risks would remain
    low if  registered.
 (c) Benefits analysis  did not evaluate the individual benefits of the two
    formulations.
                                        IV-5

-------
concentrate uses.  It also  concludes  that  this  risk  can be reduced to the
acceptably low dietary level  if  the emulsifiable  concentrate formulation is
replaced by the granular  formulation.   Therefore,  the Agency will examine
the feasibility of cancelling all  diallate emulsifiable concentrate
formulations in its regulatory options.

               For sugar  beets (the only use  for  which the granular formula-
tion is also registered), the risk/benefit balance for the granular formula-
tion is shifted favorably.  Risks  become low  because of significantly
decreased applicator exposure and  benefits remain moderate because the cost
and effectiveness of granular diallate  are approximately the same as the
emulsifiable concentrate.   Based on this finding  for sugar beets, the
Agency assumes that this  more favorable risk/benefit balance could be
achieved if the granular  formulation  were  registered for all of the other
uses.

     D.   Regulatory Options

          With regard to  the  emulsifiable  concentrate products of diallate,
three basic regulatory options have been developed for consideration:

          1.   Continue emulsifiable  concentrate  registrations.

          2.   Cancel emulsifiable concentrate  registrations immediately.

          3.   Cancel emulsifiable concentrate  registrations effective in
two years.

               Options 1  and  2 represent an absolute regulatory response.
For Option 1 it means that  sale  and distribution  of  diallate emulsifiable
concentrate products are  unconditionally continued.   Option 2 on the other
hand means that sale and  distribution of these  products are prohibited
effective as soon as the  decision  becomes  final.   Option 3 represents a
decision to phase out the emulsifiable  concentrate formulations to permit
time to extend the registrations of granular  diallate to uses where it is
not presently registered.

          1.   Option 1;  Continue Emilsifiable Concentrate Registration

               This option  would return emulsifiable concentrate formula-
tions of diallate to the  registration process,  and they would be retained
as effective means to control wild oats in sugar  beets and other crops.  By
adopting Option 1, the Agency would conclude  that the benefits associated
with the use of emulsifiable  concentrate diallate outweigh the risks and
that allowing its use would not  result  in  'unreasonable adverse effects.
                                                                       -4
               Under this option,  the potential applicator risks of 10
resulting from inhalation and dermal  exposure would  not be reduced.  There
would be no adverse economic  impacts  associated with Option 1 because use
of emulsifiable concentrate formulations would  continue.  By choosing this
                                     IV-6

-------
option the Agency would  conclude  that  the benefits would outweigh the risks
of continued use.

          2.   Option  2:   Cancel  Smulsifiable Concentrate Registrations
               Immediately

               This  option would  eliminate all uses of emulsifiable
concentrate diallate thirty days  after the final Agency decision.  3y
adopting Option  2 the  Agency would conclude that the risks associated with
the use of emulsifiable  concentrate diallate exceed the benefits, and
result in unreasonable adverse effects.

               Under this  option, applicator risks resulting from
inhalation and dermal  exposure would be reduced to the same magnitude of
risk as the general  dietary risk.  Cancelling all emulsifiable concentrate
registrations iaanediately  would imply that risks outweigh benefits.

               Hie  economic impact which result from the immediate
cancellation of  diallate emulsifiable concentrate formulations on sugar
beets is estimated  to  be more than $3 million (Table III-1).  The benefit
analysis did not attempt to identify the impact of the individual formula-
tions.  Bowever,  most  of the impact will result from the cancellation of
the emulsifiable concentrate formulations.  This is based on the fact that
the emulsifiable concentrate accounts for approximately 85% of the diallate
presently used on sugar  beets.  Flax growers are expected to suffer a
$400,000 annual  loss while growers of lentils and peas are anticipated to
receive an economic gain as expressed in Table III-1.

               This option would ignore such factors as availability of
alternatives including granular diallate since time would have to be
allowed to produce  adequate quantitites of this material.  It also ignores
the time necessary  to  register granular diallate on flax and other crops
for which the  emulsifiable concentrate is now registered if the markets
demand such action.  Lastly, this option neglects to allow time for growers
to modify or acquire the necessary equipment to apply the granular
formulation.

          3.   Option  3;  Cancel Bmulsifiable Concentrate Registrations
               Effective Two Years After the Decision Becomes Final

               This option is essentially the same as Option 2.  It differs
in that  it would eliminate all uses of emulsifiable concentrate diallate
two years after  the decision becomes final.  By adopting Option 3, the
Agency would indicate  its  unwillingness to accept the applicator risks
associated  with the use of the emulsifiable concentrate formulation
indefinitely.  This option would lessen the impact of immediate cancel-
lation by  1) allowing time to produce the necessary granular diallate to
control  wild oats in sugar beets,  2) allowing time to register granular
diallate on  crops where only the emulsifiable concentrate is now registered
and where  other  registered alternatives are not desirable, and 3)  allowing
time  for  growers to make necessary equipment adjustments.
                                     IV-7

-------
               Some granular  diallate  formulations are presently registered
for use on sugar beets only in Minnesota and North Dakota.   It will be
necessary to allow sufficient time  to  extend the registration of these
granular formulations to  the  western states.   While the registered granular
formulation with diallate as  the  sole  active ingredient is  only registered
for use in the non-irrigated  areas, other granular formulations with
diallate as one of the active ingredients are registered in other geo-
graphical areas (western  states).   Two years should also provide ample time
to register the granular  formulation on crops now being treated with the
emulsifiable concentrate  formulation if the market demands  it.

               The Agency, based  on the information available, believes 1)
that a two-year tinif frame is sufficient *-im«» for the company to produce
the granular diallate required to maintain control of wild  oats in sugar
beets, and 2) that two years  is ample  *-inu» for growers to make adjustments
in equipment necessary to apply the granular formulation (Lewis, 1980).

               Because there  is no  acceptable alternative chemical
presently registered for  use  on flax,  cancellation of the emulsifiable
concentrate is a major concern.   The expected annual economic impact to
flax growers is $400,000. The Agency  believes that the granular
formulation will provide  excellent  wild oat control with little or no
phytotoxicity and is willing  to consider a proposal for this use.

               Registrants of the emulsifiable concentrate  and granular
formulations may submit an application for amended registration to
1) convert their emulsifiable concentrate formulation to granular, and
2) expand their granular  registrations to crops for which only the
emulsifiable concentrate  is now registered.  The review of  these actions
will be expedited and should  not  require additional data.
                                  IV-8

-------
V.   Proposed Regulatory Decision

     The Agency proposes to  implement  Option 3  which is to cancel the
emulsifiable concentrate formulations  of  diallate effective two years from
the date the decision becomes  final.

     Option 1 is unacceptable  because  the level of risk Jo applicators of
the enmlsifiable concentrate formulations remains at 10   which the
Agency is unwilling  to permit  considering the low and moderate benefits.

     Option 2 is also unacceptable.  While under this Option the risk to
applicators is reduced to  the  general  dietary risk level, it does not take
into account the potential impacts that may occur due to the unavailability
of adequate supplies of granular diallate currently registered for sugar
beet use.  In addition it  does not permit any opportunity for the
registrant to extend the use of granular diallate to those uses where only
the enmlsifiable concentrate is currently registered prior to complete
cancellation.

     Option 3 would  overcome the disadvantages of both Options 1 and 2.
The Agency would welcome applications  to extend the use of granular
diallate as a replacement  for  the emulsifiable concentrate and EPA
considers a two-year time  frame appropriate to accomplish this.  At the
same time, the two-year time frame permits the production of adequate
supplies of granular diallate  to meet  expected market demand.

     Also under Option 3 registrants would be allowed to convert
emulsifiable concentrate formulations  to granular and expand granular
registrations to crops for which only  the emulsifiable concentrate is
presently registered.
                                    V-l

-------
VI.  References

Albert, R.E., 1979a.  Diallate RPAR Rebuttal Response.  Carcinogen
      Assessment Group, EPA, 1979.

Albert, R.E., 1979b.  Risk Assessment of Avadex.  Carcinogen Assessment
     Group, EPA, 1979.

Brusick, D.J. 1977a.  Mutagenicity Evaluation of CP15336, BIO-77-166 (Final
     Report July, 1977) submitted to Monsanto, Co, Inc., by Litton
     Bionetics, Inc., Project No. 2683. 9pp.

Brusick, D.J. 1977b.  Mutagenicity evaluation of CP15336 in the mouse
     lymphoma v-ssay, BIO-77-171  (Final Report, August,  1977) submitted to
     Monsanto Co, Inc., by Litton Bionetics, Inc., Project No. 2684.
     Submitted to EPA March 22,  1978.  EPA Accession No. 233353.

Carcinogenesis Bioassay Data Systems, 1975.  Preliminary Analysis Report
     for Avadex.  Prepared for the National Cancer Institute by EG&G/Mason
     Research Institute.

Development Planning and Research Associates, Inc., March, 1979.  Economic
     Analysis of Effects of Restricting Use of Diallate on Sugarbeets,
     Flax, Lentils and Dry Peas; to Environmental Protection Agency.

Gardner, R., 1980.  Estimates of human exposure to diallate.  Memo to James
     Wilson, SPRD (TS-791), Project Manager.

Innes, J.R.M., B.M. Ulland, M.G. Valeno, L. Petrucrlli, L. Fishbein,
     E.R. Hart, A.J. Pollotta, R.R. Bates, H.L. Falk, J.J. Gart, M. Klein,
     I. Mitchell and J. Peters.  1969.  Bioassay of Pesticides and
     Industrial Chemicals for Tumorigenicity in Mice:   A Preliminary Note.
     J. Nat. Cancer Inst. 42:1101.

Jensen, J. 1980.  Applicator Exposure to Diallate While Using the Granular
     Formulation.  Memo to J. E. Wilson, Jr., Project Manager for Diallate,
     April 23, 1980.

Keplinger, M.L.  1974.  Mutagenic Study with Avadex Technical in Albino
     Mice.  BTL-74-13.  Industrial Bio-Test Laboratories, Inc., August 20,
     1974.  IBT No. 622-05252.   Submitted to EPA by Monsanto Company on
     October 14, 1976.  14 pp.

Keplinger, M.L.  1976a.  Two-year Chronic Oral Toxicity  Study with Avadex
     Technical in Albino Rats BTL-74-12.  Industrial Bio-Test
     Laboratories, Inc., December  17, 1976.  IBT No. 622-05250.  Submitted
     to EPA by Monsanto Company  on February 4,  1977.  EPA Accession No.
     228094.

Keplinger, M.L.  1976b.  Neurotoxicity Study with Avadex in Chickens,
     BTL-76-84.  Industrial Bio-Test Laboratories, Inc., December 28,
     1976. %IBT No. 8580-09119.  Submitted to EPA by Monsanto Company on
     February  16, 1977.  EPA Accession No. 226094.
                                   VI-1

-------
Lewis, C., 1980.  Additional information on use of granular diallate.
    Memo to James Wilson, SPRD Project Manager.

Mauer, I., 1978.  Ph.D., Geneticist, Toxicology Branch, EPA, November 28,
      1978.  Diallate:  Preliminary Assessment of Mutagenicity, Regulatory
     Options, and Alternatives, Memorandum to Roger Gardner, Toxicology,
     EPA.

Monsanto Agricultural Products Co.,  1977.  Rebuttal to  the Rebuttable
     Presumption Against Registration and Continued Registration of
     Pesticide Products Containing Diallate.  I1A  [30000/15].

Pesticide Residues Committee,  1965.  NAS-NRC Report on  "No Residue"  and
      "Zero Tolerance,1* June  1965.

Phillips, B.M.  1977.  Final  Report to Monsanto Company  42-day Neurotoxicity
      Study with Diallate in  Chickens.   (f1B(30000/15) attachment A)

Rinkus,  S.J. and M.S. Legator,  1977.  Mutagenicity of Diallate.  Submitted
      to  EPA  (FRS/TSD).  July 6,  1977. 5  pp.

Sandhu,  S.S., Ph.D.,  Biochemistry Branch, EPA, August 30,  1978.  Evaluation
      of  the Data for  the Mutagenicity of Diallate  and Ti Lallate.
      Memorandum to William  F.  Durham, Ph.D., Director,  Environmental
      Toxicology Division  (HERL/RTP), EPA.

Selim, S. October  16,  1978.  Exposure Analysis  for Diallate.  Proprietary
      Environmental Fate Branch,  Hazard  Evaluation  Division, OPP, EPA.

SikJca, J.C.  and P. Florczyk. 1978.   Mutagenic Activity  of  Thiocarbamate
      Pesticides in a  Microbial Test  System.  J.  Agr. Food  Chem.  26:146-.

Simmon,  V.F., E.S. Riccio,  and A. Griffin.  1978.   In vitro Microbiological
      Mutagenicity Assays  of Diallate and Triallate (Final  Report,  April,
      1978).   Stanford Research Institute International, Project  LSU-3493.
      Prepared for EPA (HERL/RTP)  under  Contract  No.  68-01-2458.  23 pp.

Sochard, M.R.   1980.   Editing  of Diallate PD 2/3  Pursuant  to  Memo  From J.
      Jensen,  EFB,  to  J. E.  Wilson,  Jr.,  Project  Manager for Diallate,
      May 14,  1980.

Spurrier E. ,  1979.  Memo  of telephone  conversation with R. Troast.

Olland,  B.,  E.K. Weisburger, and J.H.  Weisburger.  1973.  Chronic Toxicity
      and Carcinogenicity  of Industrial  Chemicals and Pesticides.
      Toxicol. Appl. Pharmacol. 25:446.

USDA/State  Assessment Team on  Diallate,  1977.   Assessment  of  the Need for
      Diallate  in  Agriculture #10A[30000/15].

U.S.  Department of  Health Education and Welfare.  1969.   Report  of  the
      Secretary's  Commission on Pesticides and Their Relationship to
      Environmental  Health.  U.S.  Government Printing Office, Washington,
      D.C.

                               VI-2

-------
         UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
                      WASHINGTON, D.C. 20460
                                         OFFICE 3F =E3~!CIO£S ANO TOXiC 3U3STANCSS
            ENVIRONMENTAL PROTECTION AGENCY

                  (OPP-30000/153)

                     DIALLATS

      PRELIMINARY NOTICE OF DETERMINATION CONCLUDING
      THE REBUTTA3LE PRESUMPTION AGAINST REGISTRATION
           OF PESTICIDE PRODUCTS AND NOTICE OF
            AVAILABILITY OF POSITION DOCUMENT

AGENCY:  Environmental Protection Agency (SPA)

ACTION:  Notice of Availability

SUMMARY:  The Environmental Protection Agency  issued

a notice of rebuttable presumption  against registration

and continued registration of  pesticide products  containing

diallate on May 31,  1977.  This  notice announces  the Agency's

oreliminary regulatory position  pursuant to  40  CFR  152.11(a)

(5) and makes the document available  to registrants, users,

and other interested parties.

DATES:  Written comments  are  due on or before  (insert

date  30 days  after  date  of publication  in  the  FEDERAL

REGISTER).

ADDRESS:  Document  Control Office

          Chemical  Information Division

          SPA (TS-793)

          Room  E447

           401 M Street,  S.W.

          Washington, DC  20460

-------
FOR FURTHER INFORMATION CONTACT:




          James Wilson, Project Manager



          Special Pesticide Review Division



          401 M Street, S.W.



          Washington, B.C.  20460  (TS-791)



SUPPLEMENTARY INFORMATION:  The Agency's Position Document



(?D 2/3) reviews the risks  and benefits of the pesticide



products containing diallate  and discusses various regulatory



options available.  The preliminary  findings will be reviewed



by the FIFRA Scientific Advisory Panel and the Secretary of



Agriculture as required under the  amended FIFRA.



                  I.  INTRODUCTION



    The Environmental Protection Agency issued a notice of



rebuttable presumption against registration and continued



registration ("SPAR") of  pesticide products containing



diallate published  in the FEDERAL  REGISTER of May 31, 1977



(42 FR 27669), a thiocarbamate herbicide, thereby initiating



the Agency's public review  of the  risks and benefits of



diallate.  The rebuttable presumption was issued on the



basis of oncogeniciry.  The Agency also requested registrants



and other interested parties  to submit data on the effects



that diallate may have on plant and  animal ceil division.



    This notice constitutes the Agency's Notice of Deter-



mination (Notice) pursuant  to 40 CFR 162.11(a)(5).  This



determination is preliminary  at this point pending external



review through submission to,  and  review by, the United



States Department of Asriculture and the Scientific Advisorv

-------
Panel, pursuant to sections 6(b) and 25(d) of the Federal
Fungicide, Insecticide, and Rodenticide Act  (FIFRA) as
amended.  The action does not become final until the Agency
has reviewed these comments and  issued a  final notice.
    In broad summary,  the Agency has determined  that diallate
continues to exceed the risk  criteria outlined in 40 CFR
162.11 for cncogenicity.  The risks  that  diallate poses to
applicators of  the emulsifiable  concentrate  formulation are
of sufficient concern  to  require the Agency  to consider
whether these risks can be  reduced.  The  Agency  has  considered
benefits  information  including  that  submitted by registrants,
interested persons, and the United States Department of
Agriculture and has analysed  the economic,  social,  and
environmental benefits of the uses of  diallate.  The Agency
has weighed risks  and  benefits  together,  in-order  to determine
whether the risks  of  each diallate use are warranted by  the
benefits  of the use.   In  weighing  risks  and benefits,  the
Aaencv  considered  what risk reductions could be  achieved  and
how risk  reduction measures would  affect the benefits  of  the
use.
    The Agency  has determined that the risks of  certain  uses
of diallate  are greater than  the social,  economic,  and
environmental  benefits of these uses,  unless risk  reductions
 are  accomplished  by modifications in the terms and conditions
of registration.   Accordingly,   the Agency is proposing to
 initiate  action to cancel or deny registrations for all of
 the  uses  of diallate which involve the emulsifiable concentrate
 formulation.  The Agency has further determined that these

-------
cancellations will accomplish significant risk reductions,



and that these risk reductions can be achieved without



significant impacts on the benefits of the continued use of



the granular formulation of diallate.  The remainder of this



Notice and the accompanying Position Document set forth in



detail the Agency's analysis of comments submitted during



the rebuttal phase of the diallate RPAR, the Agency's



reasons and factual bases for the regulatory actions it is



initiating.



                 II.  LEGAL BACKGROUND



     In order to obtain a registration for a pesticide under



?I?RA, a manufacturer must demonstrate that the pesticide



satisfies the statutory standard for registration.  Section



3(c)(5) of FIFRA requires, among other things, that the



pesticide perform its intended function without causing



"unreasonable adverse effects on the environment".  The term



"unreasonable adverse effects on the environment" is defined



in FIFRA section 2(bb) as "any unreasonable risk to man or



the environment, taking into  account the economic, social,



and environmental costs and benefits of the use of any



oesticide".  In effect, this  standard requires a finding



that the benefits of each use of the pesticide exceed the



risks of use, when  the pesticide is  used in accordance with



commonly recognized practices.  The  burden of proving that  a



pesticide satisfies the registration standard is on the



proponents of registration and continues as long as the



recistra'cion remains in effect.  Under section 6 of FIFRA,

-------
the Administrator is required to cancel the registration of



a pesticide or modify the  terms and conditions of registration



whenever he determines that  the pesticide no longer satifies



the statutory standard for registration.



    The Agency generally  announces  that an RPAR has arisen



by publishing a notice in  the FEDERAL  REGISTER  After  an



RPAR is issued, registrants  and other  interested persons



are invited to review the  data  upon which  the presumption  is



based and to submit  data  and information  to  rebut  the



presumption.  Respondents  may rebut the presumption of risk



by showing that the  Agency's initial determiation  of risk



was in  error, or  by  showing  that  use of  the  pesticide  is  not



likely  to result  in  any  significant exposure to  humans or  to



the animal or plant  of  concern  with regard to  the  adverse



effect  in cues-ion.   See  40  CFR 162.11(a)(4).   Further, in



addition  to submitting  evidence to rebut  the risk  presumption,



respondents may submit  evidence^ as to  whether  the  economic,



social,  and environmental benefits of  the use  of the  pesticide



subject to  the  presumption outweigh the risks  of use.



    The regulations  require the Agency to conclude an  RPAR



bv issuing  a Notice  of  Determination  in which the  Agency



 states  and  explains  its position on the question of whether



 the risk ^resumptions have been rebutted.  If the  Agency



 determines  that the presumption is not rebutted, it will



 then  consider  information relating to the social,  economic,



 and  environmental costs and benefits which registrants and



 other interested  oersons submitted to the Agency,  and  any



 other benefits information known to the Agency.

-------
    After weighing the risks  and the benefits of  a pesticide
use, the Administrator may conclude the RPAR process by
issuing a notice of intent to  cancel or deny registration,
pursuant to FIFRA section 6(b)(1)  and  section 3(c)(6) or
by issuing a notice of intent  to hold  a hearing pursuant  to
section 6(b)(2) of FIFRA to determine  whether the registra-
tions should be cancelled or  applications  for registration
denied.
    In determining whether the use of  a pesticide poses risk
which are greater than benefits, the Agency considers
modifications to the terms and conditions  of registration
which can reduce risks, and the impacts of such modifications
on the benefits of the use.   Among the risk reduction
measures short of cancellation which are available to the
Agency are changes in the directions for use on the pesticide's
labeling and classification of the pesticide for  "restricted
use" pursuant to FIFRA section 3(d).
    FIFRA requires the Agency  to submit notices issued
pursuant to section 6 to the  Secretary of  Agriculture
for comment and to provide the Secretary of Agriculture
with an analysis of the impact of  the  proposed action on
the agricultural economy under section 6(b) of FIFRA.  The
Agency is required to submit  these documents to the Secretary
at least 60 days before making it  public.  If the Secretary
of Agriculture comments in writing within  30 days after
receiving the notice, the Agency is required to publish
the Secretary's comments and  the Administrator's  response

-------
with the notice.  FIFRA also requires the Administrator to
submit section 6 notices to a Scientific Advisory Panel for
comment on the impact of the proposed action on health and
the environment, at the same time  and under the same proce-
dures as those described above  for review by the Secretary
of Agriculture under section 25(d)  of FIFRA.
    Although not required  to do so under the statute, the
Agency has decided that it is consistent with  the general
theme of the RPAR process  and the  Agency's overall  policy of
open decision making to afford  registrants and other interest-
ed persons an opportunity  to comment on  the bases for the
proposed action during  the time that the proposed action  in
under review by the Secretary of Agriculture and  the Scien-
tific Advisory Panel.   Accordingly, appropriate steps: will
be taken to make copies of the  Position  Document  available
to registrants  and other  interested persons  at -the  time  the
decision documents are  transmitted for  formal  external
review, through publication of  a notice  of  availability  in
the Federal Register, or  by other  means.   Registrants  and
other interested person will  be allowed  the  same  30 day  com-
ment period that the  statute provides  for  receipt of  comments
from the Secretary of Agriculture  and  the  Scientific Advisory
Panel.
    After  completing  these external review procedures and
making  any changes  in the proposed action  which are deemed
appropriate  as  a result of the comments received, the Agency
will proceed  to implement the desired  regulatory  action  by
preparing  appropriate documents and releasing  them  in the
manner  prescribed  by  the  statute and by the Agency's rules.

-------
  III.  DETERMINATION AND INITIATION OF REGULATORY ACTION
      The Agency has considered information on the risks
associated with the uses of diallate including information
submitted by registrants and other interested persons in
response to the diallate RPAR.  The Agency has also considered
information on the social, economic, an environmental
benefits of the uses of diallate subject to the RPAR,
including benefits information submitted by registrants and
other interested person in conjunction with their rebuttal
submissions, and information submitted by the United States
Department of Agriculture.
    The Agency's assessment of the risks and benefits of the
uses of diallate subject to this RPAR, its conclusions and
determinations whether any uses of diallate pose unreasonable
adverse effects on the environment,  and its determinations
whether modifications in terms of conditions of registration
reduce risks sufficiently to eliminate any unreasonable
adverse effects are set forth in detail in the Position
Document.  This Position Document is hereby adopted by the
Agency as its statement of reasons for the determinations
and actions announced in this Notice and as its analysis of
the impacts of  the proposed regulatory actions on  the
agricultural economy.  For the reasons summarized  below  and
developed in detail in the Position  Document,  the  Determina-
tions of the Agency with respect to  diallate  are  as  follows:

-------
               A.   Determination  of  Risk
        The diallate  RPAR was  based  on  information  indicating
that diallate  posed oncogenic  risks  to  humans.  As developed
fully  in  the Position Document (PD 2/3),  the Agency has
determined that  the information submitted to rebut  the risk
criteria  for oncogenicity was  insufficient to overcome the
presumption against diallate for  this effect.
        Since  the  original RPAR notice  was published the
Agency received  additional evidence  to  support the  conclusion
that diallate  is a mutagen.  Although quantitative  estimates
of  the mutagenic risk to applicators are  not possible at
this timef any risk reduction  procedures  proposed to reduce
the oncogenic  risks of diallate will concomitantly  reduce
mutagenic risks.
        There  is also evidence that  diallate causes neuro-
'toxic  effects. As  in  the case of mutagenic risks quantitative
estimates of risk  are not presently  possible.  However,
based  on  current exposure estimates  there is a 600-fold  span
between the observed  effect level in chickens and  the
estimated human exposure level.
        The principal risk associated with the use  of
diallate  is oncogenicity. This risk  manifests itself in  the
general population through dietary exposures at low levels
 and to pesticide applicators through dermal  and inhalation
 exposures before and/or during application of diallate  as an
 emulsifiable  concentrate at high levels.

-------
        It is estimated that there are approximately 2400



pesticide applicators currently at risk.  This risk is


                  —4
estimated to be 10   and is of primary concern to the



Agency.  The dietary risk to the general population is



estimated to be 10~  based on tolerance levels adjusted to



reflect the percent of crop treated.  Since residues have



not been detected on diallate treated foods, these risk



estimates are almost certainly overstated.  The Agency



considers the dietary risks of diallate to be low and not of



primary concern when compared to the benefits associated



with its use.



            B.  Determinations of Benefits



        Diallate is used primarily to control wild oats in



sugar beets, flax, lentils and peas.  It has minor use with



alfalfa, barley, corn, potatoes, safflower and soybeans.



    1.  Sugar Beets.  The total annual market value of sugar



beets is $500 million.  Should diallate become unavailable,



growers are expected to experience an annual loss of $4



million.  More than 60 percent of the diallate used in this



country is applied to sugar beets.  Presently the emulsifiable



concentrate formulation is applied to approximately 85



percent of the treated acreage while the granular formulation



is applied to the remaining 15 percent.  The degree of



control of wild oats provided by each of these formulations



appears to be the same for fall applications but slightly



less for the granular formulation in spring applications.

-------
    2.  Flax .  Approximately 3 percent of the total flax
acreage is treated with diallate.  If diallate should become
unavailable for use in the control of wild oats in flax,
growers are expected to experience a $400,000 annual loss.
The emulsifiable concentrate is the only formulation presently
registered for this use.  However, the granular is expected
to be as effective.
    3.  Lentils.  Thirty-eight percent of the total lentil
acreage is treated with diallate.  Lentil production is
basically limited to two western  states, Idaho and Washington.
It is estimated that more than 43,000 pounds of diallate  are
applied to lentil acreage annually.  The emulsifiable
concentrate is the only diallate  formulation presently
registered for this use.
    4*  Peas.  Dry peas, like  lentils, are primarily grown
in Idaho and Washington. Currently,  approximately  11 percent
of the dry pea acreage  is treated with diallate.   Only  the
emulsifiable concentrate formulation of diallate  is registered
for use on peas for the  control of wild oats.
    5.  Minor uses including  alfalfa, barley,  corn, potatoes,
safflower,  and soybeans.  The  total  percent  of minor  crop
acreage treated annually with  diallate ranges  from <0.1
percent to  0.5 percent.  More  specifically,  it  is  estimated
that  0.5 percent  of  the potato acreage  is  treated, whereas
the percentage of treated  acreage for  all  other  crops  is  0.1
percent or  less.   No  economic impacts  are  expected if
diallate  is cancelled.   Only the emulsifiable  concentrate
formulation is  registered  for the minor  uses.

-------
    C.  Determinations of Unreasonable Adverse Effects
        For the reasons set forth in the accompany-
ing Position Document, the Agency has made the following
determinations about the unreasonable adverse effects
associated with the continued use of the eraulsifiable
concentrate formulations of diallate.
        The Agency has determined that the risks arising
from the use of emulsifiable concentrate formulations of
diallate to control wild oats are greater than its social,
economic, and environmental benefits.
        The Agency has further determined that modifications
in the terms and conditions of registration of the emulifi-
able concentrate formulation of diallate will not accomplish
significant risk reductions.  Accordingly, the Agency has
determined that, unless these formulations are cancelled,
the uses of these formulations will  continue to  cause
unreasonable adverse effects in the  environment, when used
in accordance with widespread and commonly recognized prac-
tices.  The Agency has also determined that benefits derived
from uses of granular formulations of diallate are greater
than its social, economic, and environmental risks.  Therefore,
these formulations will not be cancelled.
            D.  Initiation of Regulatory Action
        Based upon the determinations summarized above  and
set out in detail in  the Position Document, the  Agency  is
initiating a regulatory action which would cancel  the use of
all diallate emulsifiable  concentrate formulations  two  years
after this decision becomes  final.   Diallate registrants may

-------
apply for amended registration to convert from emulsifiable
to granular formulations as well as expanded registrations
of the granular to crops for which only the emulsifiable
concentrate is now registered.  Review of the amended
registrations will be expedited and may not require more
data.
              V.  PROCEDURAL MATTERS
    This Notice of Determination notifies the United States
of Agriculture, the Scientific Advisory Panel, pesticide
registrants and users,  and other interested parties of the
Agency's preliminary determinations relating to the risks
and benefits to the uses of diallate  and provides these
entities and individuals with the opportunity to comment on
these determinations.
    The Agency's decision to initiate regulatory action
must be referred for review by^the Secretary of Agriculture
and the Scientific Advisory Panel.  The EPA position document
setting forth the reasons and factual bases for the regulatory
actions which the Agency proposes and this Notice of Deter-
mination are being transmitted immediately to the Secretary
of Agriculture and the  Scientific Advisory Panel for comments.
The Agency also will offer registrants and other interested
persons an opportunity  to comment on  the bases for the
Agency's action by making copies of the Position Document
available upon request.

-------
    Interested persons may receive copies of the documents

by communicating their requests to James Wilson, Project

Manager, Special Pesticide Review Division, Office of

Pesticide Programs, EPA (TS-791), 401 M Street SW, Washington,

DC  20460 (703) 557-7420.  Registrants and other interested

persons have the same period of 30 day comment period

that the statute provides for the Secretary of Agriculture

and the Scientific Advisory Panel.

    All comments on the proposed actions should be sent to

the Document Control Office, Chemical Information Division,

EPA (TS-793), Room E-447, 401 M Street S.W., Washington, DC

20460.  In order to facilitate the work of the Agency  and of

others inspecting the comments, registrants and other  interested

persons should submit three copies of their comments.  The

comments should bear the the identifying notation 30000/15B and
                                 f V
should be submitted on or before  e>£
    After completion of these review procedures,  the Agency

will consider the comments received and publish  an  analysis

of them, together with any changes in  the regulatory actions

announced in this Notice which it determines  are appropriate.

Until this final review phase is concluded  in this  manner,

it is not necessary for registrants or other  interested

persons to request a hearing to contest any regulatory

action resulting from the conclusion of this  RPAR.

Dated:
                             SXe-ven D. Jellijnek
                             Assistant Administrator
                             for Pesticides
and Toxic Substances

-------
                DIALLATE
          Position Docunent 2/3
               April,  1980

Office of Pesticides and Toxic Substances
     Environmental Protection Agency
          401  'M' Street, S.W.
         Washington, D.C. 20460

-------
ACKNOWLEDGEMENTS
EPA Proj eet Support Team

James E. Wilson, Jr.,  Senior PM, SPRD
Minnie R. Sochard, Ph.D.,  PM,  SPRD
Karl 0. Bayer, OGC
Tom Edwards, HED
Roger Gardiner,  HED
Janice Jensen, HED
Dave Johnson,  HED
Charles Lewis,  BFSD
Richard Stevens, HED
Linda V. Zygadlo, BFSD

-------
                             TABLE OF CONTENTS

                                                            Page Number
I.   Background                                              1-1
     A.    Introduction                                       1-1
          1.   Organization of Position Document             1-1
          2.   Chemical and Physical Characteristics         1-1
          3.   Registered Uses                               1-1
          4.   Tolerances                                    1-2
     B.    Applicable Sections of FIFRA       .                1-2
     C.    The "•RPAR" Process                                 1-3
     D.    Regulatory History                                 1-3
     E.    Basis for the Rebuttable Presumption               1-4
II.  Analysis and Assessment of Risk                        II-1
     A.    Analysis of Rebuttal Arguments
            for Oncogenicity              -                  II-1
          1.   Rebuttal Pertaining to National
                 Cancer Institute (NCI) Rat
                 Study (Litton-Bionetics)                   II-1
          2.   Rebuttal Pertaining to Industrial
                 Bio-Test (IBT) Rat Study
                 (Sponsored by Monsanto Company)            II-5
          3.   Rebuttal Pertaining to NCI
                 Mouse Study (Innes Study)                  11-11
     B.    Analysis of Data Submitted Since PD 1
            for Other Possible Adverse Effects              11-14
          1.   Mutagenicity                                 11-14
          2.   Neurotoxicity                                11-16
          3.   Reproductive Effects                         11-17
     C.    Exposure Analysis                                 11-18
          1.   Spray Applicator Exposure                    11-18
               a.   Spray Applicators                       11-18
               b.   Granular Applicators                    11-19
          2.   Dietary Exposure                             11-19
     D.    Risk Assessment                                   11-19
          1.   Oncogenic Effects                            11-19
          2.   Mutagenic Effects                            11-24
     E.    Risks Associated with Alternative Chemicals       11-27
III. Benefit Analysis                                      III-1
     A.    Sugar Beets                                      III-2
     B.    Flax                                             III-7
     C.    Lentils                                          III-7
     D.    Peas                                             IH-8
     E.    Minor Uses                                       III-9
IV.  Development of Regulatory Options                      IV-1
     A.    Introduction                                      IV-1
     B.    Salient Risk/Benefit Considerations               IV-1
          1.   Salient Risk Considerations                  IV-1
          2.   Salient Benefit Considerations               IV-2

-------
     C.   Risk/Benefit Analysis Options                       IV-3
          1.   Dietary Risk/Benefit Analysis                  IV-3
          2.   Applicator Risk/Benefit Analysis               IV-4

     0.   Regulatory Options                                  IV-6
          1.   Option 1                                       IV-6
          2.   Option 2                                       IV-7
          3.   Option 3-                                      IV-7
 V.  Proposed Regulatory Decision                             V-1
VI.  References

-------
                       DLALLATE:  POSITION DOCUMENT 2/3

I«   Background

     A.    Introduction

           1 •    Organization of Position Document

                This  Position Document contains five parts.  Part I is the
introductory  section.   Part II is an evaluation of the risk of diallate.
It includes descriptions  of the relevant data on toxicity, exposure, and
the Agency's  present risk assessment.  Part III is a summary of the
economic benefits  of diallate.  Part IV describes the range of the
regulatory options considered by the Agency.  Part V puts forward the
Agency's recommended option.

           2.    Chemical and Physical Characteristics

                Diallate (S-(2,3-Dichloroallyl)diisopropylthiocarbamate) is
a thiocarbamate which is  also known by the trade name AVADEX®.  Diallate
acts as a  pre-emergence selective herbicide.  It is an amber-colored liquid
which is formulated  as an emulsifiable concentrate (4 pounds/ gallon ) and as
a granular (10%).—   Its  molecular weight is 270.2.  Its structural
formula is as follows:

                     trans isoner                   ds  isoaer
           H          Cl          CH3     Cl        Cl           CH
           Gl                                        .
                                                     *     ^CH— CH
                                                                  -
                                                                  3
           Foraula
           Formula Weight
          3.   Registered Uses

               Monsanto Agricultural  Products Company is the sole producer
of technical-grade diallate.  As  a  registered pesticide, diallate is  cur-
rently used to control wild oats  in sugar beets,  flax,  barley,  corn (grain
and silage), forage legumes (alfalfa,  sweet,  red  and alsike clover),
—   As a granular, diallate  is  registered for use on sugar beets  only.
                                    i-r

-------
lentils, peas, potatoes, safflower, and soybeans.   In combination with
pebulate or cycloate, it is used also  to  control other grasses and broad-
leaf weeds in sugar beets.  Diallate is incorporated into the soil in the
fall before the freeze or in the spring,  either before or after seeding,
but before emergence.  Usually only one application is made by the user per
season.

               Eight federal registrations  of  products containing diallate
are held by three registrants.

          4.   Tolerances

               Tolerances for diallate in or on raw agricultural
commodities are listed in 40 CFR 180.277  as follows:  negligible residues
on alfalfa (fresh and hay), barley (grain,  forage,  and straw), clover
(fresh and hay), field corn grain, fodder"and  forage, flaxseed, lentils,
peas, pea forage and hay, potatoes, safflower  seed, soybeans, soybean
forage and hay, and sugar beet roots and  tops  at 0.05 part per million.

     B.   Applicable Sections of FIFRA

          The Federal Insecticide, Fungicide and Rodenticide Act (7 O.S.C.
136 et seq.) , as amended, confers authority on EPA to regulate pesticide
products.  Section 3 (a)  of the Act requires all pesticide products to be
registered by the Administrator before they may be sold or distributed.
Before the Administrator may register  a pesticide,  however, he must
determine that its use will not result in "unreasonable adverse effects on
the environment," defined in Section 2(bb)  of  FIFRA to mean "unreasonable
risk to man or the environment, taking into account the economic, social,
and environmental costs and benefits of the use of  any pesticide."  In
other words, any registration decision must .take into account both risk and
benefits from the pesticide's uses.

          Section 6(b) of FIFRA authorizes  the Administrator to issue a
notice of intent to cancel the registration of a pesticide or to change its
classification if it appears to him that  che pesticide or its labeling
"does not comply with the provisions of  [FIFRA]  or, when used in accordance
with widespread and commonly recognized practice,  generally causes
unreasonable adverse effects on the environment."   Thus, the Administrator
may cancel the registration of a pesticide  whenever he determines that it
no longer satisfies the statutory standard  for registration; this standard
requires, among other things, that the pesticide "perform its intended
function without unreasonable adverse  effects  on the environment"  [FIFRA
3(c)(5)(C)].  He may also cancel the registration of a pesticide if its
labeling does not comply with the misbranding  provision of FIFRA, which
requires the labeling to contain certain  language "adequate to protect
health and the environment" (FIFRA 2(q)).
                                    1-2

-------
     C.   The "RPAR" Process

          To implement its authorized functions,  the Agency has designed
the Rebuttable Presumption Against Registration (RPAR)  process, which
involves gathering data on the  risks  and benefits associated with the uses
of suspect pesticides. By allowing all interested parties to participate by
submitting information, the process enables EPA to make balanced decisions
concerning problem pesticides.

          If the presumptions  of risk are not rebutted, the evidence
pertaining to benefits must be  evaluated and considered together with the
evidence pertaining  to risk.   Various risk reduction methods and their
costs are analyzed.  The Agency then determines whether the pesticide may
be regulated so that a balance is achieved between risks and benefits.  If
the statutory balance cannot  be reached for a use, the registrations for
that use must be cancelled.

     0.   Regulatory History

          The first  registration for a product containing diallate was
granted to Monsanto  on December 9, 1960 (EPA Registration Number 424-119).
Use of diallate on corn,  forage legumes, lentils, peas, potatoes,
safflower, and sugar beets was approved.  Because no residues were detected
in or  on  these raw agricultural commodities the use of diallate was
accepted  under the "no residue-zero tolerance" concept.

          In  1954, the Food,  Drug, and Cosmetic Act was amended to provide
for the establishment of  tolerances or exemptions from tolerances for
pesticide chemicals  in or on food crops.  The  amendments also  provided  for
the establishment of a tolerance "at zero level"  if data showed that no
detectable residues  were  present in the treated crop at the time of
harvest.  In  1965  the National Research Council of the National Academy of
Sciences  recommended that the "no residue-zero tolerance" concept be
abandoned.  This  recommendation was based on the  fact that  this concept, as
applied  in the registration and regulation of  pesticides, had  become
scientifically and administratively untenable.  Among the reasons given by
the council were  that  analytical methodology had  improved,  and that  small
levels of pesticide  residues had become detectable  (Pesticide  Residues
Committee Report  on  "No  Residue" and  "Zero Tolerance", NAS-NRC, June  1965).

          In  1966 the  USDA began to phase-out  this  concept  (FR April  13,
 1966).  Registrants  of products under  this concept  were  given  (through  a
 series of 1 year  extensions)  until 1971 to convert  the "no  residue"
tolerances  to finite tolerances.

          The Council also recommended that pesticides previously regulated
 under the "no residue"  concept be  regulated on the  basis of "negligible
residue"  tolerances.  These tolerances could be  established by supplying  a
 limited  amount of data.   It was concluded  that a  negligible residue  is  the
 amount which will produce no effects  in  test animals,  which (effects) are
                                     1-3

-------
indistinguishable from control animals.  The tolerance,  in many  cases,  may
reflect the sensitivity of the method and require only two 90-day
(subacute)  animal studies. Such was the case when the tolerance  for
negligible residues was established for diallate.

          During the phase-out period, Monsanto filed a  petition (PP
7F0607) requesting a tolerance of 0.3 ppm on corn,  forage legumes, lentils,
peas, potatoes, safflower, and sugar beets.  That petition was subsequently
withdrawn and a second petition was filed (PP 9F832) requesting  a tolerance
for negligible residues of 0.05 ppm on the above crops.  Tolerances for
negligible residues of diallate were established in or on the above raw
agiicultural commodities on July 28, 1971 (40 CFR  180.277).

          On May 31, 1977, the Agency issued in the Federal  Register   (42
FR 27669) a notice of rebuttable presumption against registration and
continued registration of pesticide products containing  diallate.
Diallate:  Position Document 1, published together  with  the  RPAR notice,
explained the background and supporting data for the presumption of risk
cited in the RPAR notice.

          Following the publication of the RPAR notice,  Monsanto Company
requested and was given a 60-day extension of the  rebuttal period.  The
extension was granted to all registrants and interested  parties.

     E.   Basis for the Rebuttable Presumption

          The rebuttable presumption against registration and continued
registration of pesticide products containing diallate was issued on the
basis of oncogenic effects in test animals as a risk criterion  [40 CFR
162.11 (a)(3)].  Specifically, the presumption was  based on  the  following
three long-term feeding studies which indicated that diallate is poten-
tially oncogenic:  1) a National Cancer Institute  (NCI)  Ulland et al.,
1973) rat study (Litton Bionetics), 2) an Industrial Bio-Test (IBT)
(Keplinger. 1976a) rat study (sponsored by Monsanto Company), and 3) an NCI
mouse study (Innes et al., 1969).

          Data from the rat study conducted by Ulland and coworkers at
Litton Sionetics were verified by the NCI and reviewed by the EPA
Carcinogen Assessment Group (GAG).  The GAG concluded that the Litton
Bionetics study showed a statistically significant  increase  in malignant
tumors at the highest dose in male rats and in carcinomas at the higher
dose in female rats.

          Industrial Bio-Test concluded from its rat study "that the
neoplastic lesions noted in the test and controls were considered normal
for rats of this age and strain."  In its evaluation, the CAG concluded
that "the Industrial Bio-Test study in rats showed  a statistically
significant excess of mammary carcinomas in•females."
                               1-4

-------
          The CAG also concluded  that the mouse study conducted by Innes
and coworkers showed an  increased incidence of liver-cell and pulmonary
tumors.  This study was  the  basis for the Mrak Commission Report recommen-
dation that human exposure to  diallate be minimized because there was
evidence of tumor induction  in mice.

          Respondents were given  an opportunity to rebut the presumption
against diallate by showing  (1)  that the Agency's initial determination of
risk was in error, or  (2) that given current use patterns, exposure to
diallate is not likely to result  in any significant chronic adverse effects
[40 CFR 162.11(a)(4)].   Respondents were also invited to submit evidence on
behalf of the social, economic,  and environmental benefits of the use of
the pesticide [40 CFR  162.11(a)(5)(iii)].
                                     1-5

-------
II.  Analysis and Assessment  of  Risk

     A.   Analysis of Rebuttal Arguments  - for Oncogenicity

          The Agency has analyzed the rebuttal-comments submitted to it in
response to to the presumption of oncogenicity—  and responded to these
comments in this section.  From  the analysis of rebuttal comments, the
Agency has concluded that  the oncogenic presumption against diallate has
not been rebutted and that humans are subj ect to the risk of developing
cancer from the use of diallate.

          1.   Rebuttal Pertaining to National Cancer Institute (NCI) Rat
               Study (Litton  Bionetics)

               a.   Errors in Tabulated Data

                    Monsanto  Company nateu errors in the data cited in the
EPA's Carcinogen Assessment Group (CAG) (Albert, 1979) report and in
rebuttal presented a retabulation of the  raw data (Monsanto Co.,
#1A[30000/15]).

                    The Agency acknowledges Monsanto's rebuttal on this
point.  The CAG has re-evaluated the raw  data and has corrected its report
accordingly.  However, these  corrections  do not change the Agency's
interpretations of the study.

                    The CAG's revised tabulation of the data (shown in
Tables II-1 and II-2) differs somewhat from the data presented by Monsanto.
Most of these differences  are due to the  classification of gliomas and
leukemias as sarcomas by the  CAG and as carcinomas by Monsanto.  These
differences in classification are unimportant since the Agency bases its
regulatory decisions on oncogenic risks which include all turners  (Albert,
1979a).

               b.   Statistical  Difference in the Malignant Tumors in Males

                    Monsanto  Company contended that according to their
analysis- of the data, there is no statistical difference in the number of
malignant tumors in the diallate-treated male rats (both dose groups
combined) compared to pooled  controls  (13/52 treated vs. 11/64 controls,p =
.21).  Monsanto asserted further that there is no statistical difference in
the number of malignant tumors or of sarcomas in either treated group of
male rats (high or low dose)  as  compared ta control groups (Monsanto Co.,
#1A[30000/15]).

                    The respondent acknowledged an increased incidence of
carcinomas in  the high-dose male group as compared to the controls, but
submitted that the incidence  in  the low-dose male group was not different
from either  control group  (Monsanto Co.,  #1A[330000/15]).
2/    In  addition to the above-mentioned rebuttals, the Agency received
      16  responses pertaining to the benefits of diallate.  Comments
      submitted on benefits are addressed in Section III, "Benefit Analysis."
                                  II-l

-------
           Table II-1.  NCI Rat Study, Incidence of Malignant Tumors  in
                        Male Rats Ingesting Avadex  (Diallate)—
            Comparison of Low- and High-Dose Groups to Pooled Controls

                No. of Rats with    No. of Rats, with   Total No. of  Rats
Dose Group        Carcinomas          Sarcomas"        with Malignant Tumors£/—/
Pooled Control
Low Dose
High Dose
4/64 (6%) .
3/26 (12%)-
4/26 (15%)
7/64 (11%)
1/26 (4%)
4/26 (15%)
11/64 (17%)
4/26 <15%lf/ /
10/26 (38%)-:i
    a/
    -'  Revised CAG tabulation.   (Albert,  1979a)
    — , Gliomas and leukemias were counted as  sarcomas.
    c/
    —  Rats with carcinomas did  not have  sarcomas.
    —/ Corrected for survival.
    —' Two of these rats with carcinomas  had  metastases  (carcinomas  of  the
       prostate metastatic to lung and lymph  nodes;  islet cell  carcinomas of
    _. the pancreas with metastases to the  heart).
    —  The tumor incidence in the high dose group  compared to the pooled
       control group is statistically significant  (p =  .032)  (Fisher Exact
       Test).
    -*  The total incidence of 10/26 includes  2  unclassified malignant
       tumors in the subcutaneous tissue.
                                    II-2

-------
           Table II-2.  NCI Rat Study,  Incidence of Malignant Tumors  in
                        Female Rats Ingesting Avadex  (Diallate)—
          Comparison of Low- and High-Dose Groups  to  Pooled Controls

                No. of Rats with    No. of Rats .with    Total  No. of Rats
Dose Group     .   Carcinomas          Sarcomas—         with Malignant  Tumors
Pooled Control
Low Dose
High Dose
— Revised CAG
— , Gliomas and
3/64 (Si).,
2/26 (8%)-/
5/26 (19%)-/
tabulation . (
leukemia s were
4/64 (6%)
2/26 (8%)
0/26 (0%)
Albert, 1979a).
counted as sarcomas .
7/64
4/26
5/26

(11%)
(15%)
(19%)

    —  The two rats with carcinomas of  the  mammary gland had metastases to
     , the lungs.
    —  The tumor incidence in the high  dose group compared to the pooled
       control group is statistically significant (p = .042)  (Fisher Exact Test)
                                     II-3

-------
                    The  Agency rej ects  Monsanto's rebuttal on this point.
The  GAG analyzed  tumor incidence  from each treatment group individually,
whereas Monsanto  Company combined the data from the high- and low-dose
groups  and in  combining  the data  the significance found in the high dose
group was  masked.   The GAG re-evaluation of the data demonstrated a statis-
tically significant increase in total malignant tumors in male rats of the
high-dose  group as  compared to pooled controls (10/26 treated vs.11/64
controls;  p =  .032)   (Refer to Table II-1) {Albert, 1979a).

               c.   Statistical Significance of Carcinomas, Sarcomas,
                    and  Total Malignant Tumors in Females

                    From their analysis of the data on female rats,
Monsanto Company  concluded that there is no statistically significant
increase in carcinomas,  sarcomas, or total malignant tumors in either
treated group  (Monsanto  Co.,  #1A[30000/15]) .

                    The  Agency rej ects  this rebuttal attempt.  From the
GAG's revised  tabulation of the data, it is apparent that there is a
•statistically  significant increase of carcinomas in the female rats of the
high-dose  group as  compared to pooled controls (5/26 treated vs. 3/64
controls;  p =  .042).  (Albert, 1979a.)  (Refer to Table  II-2.)

               d«   Low  Number of Tumors in High-Dose Male Rats

                    Monsanto Company asserted that the number of tumor-
bearing rats in each  diallate-treated group is lower than in either control
group,  and emphasized that the high-dose group has the lowest number of
animals with tumors (Monsanto Co., #1A[30000/15]).

                    The  Agency rej ects  this point of rebuttal.  The number
tabulated  by the  GAG  for the controls was 11/64 (17%) as compared with 10/
26  (38%) and 4/26 (15%)  in the high-and low^-dose male groups, respectively.
The  individual numbers of animals possessing tumors is not relevant.  The
appropriate comparison is the percentage of animals which have tumors.  The
percentage of  animals with tumors in the high-dose v,roup is statistically
greater than the  percentage of animals  with tumors in the control group.
(Albert, 1979a.)

               e.   No Apparent Effect  of Piallate on the Formation of
                    Individual Tumor Types

                    Monsanto contended  that an evaluation of individual
tumor types/sites is  necessary to conclude that a compound is carcinogenic,
and  that there was  no apparent effect of diallate on the formation of
individual tumor  types  (Monsanto  Co., #1A[30000/15]) .

                    The  Agency rej ects  this rebuttal attempt.  Although
there was  no statistically significant  incidence of individual tumor types,
there was  a statistically significant increase of total malignant tumors in
                                H-4

-------
male rats of the high-dose  group  relative to pooled controls.   Moreover,
there was a statistically significant increase  of  carcinomas  in female rats
of the high-dose group compared to pooled controls.  The use  of total tumor
data in evaluating carcinogenicity is discussed in a recent Interagency
Regulatory Liaison Group  (IRLG) document as follows:

                    "At the present time there  is  considerable
                    uncertainty about the interpretation of carcinogenic
                    responses  in  terms of the total tumor yield in contrast
                    to the  response in terms of a statistically significant
                    increase of tumors in specific target organs or
                    tissues. Traditionally, carcinogens have  been
                    recognized in human and animal studies by a decisive
                    increase in tumors of target organs.  However, it is
                    conceivable that a generalized increase in total tumor
                    yield,  in  the absence of an excess incidence in one or
                    more  target tissues, could  occur, for example by a
                    promoting  effect that generally increases the
                    spontaneous incidence of tumors in test animals or by
                    the action of a multipotent carcinogen whose response
                    did not reach statistical significance in any one organ
                    even  at the maximum tolerated dose.— "

          2.   Rebuttal Pertaining to Industrial Bio-Test (IBT) Rat
               Study  (Sponsored by Monsanto Co.)

               The existing registrations and tolerances for  diallate were
established and supported by data contained in  studies conducted by
Industrial 3io-Test Laboratories  (IBT) for Monsanto.

               In 1977, subsequent to the publication of Position Document
1, the Agency, in the Office of Pesticide Programs, established a
Toxicology Data Audit Program  (TDAP) to assure  the reliability and
integrity of data supplied  to  the Agency by pesticide manufacturers.  These
data are the integral component of the information upon which pesticides
are registered and tolerances  are established in the United States.

               Industrial Bio-Test Laboratories (IBT) was one of the
initial laboratories audited jointly with the Food and Drug Administration.
The IBT laboratory performed a large volume of  testing utilized by the
Agency in its regulatory  decision-making.  During  an audit at the facility,
numerous significant departures from acceptable laboratory practice were
noted.  As a result, the  Agency decided to reevaluate all pivotal IBT
studies used in support of  tolerances.  The workload of this  evaluation
program was shared with the Canadian Government in cases where chemicals
were registered on identical data bases.  The IBT studies for diallate were
evaluated by Canada and the results are shown in Table II-2a.
3/ IRLG  (February 6,  1979),  Scientific Bases for Identifying Potential
   Carcinogens and Estimating Their Risks.
                                    II-5

-------
Table II-2a.  Diallate IBT Studies
Type of Study
                              IBT No.
               Status
                                                         Remarks
Fish & Wildlife
8-Oay LC   Mallard Duck

8-Day Dietary LC5Q Bob-
  white Quail

8-Day Dietary LC5Q Bob-
  white Quail
8-Day Dietary LC^Q Bob-
  white Quail

4-Day Fish Tox

Fish & Wildlife

Fish S Wildlife

Fish & Wildlife

Acute Studies

4-week Pilot oral study

Pilot feeding study/mice

Acute Inhalation

Acute cholinesterase
651-3026

J-6672


J-6673


651-3025
Valid

Valid


Valid


Valid
665-3027
A-6675
A-6674
A-6849
3531-9714
8532-9820-
59-13-3
8530-9030
Valid
Invalid No raw data
Invalid No raw data
Invalid No raw data
Valid
Invalid Incomplete raw data
Invalid No raw data
Valid
                                    II-6

-------
Table II-2a. (continued).  Diallate 1ST  Studies
Type of Study
Subacute Studies
Subchronic oral
Subchronic oral
Subchronic oral
Subchronic oral
Chronic Studir.s
Teratogenicity/rabbits
Mutageni city/mi ce
Neurotoxicity/ chicken
Neurotoxicity/ chicken
Chronic feeding/rats
IBT No.
59-13A
59-13-2
59-13
59-13-1
651-5254
622-5252
8580-9119
8580-10813
622-5250
Status
Invalid
Invalid
Invalid
Invalid
Invalid
Invalid
Valid
Invalid
Invalid
Remarks
No raw data
No raw data
No raw data
No raw data
Deficiencies in
exper imental
design
No raw data

No raw data
Incomplete raw
                                                            data
                                    IX-7

-------
               TDAP Evaluation of 1ST Rat Study

               The IBT study sumitted by Monsanto was  evaluated  by  the
Office of Pesticide Program's Toxicology Data Audit  Program  (TDAP).   The
analysis of the study indicates that the 2-year feeding  study  is invalid
because of the lack of histopathology data, food consumption data,  organ
weight data on surviving animals, and dietary preparation  data to show
unequivocally that doses were correctly prepared.  Therefore,  no scientific
conclusions can be drawn from this study.

               a.   Errors in Tabulated Data

                    Monsanto observed errors in the  data cited in the CAG
report and in their rebuttal submitted a retabulation  of the raw data.
Monsanto contended there are no statistically significant  differences in
the number of animals with tumors in the diallate-treated  groups as
compared to the controls (Monsanto Co.,- #1A[30000/15]).

                    The CAG has re-evaluated the raw data  and  has corrected
its report accordingly.  The revised CAG tabulation  (shown in  Tables II-3
and II-4) now agrees with the retabulation submitted by  Monsanto Company.
The data indicate that there is no statistically significant increase
either of total tumors (benign and malignant) or of  tumors of  any
anatomical site in any diallate-treated group of male rats as  compared to
controls.

                    As Table II-5 indicates, however,  female rats treated
with  100 ppm diallate  (middle dose) showed a statistically significant
increase of benign mammary tumors (p =  .021) and hence of total  mammary
gland tumors (Albert,  1979).  Therefore, the last portion of this rebuttal
is rej ected.

               b.   No Statistical Increase  in  Mammary Tumors

                    The incidence of mammary tumors  in diallate-treated
female rats is shown in Table II-5.  Monsanto Company pointed out that
there was no statistically significant  increase in  the number of mammary
tumors in the treated  female rats of the high-  or  low-dose groups as
compared to the controls.  The respondent  argued  that the admittedly
statistically significant increase in mammary  tumor among the middle-dose
groups represents a random event, since there was  no dose response.
Monsanto pointed out further that there is  no  statistically significant
increase of mammary carcinomas.   (Monsanto  Co., #1A[30000/15])

                    As Monsanto  noted,  the  female  rats of the middle  dose
group exhibited a statistically  significant  increase of total mammary
tumors  (p = .021), which was attributable  to  the  statistically  significant
increase in benign mammary tumors  (p =  .021). This  increase of benign
mammary tumors in the  middle-dose  group may indicate  an adverse  effect,
although the possibility that  this response may be a  random event cannot be
                                 11-8

-------
    Table II-3.   IBT Rat Study, Incidence of Benign and Maligngnt
                      Tumors in Male Rats Ingesting Diallate	
Dose Group
No. of Rats with
Benign Tumors
No. of Rats with  ,  .
Malignant Tumors	
Total No. of Rats
with Tumors
Control
50
100
200
1/50 (2%)
0/50 (0%)
2/49 (4%)
1/50 (2%)
4/50 (8%)
7/50 ( 14% )
4/49 (8%)
6/50 (12%)
5/50 (10%)
7/50 (14%)
6/49 (12%)
7/50 (14%)
       Revised GAG tabulation  (Albert,  1979a).
    —  Rats with endocrine tumors  are not  included.
    —  Rats with both benign and malignant tumors  were  tabulated as  having
       malignant tumors.
                                     II-9

-------
         Table 11-4.
      1ST Sat Study, Incidence of Carcinomas  and., .
      Sarcomas in Male Rats Ingesting Diallate^ — —
Dose Group
No. of Rats with
Carcinomas
No. of Rats with
Sarcomas"
   Total No. of Rats
with Malignant Tumor
 Control
   50
  100
  200
   0/50 (0%)
   3/50 (6%)
   1/49 (2%)
   2/50 (4%)
     4/50 (8%)
     4/50 CS.%)  ,
     3/49 (6%)~
     4/50 (8%)
       4/50 (8%)
       7/50 (14%)
       4/49 (8%)
       6/50 (12%)
    ^' Revised CAG tabulation (Albert, 1979a).
    —  Table II-4 is an elaboration of Total Malignant  Tumors column in
      , Table II-3.
    —  Rats with ondocrine tumors are not included.
    —  Rats with both benign and malignant tumors  were  tabulated as having
       malignant tumors.
    S/
       One rat had metastatic fibrosarcoma to the  lung and liver, another
       to the lung.  No primary sarcomas were found in the two rats.
                                11-10

-------
unequivocally refuted (Albert,  1979a).   However,  the  high percentage  of
test animals with tumors in the  high  dose  group  strongly suggests  that
there is an adverse effect.

               c.   Spontaneous  Pituitary  Adenomas

                    Monsanto Company  pointed  out that there  is -a high
spontaneous- incidence of pituitary  adenomas in the  I3T rat study.   The
respondent pointed out further that there  is  no  apparent significant
increase in the incidence of this tumor  in any treated group, nor  any
linear increase by dose observed (Monsanto Co.,  *1A[30000/15] ).

                    The Agency agrees with this  conclusion  (Albert, 1979) .

          3.   Rebuttal Pertaining  to the  NCI Mouse Study   (Innes  Study)

               a.   Study Invalid

                    Monsanto Company  claimed  that the Innes  study  is
invalid for the following two reasons:

                    1)   Only one dose level, the maximum  tolerated dose
(MTD), was used in the study.   The  rebuttal contended that  the  MTD has  been
redefined, thereby making the doses used in the  study higher than  what  is
now considered the MTD and  invalidating the experiment.

                    2)   The experimental design necessitated the  dosing
of litter-mates.  The rebuttal  contended that biological and statistical
significance cannot be drawn from this inferior  experimental design.
Monsanto suggested that a bias  of inherited tendencies (e.g., predisposi-
tion to hepatoma formation) cannot  be eliminated because the litter mates
were not randomly distrihnted among the test  groups (Monsanto Co.,
*1A[30000/15).

                         The Agency rej ects  this rebuttal  argument.
Concerning Monsanto's point 1)  above, the Agency points out  that the MTD  is
defined in  the NCI Guidelines as "... the highest dose of the test  agent
given during the chronic study  that can be predicted not to  alter  the
animal's normal  longevity from  effects other  than carcinogenicity."—

                         Innes  et al. reported  that the MTD was
selected on the  basis of a  series of  studies  in  which the  maximal  level
given in a5single dose,  in  6  daily  and in 19  daily doses,  resulted in zero
mortality.—
—  NCI Guidelines  for Carcinogen Bioassay in Small Rodents, NCI Tech.
    Report  Ser.  No.  1, Feb., 1976, p. 14. U.S. Dept. of Health, Education
    and Welfare, PHS,  NIH, NCI-CG-TR-1.

—^  Innes,  J.R.M. et al (1969)  Bioassay of Pesticides and Industrial
    Chemicals  for Tumorigenicity in Mice:  A Preliminary Note.  J. Nat.
    Cancer  Inst. Vol.  42,  p. 1102.
                                  H-11

-------
    Table II-5.
IBT Rat Study, Incidence of Benign and Malignant Tumors of
the Mammary Gland in Female Rats Ingesting  Diallate—
Dose Group'(pom)
  No. of Rats with
               D/
  Benign Tumors—
No. of Rats with
Malignant Tumors
Total No. of Rats
with Tumors
Control
50
100
200
14/50 (28%)
19/49 (39%)
24/48 (50%)
15/46 (33%)
5/50 (10%)
4/49 (8%)
5/48 (10%)
10/46 (22%)
19/50 (38%)
23/49 (47%)
29/48 (60%)-
25/46 (54%)
    ~' Revised CAG tabulation (Albert, 1979a) .
    —  Rats with both benign and malignant tumors were  counted as
       malignant.
    —  The tumor incidence in the 100 ppm dose  group  compared to the
       control group is statistically significant (p  =  .021)  (Fisher Exact Test),
                                     11-12

-------
                         A comparison of the  survival data in the carcino-
genicity study indicates that  the  number of 18-month survivors in the
diallate-treated groups was  similar to that of the vehicle (0.5% gelatin)
and untreated control groups for  each respective species/sex.  Hence the
doses given in the study were  tolerated by the treated animals and
therefore did not exceed the MTD.

                       .  In  reference to Monsanto's point 2)  above,  Innes
administered diallate to mice  (in  0.5% gelatin)  by stomach tube from 7  days
to 4 weeks of age.  Thereafter, the mice were weaned and were given
diallate (without a vehicle) in the diet.   Since the study began prior  to
weaning, each test group (e.g., diallate-treated, positive or negative
control group) was comprised of litter-mates  instead of a random assortment
of litter-mates, which was precluded by the study design.

                         During the MRAK Gommission review of this study,
which recommended that human exposure to diallate be minimized,—  Mr.
Carrol Weil presented a dissenting opinion which included criticism  of  the
non-random allocation of animals.   Monsanto cited  Mr. Weil's criticism in
its rebuttal.

                         In  response to Mr. Weil's criticism  on non-
randomization, the MRAK Commission reported that "...the data were
reexamined on a litter basis,  in  keeping with the Epstein-Mantel approach,
rather than on the single-animals  basis employed in the Journal of the
National Cancer Institute report.   All compounds which had been judged
positive for tumor ^nduction (significant at  the 0.01 level,  or stronger),
remained positive."—   Thus, although non-randomization of litter-mates
is not the optimal experiment  design, there is no evidence in this study
that a bias existed for a genetic  predisposition to tumor occurrence
(Albert, 1979a).

               b.   Sex Specificity—•Increased Hepatomas in Male Mice

                    Monsanto Company acknowledged the statistically
significant increase in hepatomas  in male mice in the Innes study, but
contended that the apparent  sex-specificity is unusual (Monsanto Co.,
*  1A[30000/15]).

                    The Agency rejects this rebuttal point.  The data from
the Innes study indicate a statistically significant increase in hepatomas
in both strains of male mice when  compared with either their  respective
matched (vehicle) control, negative control,  or pooled negative control
6/
—   Report of the Secretary's  Commission of Pesticides and Their
    Relationship to Environmental  Health,  Parts  I  and II.   U.S.  Dept. of
    Health, Education  and  Welfare,  Washington,  D.C.   (1969),  p.  470.

—   Ibid p. 483.
                                  11-13

-------
 groups;  and a statistically significant  incidence in female mice of strain
 X when  compared  with  the  pooled  negative control group only.  Contrary to
 Monsanto's  assertion,  hepatic tumors  are known to occur raore frequentlv in
 male  mice than in  females.— (Albert,  1979a).

               c.   No Statistically  Significant Increase in Pulmonary
                    Adenomas

                    Monsanto Company  argued  that there was no statistically
 significant increase  in pulmonary  adenomas in either sex of either strain
 of mouse, whereas  the  CAG report indicated a small statistically
 significant increase of .lung adenomas in both sexes of strain X and in
 males of strain  Y  (Monsanto Co., #1A[30000/15]).

                    The Agency rejects this  point of rebuttal.  Having re-
 analyzed the data,  the CAG finds a borderline statistically significant
 increase of pulmonary  adenomas in  the diallate-treated males of strain X as
 compared with the  matched (vehicle) control  group     (p = .051) and with
 the pooled  control groups (p » .041).  (See  Table II-6) (Albert, 1979a).

      B.   Analysis of  Data Submitted  Since PP 1  for Other Possible
          Adverse  Effects

          1. Mutaqenicity

               In  PD  1 the Agency, based on  the  two available studies,
 stated  that it had insufficient  data  to  indicate diallate is mutagenic.
 The two studies  were  one,  an Ames  test in bacteria, and the second, a
 dominant lethal  study  in  rice (Keplinger 1974).   On the basis of this
 conclusion,  the  Agency requested comments and information on diallate's
 mutagenic potential.   Additional studies were submitted and evaluated, and
 the Agency  now concludes  that diallate meets the criteria stated in 40 CFR
 162.11  for  mutagenicity by multi-test evidence.   The additional studies
 which led to the Agency1 s finding  are discussed below.

               In  response to the  Agency's request for additional
 information with regard to the possible  mutagenic properties of diallate,
 Rinkus  and  Legator (1977)  submitted a study  entitled "Mutagenicity of
 Diallate."   This study, an Ames  test,  investigated the mutagenic effects of
 diallate in five strains  of Salmonella.   Diallate exhibited mutageni:
 activity in strains TA 1535 and  TA 100 which exceeded the mutation
 frequency of controls  by  factors of approximately 20 and 12 respectively.
 This  activity was  only observed  in the presence  of a microsomal activation
 system.  No activity  was  observed  in  strains TA 1537, TA 1538, and TA 98 in
 identical tests.
8/
—   Stewart, H. I>.  (1976).   Comparative  aspects of certain cancers.
    Chpt.  10 in Cancer, A Comprehensive  Treatise,  Vol.  4,  F. F. Becker
    (ed.), Plenum Press, New York.


                                 H-14

-------
                         Table  II-6.   NCI  Mouse Study (Innes Study):   Lung Tumors in Mice
                                           Ingesting Avadex (Diallate)
Dose
Group
                                                                 Strain X
                                                            Strain Y
Male
                                                                          Female
                                                            Male
Female
                          Pulmonary Adenoma   Matched Control
                          Pulmonary Carcinoma
                                        0/16*(0%)  0/16 (0%)2/18 (11%) 1/17(5%)
                                        0/16 (0%)  0/16 (0%)0/18 (0%)  0/17(0%)
                          Pulmonary Adenoma   Negative Control
                          Pulmonary Carcinoma
                                        2/17 (12%) 1/18 (5%)2/18 (11%) 0/17(0%)
                                        0/17 (0%)  0/18 (0%)0/18 (0%)  0/17(0%)
                          Pulmonary Adenoma   Pooled Control
                          Pulmonary Carcinoma
                                        5/79**(6%) 3/87 (3%)10/90(11%) 3/82(4%)
                                        0/79 (0%)  0/87 (0%) 0/90 (0%) 0/82(0%)
i
t~*
in
                          Pulmonary Adenoma     560  ppm
                          Pulmonary Carcinoma
                                        4/16 (25%) 2/16 (13%)4/18 (22%]
                                        0/16 (0%)  0/16 (0%0 0/18 (0%) 0/15(0%)
* p = .051 (Fisher Exact Test) for the incidence of pulmonary adenomas in
the treated group compared with the matched control.
                         **
                            p = .041  (Fisher Exact Test)  for  the  incidence of  pulmonary adenomas in
                         the treated  group compared with  the  pooled control.

-------
               In PD 1 an unevaluated study by  Sikka  and Florczyk (1978)
was mentioned.  The study investigated the ability  of diallate to induce
mutations in four strains of Salmonella typhimurium (TA 100,  TA 1535,  TA
98, and TA 1538) with and without a rat-liver microsomal activation system.
The study has now been evaluated and found to show  activity at the 1 ug per
plate level in the TA 100 and TA 1535 strains indicating base-pair substi-
tutions with metabolic activation.  Diallate did  not  cause mutation in
strains TA 98 and TA 1538 {frameshift mutants).   These results confirm the
findings of Rinkus and Legator.

               Litton Bionetics, Inc. (LSI)  (Brusick, 1977b)  investigated
the effects of diallate in the L5178Y mouse lymphoma  cell.  Ths study
concluded that "The test compound, CP 15336, induced  forward mutation at
the TK locus in L5178Y mouse lymphoma cells in  the  presence of an uninduced
mouse liver S-9 metabolic activation system."   No dose-related effects were
observed in the absence of metabolic activation.

               Studies by SRI International  (Simon, 1978) for EPA show that
diallate does not induce gene mutation in £_._ coli (WP2).  The bioassay was
designed to monitor induced genetic alteration  in E.  coli at the tryptophan
locus.  However, tests in JE. coli are not as sensitive as tests in
Salmonella and, therefore, positive findings may  not  be manifested through
experiments in this organism (Sandhu, 1978).

               Diallate's potential to cause primary  DNA damage was studied
in two strains of Saccfaaromyces cerevisiae.  SRI  International (Simon,
1973) employed strain D_ to measure induced mitotic recombination and LSI
(Brusick, 1977a) used strain D  to measure gene conversion,  while posi-
tive results were reported for mitotic recombinations in the SRI study, the
LSI study results were negative.  However, this mixed finding does not
detract from the finding of induction of mitotic  aberrations, a supportive
finding for mutagenesis.

               Monsanto submitted a dominant lethal study in mice in 1975
(Keplinger, 1974).  This study done by 1ST was  reported by Monsanto as a
negative study.  Th*» TDAP program reviewed this study and found it to be
invalid because of the lack of raw data.

          2.   Neurotoxicity

               The Agency concluded in the PD  1 that  diallate is neurotoxic
at the dose levels tested.  This conclusion was based on an IBT study in
chickens {Keplinger,  1976b).  That study did not  provide a dose-response
relationship nor a no observable effect level which is needed to determine
a margin of safety.

               In the diallate  PD  1, published on  May 31, 1977,
registrants were given  180 days to complete  appropriate neurotoxicity
studies and submit the r-.suits to the Agency or face  possible cancellation
under the provisions of FIFRA Section 6(b)(1)—  .   Monsanto submitted an
9/
—    In  1978 FIFRA was  amended  to  provide for suspension under Section
     3(c)(2)(B).
                                   11-16

-------
IBT study in which diallate was administered  to  chickens—'  at  dose
levels ranging from 0.01 to 0.32  gm/kg  which  were  administered  twice  daily
for 3 consecutive days  (Phillips,  1977).  Twenty  days  following  the  initial
dose, all surviving birds were again  given  the same dose regimen.   Controls
were dosed with 0.32 gm/kg corn oil  and positive controls received  500 mg/
kg TOCP on day 0.

               All positive controls  exhibited lesions typically associated
with delayed neurotoxicity (Phillips, 1977).   No such lesions were  found in
the negative controls.

               Two test birds,  one in the 0.04 gm/kg and one in the 0.16
gm/kg group showed focal lesions  of  axonal degeneration and secondary
demyelination  in  the  sciatic  nerve.   While these lesions were described as
morphologically indistinguishable from those observed in the positive
control birds, the affected birds showed no clinical signs which could be
characterized  as  delayed neurotoxicity prior to sacrifice.  No dose
response  relationship was established nor were there any reasons given for
the  absence of lesions at the highest dose.

               Both  of these  I3T studies were reviewed by the TDAP.  The
original  study, discussed in  PD1, was validated and therefore does demon-
strate  that diallate given at 312 mg/kg causes neurotoxic effects.  How-
ever, as  stated in PD 1, this study does not show any dose response with
which to  determine ultinate  safety.  The second IBT study on neurotoxicity
was  declared  invalid.   TDAP  found that raw data were totally missing on
this experiment.

               The Agency's  concerns on neurotoxic effects of diallate have
not  been  addressed by the registrant.  While attempting  to satisfy the
needs of  EPA,  Monsanto has failed to take  adequate precautions  to insure
validity  of their data.  Therefore, although the  Agency  does not have
 enough  data to quantify the potential  neurotoxic  risks  of diallate, it has
 calculated the annual applicator exposure.  The Agency  has determined that
 the  effect level is  600 times greater  than the  exposure  level.

           3.    Reproductive Effects

                Prior to the RPAR review  the  Agency requested Monsanto to
 perform a 3-generation  reproduction  study  in rats.   However, no time  limit
 was imposed on the registrant.  At the time  of  PD 1  only the data  in  the
 F° (parental) generation had been submitted.  The final report  is
 expected in 1980.  This study will be  evaluated by the  Agency  and  its
 results will be  included in PD 4.
 —/  The  study used  5  birds  per dose group with dose given at 0.01,
     0.02, 0.04,  0.08,  0.16,  and 0.32 gmAg-
                                   H-17

-------
     C.   Exposure Analysis

          The Agency analyzes exposure to a pesticide as part of  the
overall risk assessment.  The complete exposure analysis includes
incremental exposures to various populations depending on  the route of
exposure (e.g. dietary exposure to the general population,  occupational
exposure to applicators, drift to farm families, etc.).  In compiling the
analysis, the Agency takes into account the use patterns and methods of
application so all populations likely to receive exposure  are included  in
the analysis.

          1.   Spray Applicator Exposure

               a.   Spray Applicators

                    Information concerning spray applicator exposure was
provided by the Environmental Fate Branch of the Agency's  Hazard  Evaluation
Division (HfiD) (Selim, 1978).  Diallate is applied primarily as an emulsi-
fiable concentrate with ground equipment by boom sprayer.   Spray  applica-
tors are exposed to diallate by 1) dermal exposure during  the loading of
the sprayer and during the application process, and  2) inhalation of the
volatilized compound.

                    Exposure from diallate has not been maasurad,  conse-
quently the Agency used published data on triallate  to prepare  axposur-3
analysis for diallate (the mode of application ar.d chemical-physical
propertiss of triallate are similar to diallate).  To further check these
results, the Agency validated the assumptions used for diallate by an
exposure analysis which was published for paraquat.   The pattern  of
exposure to paraquat is similar for the applicator wording with diallate.
Therefore, the Agency felt justified in utilizing the paraquat  data for
extrapolation to diallate.  Both of these extrapolations produced very
similar results.

                    The estimated dosages were reported  in mg/hr, then
converted co mgAg/year.  The conversion from dermal exposure expressed in
mg/kg/year to equivalent lifetime dietary exposure expressed in ppm in  the
diet is as follows:
               60 kg  (worker  dermal  exposure- in mg/kg/yr)  x 40 yr
               365 d/yr  ^70  yr x  1.5  kg/d
               6.26 x 10    x  (worker dermal  exposure in mg/kg/yr)
               lifetime  dietary exposure  in  ppm
where X  is ppm  in  diet,  60  is  average body weight in kg, and 1.5 kg is the
average  daily dietary  intake.   A  40-year working history and a 70-year
lifetime is  assumed  for  the applicators.  This value (X) will be used to
calculate lifetime probability in the Risk Assessment Section (II.D.1).
                                 11-18

-------
                    Table II-7 presents  data on the absorbed dose in mg/kg/
year and parts per million  for spray applicators exposed to  diallate.
Rubber gloves and coveralls were  considered as  protective clothing when
calculating the exposure reduction levels.

                    The annual exposure  to  applicators of diallate is 0.516
mg/kg/yr (.0323 ppm) and 0.018 mg/kg/yr  (0.0011 ppm)  from dermal and
inhalation exposure respectively  (Selim, 1978).  These estimates are based
on 10% absorption from both routes (Gardner, 1980).  Based on the assump-
tion that dermal exposure would be reduced  by a factor of 4  if protective
clothing is used, the dermal  exposure level can be reduced to 0.13 mg/kg/yr
(8.1 x 10   ppm) (Selim, 1978).

               b.   Granular  Applicators

                    An applicator's exposure from granular diallate
applications would be lower than  an applicator's exposure using the
emulsifiable concentrate formulations.  With the granular formulation there
is no chance of exposure from spray drift or from spray splash as there is
with the enulsifiable concentrate formulations.  The granules do not
adhere to the skin as the emulsifiable concentrates would.  Additionally,
because diallate is applied during the late fall or early spring in the
northwestern states, it is  likely that most diallate applicators would be
wearing clothing such as long sleeved shirts and trousers to protect
themselves from rhe cold as well  as protect uhemselves against the
diallate.  However, for the brief period of loading the (ore-mixed)
granular diallate formulations there is  a potential dermal and inhalation
exposure hazard to  the applicator due to dust from the granules.  To
mitigate this potential hazard,  the Agency  suggests the use of rubber gloves
and cloth face masks for applicators during the loading process.

          2.   Dietary Exposure

               The human population encounters direct dietary exposure to
diallate. residues through  consumption of the following foods:  barley,
lentils, peas, soybeans, sugar beets, corn   (grain), and flax (seed).

               Maximum worst-case exposure  was developed from tolerances
established  for residues of diallate in foods.  The FDA, in its Market
Basket Survey, has  not analyzed raw agricultural commodities specifically
for diallate.   It is assumed  that residues  are present in all individual
raw agricultural  concnodities  to che extent  permitted by the tolerances and
that  the commodities are  uniformly distributed throughout the country.
Table II-8 presents  the dietary exposure of the entire U.S. population to
diallate.

                A  second  set of estimates were-developed which were based on
available  information  concerning the percentage of crops treated and were
provided by  the Agency's  Benefits and Field Studies Division (Lewis,  1978).
Table II-8 presents the exposure estimates  when the percentage of crop
treated  is considered.
                                   11-19

-------
     D.   Risk Assessment

          1.    Oncoqenic Effects

               The cancer risk assessment of diallate is based on the
principles and procedures outlined in the EPA interim cancer risk assessment
guidelines (41 FR 21402, May 25, 1976).  These guidelines specify that a
substance will be considered a "presumptive cancer risk when it causes a
statistically significant excess incidence of benign or malignant tumors  in
humans or animals." Current and  anticipated exposure levels are appro-
priate considerations for establishing cancer risk estimates.  These
estimates may be derived from a variety of risk extrapolation models such
as the log-probit and linear non-threshold models.
                                  11-20

-------
      Table II-7  Diallate Dermal  and Inhalation  Exposure  to  Spray
                  Applicators and  Equivalent  Lifetime  Dietary Exposure
Route of
Exposure
Estimated
Dose
mg/hr
Duration
of
Exposure
Dose
in
mg/kg/yr
Dose in
ppm
DERMAL

Absence of
protective cloth-                                                 __
ing (Assuming       2.58         6-12             .258       2  x 10
10% dermal   .                    hr/yr.           .516       3  x 10
absorption)—

With protec-
tive clothing                                                      _
(Assuming 10%         .65         6-12            0.065       4  x 10
dermal     ^.  .                  hr/yr.           .13        8  x 10~
absorption)— —

INHALATION
                                                                  -4
Without mask          .09         6-12             .009       6  x 10
                                 hr/yr.           .018       1  x 10

With mask                                                         _5
(10% of               .009       6-12             .0009      6  x 10_4
exposure                         hr/yr.           .0018      1  x 10
without
mask)
 1/    10%  absorption is assumed in absence of data on diallate (Gardner,
~"     1980).
 2/    Assuming  that protective clothing provides a four fold reduction in
~     exposure.  (Selim, 1978)
                                   11-21

-------
     Table II-8.
    Annual tJ.S. Population Dietary Exposure to Diallate, Based
      On Tolerance Levels and Percent of Crop Treated
Exposure Based on
100% of Crop Treated
with Residues at
Tolerance Levels
Source ppm
Barley 0.000013
Lentils 0.00002
Peas 0.00035
Potatoes 0.00271
Saf flower* 0.000013
Soybeans 0.00046
Sugar beets Q. 00001
Corn, grain 0.0005
Flax, seed 0.000013


Percent of Crop
Treated with
Diallate
0.10
33
10
0.46
100
0.20
14.3
0.009
3
Exposure Based on Actual
Percent of Crop Treated
with Residues at
Tolerance Levels
ppm
0.00000
0.00001
0.00005
0.00002
0.00002
0.00000
0.00012
0.00000 **
o.ooooo **.
Total
0.004089
                                                            0.00022
     *    100% is assumed in the absence of data

     **   Not actually "0", remaining significant figures  truncated.
                                  11-22

-------
               In accordance with  these principles,  the EPA Cancer
Assessment Group (GAG)  (Albert,  1979b)  developed risk estimates  based on
several different models and a  range of exposure estimates.   After
reviewing the data sources  and  the preliminary risk  estimates, the GAG
concurred in recommendations that  the final quantitative risk estimates be
based on data from the  NCI  mouse study (Innes  study)  using the one-hie
model.  The CAG used the Innes  data because animals  in the Innes study were
fed the compound beginning  at a younger age than were animals in the NCI
(rat) or IBT studies, and therefore provided the most sensitive  animal upon
which to base the conservative  analysis.

               To develop a risk estimate,  CAG evaluated the animal test
data and the human exposure data using several different models.  They
selected the one-hit model  because it provided the most conservative
estimate of potential risk  to humans.

               As explained above/ the animal  bioassay data used for the
quantitative risk assessment were  based on  the Innes oral feeding study in
mice.  In this study one treated group of mice were  fed 560 ppm  diallate in
the diet.  A statistically  significant higher  incidence of hepatomas in
males of both strains X and Y was  observed, as compared to matched controls
(see Table II-9).

               The proportion of hepatomas  observed  in Strain X  males was
used to calculate the slope parameter, for the  one-hit model, adjusting for
background tumor incidence. Therefore, using the proportion of  hepatomas
in the matched control  9TT°up and the treated group,  the one-hit  slope
parameter is as follows:—

                   B -  • In [(1-Pt)/(1-Pc)] /y
                              B -  - In [(1- 13/16)/(1-0/16)1/570
                                 »  2.989.  x  10

Prom the slope, the estimate life-time probability can be estimated from
the following equation.

                          P » BX * (2.989 x 10~3)X
                   (where X is  the ppm in the  diet from actual exposure
                   and  from equivalent dermal  exposure as calculated in
                   Section  II.C.1.)
 —        Where:  B *  slope coefficient of the one-hit model
                  P *  (Pt-Pc)(1-Pc)
                 PC *  Incidence of hepatomas in control animals
                 Pt »  Incidence of hepatomas in test animals
                  Y »  Test animal exposure (ppm)
                  x *  Potential human exposure (ppm)
                                 11-23

-------
Table II-9.  NCI Mouse Study (Innes Study):  Liver Tumors (Hepatomas) in
             Mice Ingesting Diallate
Dose Group
Matched Control (vehicle)
Negative Control
Pooled Control
560 ppm
Strain
Male
0/16*
1/17*
8/79*
13/16
X
Female
0/16
0/18
0/87**
2/16
Strain
Male
1/18*
3/18**
5/90*
10/18
Y
Female
0/17
0/17
1/82
1/15
*  Statistically significant when p is £ .01,
** Statistically significant when p is £ .05,
                                   11-24

-------
               The exposure  estimates  for dietary and applicator  exposure
were factored into the above equation.   Risks  from these  exposures  are
shown in Tables 11-10 and  11-11.   A dietary worst-case risk  of  10
occurs at tolerance levels assuming that 100%  of  the crops are  treated.  It
is not likely that 100% of the  crops will be treated with diallate.   There-
fore using the projected percentages of  treated crops from Section  III,
results in a reasonable worse case risk  of 10

               Risk to applicators without protective clothing  is estimated
at 10  , a relatively high risk.   With protective clothing,  risk  to
applicators is improved to 10   (Table 11-11).

               Aggregated, these  risks imply that 1 in 10,000 applicators
might have a lifetime risk of developing a diallate induced  tumor taking
into account both occupational  and dietary exposure.  It  is  estimated that
there are 2380 diallate applicators (Selim, 1978).  The general population
would have a lesser risk  (10-7),  but this is based on current usage and
should any increase in usage occur, the  dietary risks would  increase.

          2.   Mutagenic Effects

               While adequate evidence exists to  establish the  rtiutagenicity
of diallate in in vitro systems,  no quantitative  assessment  of  risk can be
made because of (1) insufficient  data  from mammalian test systems(Mauer,
1978), and (2) no generally,  applicable method has been developed  to
quantify autagenic risk.—   Recent studies have  established a  strong
correlation between a chemical's  carcinogenic potential and  its ability to
induce mitotic recombination (Sandhu,  1978).
          The Agency  has  not yet developed a standard procedure for
          defining mutagenic risk in quantitative terms.   At the present
          time/ much  attention is being focused on developing a battery of
          test systems  and other data that are predictive of mutagenic risk
          in humans.  Until such time as more quantitative methods and
          procedures  for  risk estimation are developed for each mutagenic
          endpoint of concern, the Agency will evaluate each mutagenic
          chemical on a case-by-case basis,   taking into  account all
          available test  data.  The approach taken by the Agency will of
          necessity be  conservative in order to assure that man and the
          environment are protected from the risk of "unreasonable adverse
          effects" through the action of mutagenic agents.  The evolving
          nature  of methodology in the field of mutagenicity testing
          dictates that the Agency will revise its risk estimation
          procedure for future chemicals under evaluation as superior risk
          predictive  models and other relevant information become
         - available.  As  well, the Agency will revise its risk estimates
          for chemicals which have previously been subjected to risk
          assessments if  additional more relevant test data and other
          predictive  information are developed.
                                   11-25

-------
Table 11-10.  Lifetime Spray Applicator Cancer Risk from Dermal and
              Inhalation Exposure to Diallate
                    Equivalent Lifetime
                    Dietary Dose Assuming
                    12 Hour Annual Exposure       Lifetime Probability  of
Route (ppm)
DERMAL
Without protective 3 x 10
clothing*
With protective
clothing (10% -
absorption)* 8 x 10
INHALATION
Without aask 1 x 10"
With aask (10%
of exposure .
without aask)* 1 x 10
Cancer Due to Diallate
1 x 10~4
2 x 10~5
3 x 10~6

3 x 10~?
* 10% absorption is assumed in the absence of data on diallate  (Gardner
  1980).
                                   11-26

-------
   Table 11-11.  Cancer Risk to U.S. Population from Dietary

                      Exposure to Diallate
Source
Lifetime Probability
of Cancer Based on
100% of Crop Treated
Residues at Tolerance
Levels
Lifetime Probability
of Cancer Based on
Estimated Percent of Crop
Treated with Residues
at Tolerance Levels

Sarley

Lentils

Peas

Potatoes
Saf flower
Soybeans

Flax seed

Beet Sugar

Corn

Total

4

6

1

3
4
1

4

1

1

1
-8
x 10
-8
x 10
-6
x 10
-6
x 10
x 10~8
-6
x 10
-6
.9 x 10
-6
x 10
-6
X 10
-5
x 10
-11
4x10
-8
2 x 10
-7
1 x 10
-8
4 x 10
4 x 10~8
3 x 10~9

Negligible
-7
2 x 10

Negligible
-7
4 x 10
                               11-27

-------
     E.   Risks Associated with Alternative Chemicals

          Several chemicals have been proposed  as  alternatives  should
diallate become unavailable.  In non-irrigated  areas,  eptam would be the
major substitute with barban and dalapon used to a lesser  extent.   In
irrigated areas cycloate and barban are the major  substitutes.   Eptam and
barban are substitutes used on flax.  Triallate, prophaai and barban are
alternative chemicals, for use on lentils and peas.

          The data bases for the alternative chemicals are not  complete.  A
complete list indicating the studies which the  Agency  has  on hand appears
in Table 11-12.   Based on the studies reviewed for all alternatives other
than triallate, no unreasonable adverse effects were found associated with
the proposed alternatives.  Because of the lack of chronic studies no quali-
tative ranking of alternatives can be made with regard to  their relative
toxicities.

          The battery of toxicological tests performed on  triallate include
mutagenie, chronic feeding, teratogenic, and neurotoxic studies. Triallate.
was found to exhibit mutagenic activity in the  Ames test in bacteria  with
metabolic activation.  It was also found to be  positive in yeast when
tested for mitotic recombination.  Negative findings were  reported to gene
conversion in yeast, mouse lymphoma, and dominant  lethal assays.  When
triallate is compared to diallate in tests which are positive for both,
diallate.is at least 3 times more active.

          Negative findings for triallate were  reported in a chronic
feeding study in rats.  This study was evaluated by TDAP and found to be
valid only as an oncogenic screen test because  of  ieficiences in the
experimental design.

          The teratogenic and neurotoxic studies on triallate were
performed by IBT.  These studies were evaluated by TDAP and found to be
invalid.
                                     11-28

-------
                                         TOXXCOIXX3ICAI, DATA SUMMARY  TOR DIALLATE ALTERNATIVES (CHRONIC DATA)
M
iD
CHEMICAL ONCOGENICITY

Trialiate **Rat






Barban N/A
Cycloate N/A
Oalapon N/A

EPTC*

CHRONIC REPRODUCTIVE
MUTAGENZCITY FEEDING TERATOGENICITY EFFECTS
Gene Conversion (-)
Mitotic Recom-
bination (t) **Uat (-) Rabbit (Invalid) N/A
bacteria 4+ (with
activation)
2- (no activation)
Dominant Lethal (-)
Mouse Lymphoma (-)
N/A Dog (-) N/A N/A
N/A Rat (-) N/A N/A
N/A Rat (-) Dog (-) N/A
Rat (-)


NEUROTOXICITY


Chicken (2 In\
Studies)




N/A
N/A
N/A



    (+) Positive Study
    (-) Negative Study
    *  EPTC has negligible residue tolerances, therefore no chronic data have been
      submitted.
    ** This IBT study listed under "Oncogeniclty" and "Chronic Feeding," was
       reviewed by TDAP.  While the study was considered invalid as a  chronic
       feeding study, it was found valid as an oncogenic screening study.

-------
III. Benefit Analysis

     As a pre-emergence  selective herbicide,  diallate is used as a soil
treatment on field crops  for  control  of wild  oats.   Because Monsanto is the
sole producer of diallate,  production and marketing data are confidential.
However, it is estimated  that approximately 390,000 pounds active
ingredient of diallate were applied annually  to 319,000 acres of sugar
beets, flax, lentils, and peas,  the major use sites of diallate between
1976 and 1978.  Small amounts of diallate are also applied to potatoes,
soybeans, barley, corn,  and forage legumes.—

     The Agency has received  16  submissions from registrants and interested
parties pertaining to the benefits of diallate, in particular as it is used
on sugar beets, lentils,  and  peas. The Agency considered this information
in analyzing the benefits of  diallate.

     For sugar beets, flax, lentils,  and peas, the estimates of acreage
treated, the identification of biologically viable alternatives, and the
use data were based primarily on Assessment of the Need for Diallate in
Agriculture, USDA/State  Assessment Team on Diallate (September, 1977, and
1979 modifications).  The economic analysis based on these biological data
was prepared by Development Planning  and Research Associates, Inc. in March
of 1979.

     However, lack of published  data on yield changes limited certain
aspects of the analysis.  Expert opinion was  used in place of these data.
Alternate weed control strategies also lacked firm data, necessitating the
use of expert opinion to  generate impacts of  alternate control programs
(USDA/State Assessment Team,  1977).

     The alternatives to  diallate were selected on the basis of cost,
efficacy, and availability.   Partial  budgeting was employed to assess the
economic impacts of diallate  cancellation.—   The partial budgeting
methodology allowed the  change in the cost of weed crntrol to be measured,
together with the effect  on gross returns associated with substituting
alternative weed control  practices while all  other inputs were held
constant.  The economic  analysis also assumed that in some instances the
cancellation of diallate  would cause  growers  to shift production to
alternative crops.  In these  cases, net returns to the producer associated
with the production of alternative crops were evaluated.
1/   The OSDA/State Assessment  Team on Diallate (August 31,  1977)  esti-
"~    mated that approximately  10,000  acres or less of each of these crops
     were treated with -diallate annually,  accounting for 0.5% or less of
     the total crop planting in each  case.  The Monsanto ;-ubmission filed
     September 9, 1977,  did not address the benefits of diallate use on
     any of these minor  crops.

2/   The partial budgeting methodology allows the measurement of the
""    change in the cost  of pest control and the effect on returns  associated
     with substituting alternative chemical and non-chemical pest  control
     practices into the  budget  with all other inputs held constant.


                                     III-l

-------
     If the major uses of diallate on sugar  beets,  flax,  peas, and lentils
were cancelled, varied effects on producers  would  result.  Because an
acceptable substitute herbicide is available,  peas  and lentils would
actually return more income to producers if  diallate  were cancelled.  The
positive economic impact is based on increased yields which are due to a
decrease in phytotoxicity and better wild oat  control in  some instances,
see further discussion in Section III.C.  Sugar beet  producers would suffer
an estimated adverse impact of $4.0 million.   Flax  producers are expected
to experience an estimated $0.4 million economic loss.  The net adverse
impact upon all affected user groups is approximately $3.2 million
annually.—   These aggregate economic impacts  are  summarized in Table
III-1.  The following subsections (A through E)  briefly explain the
economic impacts involved should the major and minor  uses of diallate be
cancelled.

     A.   Sugar Beets

          Sugar beet production subject to wild oat infestation is located
in the non-irrigated acreage of Minnesota and  North Dakota, and irrigated
acreage in Montana, Wyoming, Idaho, Utah, Washington, Oregon, and
California.  As an average for the period 1976-1978,  see  Tables III-2 and
III-3, these nine states planted 995,000 acres of  sugar beets annually or
72.6% of the U.S. total acreage.  Of the nine-state total, 418,000 acres,
or 42%, rfere in Minnesota and North Dakota;  577,000 acres, or 58%, were in
the seven western states subject to wild oat infestation.

          Diallate is a major herbicide used to control wild oats in sugar
beets.  It is most widely used in non-irrigated acres.  As an average
during the 1976-1978 period, diallate was estimated to  have been applied to
185,175 acres of non-irrigated beets, and to 35,800 acres of irrigated
beets.  In total, diallate was applied to 220,975 acres,  or 22% of the
total sugar beet acreage subject to wild oat infestation.

          Annual use of diallate to control  wild oats in  sugar beets, as an
average for the period 1976-1978, was estimated to  have been 231,470 pounds
(active ingredient basis) in non-irrigated areas and  44,750 pounds in
irrigated areas.  Total estimated use of diallate on  sugar beets is thus
276,220 pounds.

          Two formulations of diallate are used en  sugar  beets.  Granular
diallate is applied to approximately 15% of  the treated acreage while the
eoulsifiable concentrate is applied to approximately  85%  (Lewis, 1979).
The degree control of wild oats provided by  each of these formulations
appears to be the same for the fall application. There is some decrease in
control of wild oats when the granular is substituted for the emulsifiable
concentrate in the spring application.
3_/   This reflects an estimated 1.2 million  net  increase  in revenues  when
     triallate is substituted for diallate where  it  is  also registered for
     these uses.
                                      IIX-2

-------
                  Table  III-1.  Annual  Economic  Impact of Cancellation of Diallate
                                on  the  Major  Use Sites -
H
H
U)

Site
Sugar
beets
Flax
Lentils
Peas
Maj or use
i-nt-al

Extent
of Use
(thousand
pounds)
276.2
38.4
43.3
30.4
IPO ^

Units
Treated
( thousand
acres)
221.0
30.7
43.3
24.3
1 1Q 1

Percent of Aggregate
Total Units User
(million
dollars)
16.7 4.0 loss
3.0 .4 loss
38.0 .7 gain
10.5 .5 gain


Economic Impacts
Market & Consumer .
(million dollars)-7
none
none
none
none


Total Market Value
(million dollars)
500
59
34
23.6


     —   Source:   Economic  Analysis  of  Effects  of  Restricting  Use of
                       Diallate  on  Sugar Beets,  Flax,  Lentils,  and Dry Peas.
                       Development  Planning  &  Research Associates, Inc.
                       March,  1979.

     -   1978 price  levels

-------
                Table III-].  Katlaatad annual aggregate uae coat* tot  eliwluattiui diallete on icclgalud aityar b««tB, 1»76-1»75
n
H





With dlellata
Without diallata
Cycloate
Propham
(ulapon
Bactan
Paraquat
t'nt rout t»l
tlauJ uoudinq a/
fJJcd cultivation b/
DuliycJ auciliny
St. k(t to othur crop*
tub-total w/o dlallate
kUt change w/o dtallate

Acrga*


35, 800

a, 200
300
1,800
MOO
1,000
5,600
8,200
10,700
900
6,400
45,700


Material

(1)
225,540

112.536
3,860
a, 262
20.118
36.000
	
...

	
— --
361.076
15,516
Coat*
Application


125,658

28,782
1,051
2,862
2,862
2.862
	

	
	
	
18.421
(87,217)

Yield loge* • Other coat a

<« (f)
	

167,112
42,660
42,660
122,860
111.100
665,350
164.000
129,418 27.711
42,660
1.652,000
1,145,840 1.841.711
1,145.840 1.841.711

Total cost*

(1)
151.198

388.450
47.671
51.784
146.040
171.962
665.350
164.000
157, 111
42,660
1,652,000
S, 489, 040
1,117.852
       •  USOA/Statu Kuei.ai.nunt Study, perc«itt«<)* of original acr«ag« ••tlmatad  to IM treated with dlalUt*.
      *•  Loss (.ur *cr« tram USDA/Stat* Agtunamor.l; Study, 1977.  Sugar boat*  valued at $21.70/ton.
      a/  Chaiq* at $20 pur acre, baaed on interview Information Cram augar bout proce*sora.
      b/  north-Dakota Crop and Llvaatock Hcportlng Service, 1&J7.

-------
                       III-).   KHtliMtud annual ai>n
165,175
	
30,040
97,270
25.565
Cultural i:ifuui —
..


b.



lUitU
a.


b.



with dial lale
1 . iiUU'l
2. culiivjlo (3x)
witliuui iliultatu
1 . btiuil
2. .ciiliiviitu (3x)
3. <;ullivulti (4xt
I^UM
with 'Hal Idtu
1 . I liinul ntj
2. (^C4lii,.j (2H)
uitti'»ji ilidltutu
I. iliiiiuiii'j
2. w.^Ji,,., |2x)
). ,x.,., w,.,ll^
- — —
1U-...175
1U5.175
	
IUL.OU5
IU5.175
00,550
	
	
74,070
in. \'<>
-',;I5
Couta
Hatorlal Application Yield loaa Total cost Clianga In cott

(() ($) (»> It)
* 116.292
1.166.60U 649,964 	 1.016,572
1,932,064
345,960 133,520
1.104,060 154,659
152.419 42,230 	
	 * 121,047
	 4.466.420
2.4U1.345 546,266
1,438.809 	
4,700,267
2.503.IU7 551.075 65,fl52*
1,430,009
—. 224,344
	 	 	 » 430,126
4,710,060
1,917.672
2.703,100
	 	 	 5,140.906
2.051.739
2,'JUU,177
K)J,070 	
      *  h(.Mluc..| yi.,|.| of  2  Tiiii/acru jl  *.'<). ^ti (,ur ton  (aoutco of ylaM louu tiul Iwatu,  llSDA/Stato  AnuoBsnont  Tuam, 1977)

-------
          Should diallate use  on  sugar beets be cancelled, procedures will
substitute an integrated chemical  and  cultural strategy for wild oat
control.  In non-irrigated  areas,  eptant is a major substitute for diallate,
with barban and dalapon also used  for  wild oat control.  In irrigated
areas, cycloate and barban  are the major substitute herbicides.  Sugar beet
producers would also need to increase  the amount of mechanical and hand
labor used to cultivate their  beets.   Some producers may experience yield
losses with alternative•weed control strategies (including use of
alternative herbicides and  delayed seeding),  while other producers may
shift to alternate crops.

          Economic impacts  would result from changes in herbicide costs,
increased mechanical and hand  cultivation costs, additional reseeding
costs, decreased yields, and shifts to other crops.  Should diallate be
cancelled, sugar beet producers may experience estimated economic losses
potentially as high as $4.0 million (Table's  III-2 and III-3).  Of that $4.0
million inpact, approximately  $3.1 million would result from adverse
effects in the irrigated sugar beet areas of the Western U.S., and $0.9
million would be attributed to ramifications  in the non-irrigated Red River
Valley area of North Dakota and Minnesota.  In the Red River Valley most of
the adverse impact would derive from possible increased hand labor costs
and possible increased mechanical  cultivation costs.  In irrigated areas,
losses would be nearly equally divided between yield losses and lost
revenue resulting from changes in  crop production from sugar beets to other
crops.

          The effect of diallate cancellation on a typical Red River Valley
farm with 135 acres of sugar beets would be  increased costs of $870
annually.  The average cost increase per acre of sugar beets is $4.69.
This, of course, assumes the typical farm would make all herbicide,
cultural, and labor adjustments in the same  proportions as the entire
Valley area.  Changes in net yield are not anticipated.  The average
producer could expect his net  returns  to land,  management, and labor to
decrease by only 2.1%, from $41,102 to $40,232.

          The effect of diallate cancellation on a typical irrigated sugar
beet farm would be much more severe, with an  average impact of $87.65 per
acre of sugar beets.  On a  farm which  ordinarily treated 100 acres of sugar
beets with diallate, the adverse annual  effects  would amount to $8,765,
most of which would be due  to  reductions  in  net  returns (over variable
cost).  For a typical farm, crop returns  over variable cost would be
reduced nearly 18%.  This loss, however,  is  to the individual fanner.
There is little overall yield  loss since  diallate is not used extensively
in irrigated plantings.  In a  switch to  the  granular formulation of
diallate alone for the control of wild oats  in sugar beets, the increased
cost to growers in the short run will  be  approximately $6-$7 per acre
treatment, given current prices.

          While the cancellation of diallate  may pose significant problems
to the local grower, this effect will  not seriously reduce total U.S.
                                  Ill-6

-------
production because the sugar  from  sugar  beets  represents only a small
percentage of total U.S. sugar  consumed.   Over 50%  of the U.S.  sugar is now
imported due to the favorable tariff  for imports.   A decline in profit
because of sugar imports is already reducing sugar  beet acreages.

     B.   Flax

          United States.flax  production  is concentrated in the states of
Minnesota, North Dakota, and  South Dakota.   This area produced 98%  of U.S.
flax on an average of 1,025,000 acres between  1976  and 1978.  Approximately
3% of the total flax acreage  (30,750  acres)  was estimated to have  been
treated with diallate during  this  period - an  estimated 38,440  pounds
(active ingredient basis) of  diallate annually.  The emulsifiable  concen-
trate formulation is the only form of diallate currently registered for use
on flax; however, granular diallate could be used on flax if it were
registered, but at slightly higher costs. • Herbicide use on flax is very
limited because of the extreme  phytotoxic reaction  of flax to any  herbi-
cide, including diallate.  Diallate is,  however, used in preference to
other herbicides.

          Eptam and barban are  the most  common herbicides which can be
substituted for diallate to control wild oats  in flax; however, both of
these chemicals have characteristics  which limit their use on flax.  Sptam
is phytotoxic in flax, and barban  can only be  applied when the wild oats
are at the - two leaf stage  (2-4  days).  If weather is bad for that  period
the effectiveness of the chemical  is  lost.  A cultural method of wild oat
control is delayed seeding; however,  this non-chemical method of• weed
control reduces flax yields by  about  33%.

          Cancelling the use  of diallate on flax is anticipated to result
in annual losses of approximately  $0.4 million to flax producers.   The
economic impacts result from  a  combination of  changes in herbicides costs,
shifts in production from flax  to  alternative  crops, and yield losses
resulting from delayed  seeding. The  average loss in returns per acre of
flax treated with diallate substitutes would be $13.59.  This loss
represents 25% of the expected  returns to land, labor, and management with
diallate available.

          Since only 3% of the  total  flax acreage is estimated to  be
treated with diallate, the cancellation  of diallate would not have a sig-
nificant effect on total flaxseed  supplies or established marketing systems
or patterns.  Moreover, some  shift in production away from flax is already
occurring because of decreasing demand for flaxseed and flax straw.  (USDA,
Crop Production Annual Summary, 1978.)

     C.   Lentils

          Commercial lentil production is located almost entirely in
Washington and Idaho.   During the  1976-1978 period, an average of  114,000
acres of  lentils were planted annually in these two states.  Both states
                                    III-7

-------
are subject to wild oat: infestations, and  dial late is a major herbicide
used to control wild oats in lentil plantings.   It is estimated that 43,320
acres or 33% of the total lentil acreage were  treated with diallate.  With
a normal use rate of one pound  (active  ingredient basis)  of diallate
applied per acre, this would require 43,320  pounds of diallate.

          Use of triallate provides an  excellent substitute for diallate
use on lentils.  Triallate, although increasing the overall costs of wild
oat control, offers increased yields which more than offset the increase in
costs.  The positive economic impact is based on increased yields resulting
from triallate being less phytotoxic and in  some cases better control of
the pest.  Triallate would pick up an estimated 93% of the diallate usage
if diallate were cancelled.  Therefore, cancellation of diallate and a
shift to triallate would have a positive impact to growers by increasing
returns by $512,000 annually, with an increase  in production of 1.6%.

          The growers reluctance to switch to triallate may be 1) unfamil-
iarity with triallate, 2) inconvenience of stocking additional chemicals
when diallate can treat all crops, and  3)  marketing preferences.

          Other herbicides which are expected to offset the impact of
diallate's cancellation on lentils include prophaa and barban.  However,
these two herbicides are not as useful  as  triallate because of altered
efficacy characteristics (e.g. critical timing  for effectiveness and rapid
biodegradation in soil).  Therefore, these herbicides are not considered as
complete alternatives for diallate.

     D.   Peas

          Dry peas (including Austrian  winter peas and wrinkled seed peas)
are produced primarily in Washington and Idaho.   During the 1976-1978
period, an average of 231,000 acres of  dry peas  were produced in these two
states.  Wild oats are a major pest in  dry pea production.   Approximately
11% of the total acreage is treated with diallate for wild oat control,
using 30,375 pounds (active ingredient basis) of  diallate.

          Triallate is considered the primary substitute  for diallate.
Triallate is already used on 45% of the dry peas  produced and offers
growers a larger return on the land through  increased yields (see
discussion in III.C.), so that the total return  to the grower increases  by
nearly $730,000 annually.

          Propham and barban are also registered  for use  in dry peas.
However, these two chemicals have limitations which preclude their total
effectiveness as alternatives.  Upon diallate's  cancellation,  the usage  of
both propham and barban would increase above the  current  2% (propham)  and
6% (barban) usage on dry peas.

          The cancellation of diallate  is  anticipated to  result in U.S.  dry
pea production increases of less than 0.01%.  Therefore no  economic impacts
would occur at either the market or consumer levels.
                                   III-8

-------
     E.   Minor Uses

          Diallate, in addition  to  the  foregoing major  uses,  is also
registered for use on soybeans,  corn, barley,  potatoes,  safflower,  and
alfalfa.  The percentage of use  on  each of  the above crops is quite small;
thus, use of diallate has been determined to  be minor on these crops.   Use
of diallate on these crops is basically limited to North Dakota, Minnesota,
Montana, and Idaho.

          The estimated use of diallate on  potatoes is  0.5% of the  total
potato acreage.  For all other minor  crops  except safflower and alfalfa,
the total treated acreage is thought  to be  0.1% or less.  Data are  not
available to show the total amount  of diallate used on  safflower and
alfalfa acreage.

          If diallate becomes unavailable,- batban would be used to  control
wild oats on soybeans, and eptam would  be the chemical  of choice in corn
and potatoes.  Triallate is registered  for  use on barley.  Monsanto has
expressed a desire to register triallate for all crops  for which diallate
is now registered  (Spurrier,  1979).

          The cancellation of  diallate  on the crops discussed in this
section is not expected to have  any economic impact upon the grower or
aarket prices.  Only  in the case of potatoes does the use approach 0.5% of
the total acreage.  In other  cases, diallate treated acreage amounts to
0.1% or less of the total  acreage.
                                    III-9

-------
IV.  Development of Regulatory Options

     A.   Introduction

        .  The risks and benefits  associated with the uses of diallate have
been identified in Sections  II and III.   As explained in Section I,  FIFRA
mandates that the Agency  achieve  a balance between the competing considera-
tions of risks and benefits.  In  order to carry out that mandate, the
Agency has developed various regulatory  options and has evaluated each
option for its impact on  both sides of the risk/benefit equation.

          This section of Position Document 2/3 briefly summarizes the
salient risks and benefits associated with the uses of diallate and
describes the process by  which  the Agency developed potential courses of
action.

     B.   Salient Risk/Benefit  Considerations

          In performing  a risk/benefit analysis of the uses of diallate,
the  Agency  identified several salient factors, on both sides of  the risk/
benefit equation, which  became  determining considerations in the
development  of  regulatory options.  These considerations are reviewed below.

           1.    Salient Risk Considerations

                As  detailed  in Section II of this  document the  original  risk
criteria  cited  in  the RPAR  notice  as the basis for the Agency's  presumption
against diallate stands unrabutted.  The principal  risk  associated  with the
use  of diallate is  oncogenicity.   This risk manifests  itself in  the general
population  through dietary  exposures at very  low  levels  and to pesticide
applicators through dermal  and inhalation exposures  when applying diallate
as an emulsifiable concentrate.

                It is estimated that there  are approximately 2400 pesticide
 applicators currently at  risk.   This risk  is  estimated to be 10   and is
 of primary concern to the Agency (Table  IJ-10).   The dietary risk to  the
 general population is estimated  to be 10    based  on tolerance  levels
 adjusted to reflect the percent  of crop  treated  (Table 11-11).   The Agency
 considers the dietary risks of diallate  to be low and not of primary
 concern when compared to  the benefits associated  with its use.

                Since the  original RPAR  notice was published the  Agency
 received additional evidence to  support  the conclusion that diallata  is a
 mutagen.   Although quantitative estimates of the mutagenic risk to
 applicators are not possible at  this time,  any risk reduction  procedures
 proposed to reduce the oncogenic risks  of diallate will concomitantly
 reduce mutagenic risks.

                There  is  also evidence  that diallate causes  neurotoxic
 effects.  As in the  case of mutagenic  risks quantitative estimates  of risk
                                      IV-1

-------
are not  presently possible.   However,  based on current'exposure estimates
there is a 600-fold span  between  the observed effect level and the exposure
level.

          2.   Salient Benefit Considerations

               The benefits of diallate  were assessed in terms of the
economic impacts which would  result if its  uses were cancelled and users
were thereby forced to employ available  alternatives.  As detailed in
Section  III, the economic impacts associated with the cancellation of
diallate total just over  $3 million (Table  III-1).

               Sugar Beets

               The total  annual market value of sugar beets is $500
million.  Should diallate become unavailable,  growers are expected to
experience an annual loss of  $4 million.  More than 60%  of the diallate
used in  this country is applied to sugar beets.   Presently the emulsifiable
concentrate formulation is applied to  85% of the treated acreage while the
granular formulation is applied to the remaining 15%. The degree control
of wild  oats provided by  each of  these formulations appears to be the same.

               Several alternative chemicals were identified in Section III
of this  document.  Specifically mentioned were,  1)  cycloate for control in
irrigated areas, and 2) eptam for control in non-irrigated areas,  and 3)
barban which can be used  in both  areas.  However, due to certain limita-
tions (see Section III.A)  none of these  chemicals provide adequate
protection against wild oats.

               In changing to granular diallate alone for control of wild
oats in  sugar beets, the  increased cost to  growers  in the short run will be
approximately $6-$7 per acre-treatment given current prices.

               Flax

               Approximately  3% of the total flax acreage is  treated with
diallate.  If all forms of diallate should  become unavailable for use in
the control of wild oats  in flax, growers are  expected to experience a
$400,000 annual loss.  The emulsifiable concentrate is the only formulation
presently registered for  this use.

               Eptam and  barban are the most commonly used alternatives for
diallate.  Soth of these  chemicals have limitations which reduce their
desirability and effectiveness in the  control  of wild oats (see Section
III.B).  Granular diallate is anticipated to provide  effective control of
wild oats in flax; however, this formulation is  not currently registered
for this use.
                                     IV-2

-------
               Lentils

               Thirty-eight  percent of the total lentil acreage is treated
with diallate.  Lentil production  is basically limited to two western
states, Idaho and Washington.   It  is estimated that more than 43,000 pounds
of diallate are applied to lentil  acreage annually.  The emulsifiable
concentrate is the only diallate  formulation presently registered for this
use.

               The major  alternate chemical used to control wild oats in
lentils is triallate.  Triallate has proven to be an effective alternate
and provides control of wild oats  equal to or greater than that provided by
diallate.  Triallate is also less  phytotoxic to lentils than diallate.
Propham and barban are alternative chemicals but do not provide acceptable
control of wild oats.

               Peas

               Dry peas,  like  lentils, are primarily grown in Idaho and
Washington.  Currently, approximately 11% of the dry pea acreage is treated
with diallate.  Only the  emulsifiable concentrate formulation of diallate
is registered for use on  peas  for  the control of wild oats.

               Triallate-is  the primary alternate chemical for diallate on
peas.  Presently, triallate  is used on 45% of the dry pea acreage.
Triallate is less phytotoxic to peas than diallate.

               Minor Uses (Alfalfa, Barley, Corn, Potatoes, Safflower,
               and Soybeans)

               The total  percent  of minor crop acreage treated annually
with diallate ranges from <0.1% to 0.5%.   More specifically, it is
estimated that 0.5% of the potato  acreage is treated, whereas the
percentage of treated acreage  for  all other crops is 0.1% or less.  No
economic impacts are expected  if  diallate is cancelled.  Only the
emulsifiable concentrate  formulation is registered for the minor uses.

               Barban and eptam are considered as possible substitute
chemicals.  Triallate is  now registered for use on barley.

     C.   Risk/Benefit Analysis

          1.   Dietary Risk/Benefit Analysis

               As indicated  in Section II.D.1.  the dietary risk from
diallate is estimated as  10  . This estimate is based on assuming
residues exist on treated crops at the tolerance level.  This is a worst
case assumption and to date, residues have not been found on any crops  at
the level of detection (.02  ppm).   More than 60% of the dietary risk is
attributable to the use of diallate on sugar beets.  The dietary risk is
                                     IV-3

-------
considered to be low and the benefits of  diallate use are moderate for
sugar beets and flax, and low for lentils,  peas,  and minor crops.  In sugar
beets the average cost increase per acre  will  range from $4.69 to $97.65.
For flax the loss in returns per acre would be about $14.  For all other
uses economic benefits would accrue due to  minor  production increases from
either more effective control of wild oats  or  decreased phytotoxic effects
of alternate chemicals.  Therefore, the Agency concludes that the benefits
(low to moderate) outweigh the dietary risk (low),   in view of this,
regulatory action on the basis of dietary risk alone is. not warranted,  ^s
indicated above the principal risk of diallate is to pesticide applicators
through dermal and inhalation exposure when applying diallate as an
emulsifiable concentrate.

          2.   Applicator Risk/Benefit Analysis

               The applicator risk/benefit  matrix for diallate, expressed
in qualitative terms, is shown in Table IV-1.   For all presently registered
use patterns the applicator risks are high  for the  emulsifiable concentrate
formulations while the risks associated with the  granular formulation are
no greater than the general populations risk from dietary exposure.  The
benefits of diallate use (either formulation)  on  sugar beets and flax are
moderate and all other uses are low.

               Applicators of the emulsifiable concentrate formulation nay
be exposed both derraally and via inhalation as the  result of  splashing,
vaporization, or accidental spills.  Likewise  during application, exposure
may occur both dermally and via inhalation. Table 11-10 identifies the
dermal risk to applicators of the emulsifiable concentrate formulation as
10  .

               One potential risk reduction measure is to require applica-
tors using emulsifiable concentrate diallate to wear protective clothing.
Protective clothing in this instance is defined as  rubber gloves and
coveralls.  These item? would reduce the  exposure level by a factor of 4
(see Section U.C.1.).  ThJs reduction would result in a risk of 10
which the Agency still considers unreasonable  when  compared to the low and
moderate benefits and, therefore, unacceptable.  Therefore, a protective
clothing requirement will not be considered as a  viable risk reduction
measure in this analysis.

               Thus the risk/benefit picture for  the emulsifiable concen-
trate formulation is essentially the same for  all uses.  The applicator
risk is high and the benefits are, at best, moderate.  Since the risks out-
weigh the benefits in every case, the Agency must consider regulatory
options for reducing the risks associated with the  emulsifiable concentrate
formulation, in particular, the high applicator risks.

               From this analysis the Agency concludes that the high risk
involved in the use of the emulsifiable concentrate is unacceptable when
considered against the low to moderate benefits of  all emulsifiable
                                     IV~4

-------
Table IV-1.  Applicator Risks versus  Economic  Benefits  of  Diallate
Uses
Sugar Beets
Flax
Lentils
Dry Peas
Minor Uses
Barley,
Potatoes ,
Saf flower,
Soybeans ,
Corn, Alfalfa
Applicator
Emulsifiable
Concentrate
High (a)
High (a)
High (a)
High (a)
High (a)
Risk
Granular Economic Benefits (c)

Low Moderate
(b) Moderate
(b) Low
(b) Low
(b) Low
                   -4
 (a)  High Risk >_ 10
 (b)  Although not registered for these uses the applicator risks would remain
     low if registered.
 (c)  Benefits analysis did not evaluate the individual benefits of the two
     formulations.
                                         IV-5

-------
concentrate uses.  It also concludes that this risk  can be  reduced to the
acceptably Low dietary level if the emulsifiable  concentrate  formulation is
replaced by the granular formulation.  Therefore, the Agency  will  examine
the feasibility of cancelling all diallate emulsifiable concentrate
formulations in its regulatory options.

               For sugar beets (the only use for  which the  granular formula-
tion is also registered!, the risk/benefit balance for the  granular formula-
tion is shifted favorably.  Risks become low because of significantly
decreased applicator exposure and benefits remain moderate  because the cost
and effectiveness of granular diallate are approximately  the  same  as the
emulsifiable concentrate.  Based on this finding  for sugar  beets,  the
Agency assumes that this more favorable risk/benefit balance  could be
achieved if the granular formulation were registered for  all  of  the other
uses.

     D.   Regulatory Options

          With regard to the emulsifiable concentrate products of  diallate,
three basic regulatory options have been developed for consideration:

          1.   Continue emulsifiable concentrate  registrations.

          2.   Cancel emulsifiable concentrate registrations  immediately.

          3.   Cancel emulsifiable concentrate - registrations  affective in
two years.

               Options 1 and 2 represent an absolute regulatory response.
For Option 1 it means that sale and distribution  of  diallate  emulsifiable
concentrate products are unconditionally continued.  Option 2 on the other
hand means that sale and distribution of these products are prohibited
effective as soon as the decision becomes final.  Option  3  represents a
decision to phase out the emulsifiable concentrate formulations  to permit
time to extend the registrations of granular diallate to  uses where it is
not presently registered.

          1.   Option 1;  Continue Emulsifiable Concentrate Registration

               This option would return emulsifiable concentrate formula-
tions of diallate to the registration process, and they would be retained
as effective means to control wild oats in sugar  beets and  other crops.  3y
adopting Option  1, the Agency would conclude that the benefits associated
with the use of emulsifiable concentrate diallate outweigh  the risks and
that allowing its use would not result in unreasonable adverse effects.
                                                                       -4
               Under this option, the potential applicato-  risks of 10
resulting from inhalation and dermal exposure would  not be  reduced.  There
would be no adverse economic impacts associated with Option 1 because use
of emulsifiable concentrate formulations would continue.  3y  choosing this
                                     1V-6

-------
option the Agency would  conclude  that the benefits would outweigh the risks
of continued use.

          2.   Option  2;  Cancel  Smulsifiable Concentrate Registrations
               Immediately

               This option would  eliminate all uses of emulsifiable
concentrate diallate thirty  days  after the final Agency decision.  By
adopting Option 2 the  Agency would conclude that the risks associated with
the use of emulsifiable  concentrate diallate 'exceed the benefits, and
result in unreasonable adverse  effects.

               Under this option,  applicator risks resulting from
inhalation and dermal  exposure  would be  reduced to the same magnitude of
risk as the general dietary  risk.   Cancelling all emulsifiable concentrate
registrations immediately would imply that" risks outweigh benefits.

               The  economic  impact which result from the immediate
cancellation of diallate emulsifiable concentrate formulations on sugar
beets is estimated  to  be more than $3 million (Table III-1).  The benefit
analysis did not attempt to  identify the impact of the individual formula-
tions.  However, most  of the impact will result from the cancellation of
the emulsifiable concentrate formulations.  This is based on the fact that
the emulsifiable concentrate accounts for approximately 35% of the diallate
presently used on sugar  beets.  Flax growers are expected to suffar a
$400,000 annual  Loss while  growers of lentils and peas are anticipated to
receive an economic gain as  expressed in Table III-1.

               This option  would  ignore such factors as availability of
alternatives including granular diallate since time would have to be
allowed to produce  adequate  quantitites of this material.  It also ignores
the time necessary  to  register granular diallate on flax and other crops
for which the emulsifiable  concentrate is now registered if the markets
demand such action.  Lastly, this option neglects to allow time for growers
to modify or acquire the necessary equipment to apply the granular
formulation.

          3.   Option  3; Cancel  Emulsifiable Concentrate Registrations
               effective Two Years After the Decision Becomes Final

               This option  is essentially the same as Option 2.  It differs
in that it would eliminate  all uses of emulsifiable concentrate diallate
two years after  the decision becomes final.  By adopting Option 3, the
Agency would indicate  its unwillingness to accept the applicator risks
associated  with the use of  the emulsifiable concentrate formulation
indefinitely.  This option  would  lessen the impact of immediate cancel-
lation by  1) allowing  time  to produce the necessary granulir diallate to
control wild oats  in  sugar  beets, 2) allowing time to register granular
diallate on crops where  only the  emulsifiable concentrate is now registered
and where other  registered  alternatives are not desirable, and 3) allowing
time  for growers to make necessary equipment adjustments.
                                     IV-7

-------
               Some granular diallate formulations  are  presently registered
for use on sugar beets only in Minnesota and North  Dakota.   It will  be
necessary to allow sufficient time to extend the  registration of these
granular formulations to the western states.  While the registsrad granular
formulation with diallate as the sole active ingredient is  only registered
for use in the non-irrigated areas, other granular  formulations with
diallate as one of the active ingredients are registered in other geo-
graphical areas (western states).  Two years should also provide ample  time
to register the granular formulation on crops now being treated with the
emuisifiable concentrate formulation if the market  demands  it.

               The Agency, based on the information available, believes 1)
that a two-year time frame is sufficient time for the company to produce
the granular diallate required to maintain control  of wild  oats in sugar
beats, and 2) that two years is ample time for  growers  to make adjustments
in equipment necessary to apply the granular formulation (Lewis, 1980).

               Because there is no acceptable alternative chemical
presently registered for use on flax, cancellation  of the emuisifiable
concentrate is a major concern.  The expected annual economic impact to
flax growers is $400,000.  The Agency believes  that the granular
formulation will provide excellent wild oat control with little or no
phytotoxicity and is willing to consider a proposal for this use.

               Registrants of -the emuisifiable  concentrate  and granular
formulations may submit an application for amended  registration to
1) convert their emuisifiable concentrate formulation to granular, and
2) expand their granular registrations to crops for which only the
emuisifiable concentrate is now registered.  The  review of  these actions
will be expedited and should not require additional data.
                                  IV-8

-------
V.   Proposed Regulatory Decision
         Agency proposes  tol implement Option 3  which is to/cancel the
emulsifiable concentrate  formulations of  diallate effectiveTwo years from
the date the decision becomes  final.  I

     Option 1 is unacceptable  because the level of risk to applicators of
the emulsifiable concentrate formulations remains at 10   which the
Agency is unwilling to permit  considering the low and moderate benefits.

     Option 2 is also unacceptable.   While under this Option the risk to
applicators is reduced to the  general dietary risk level,  it does not take
into account the potential impacts  that may occur due to the unavailability
of adequate supplies of granular  diallate currently registered for sugar
beet use.  In addition it does not  permit any opportunity for the
registrant to extend the  use of granular  diallate to those uses where only
the emulsifiable concentrate is currently registered prior to complete
cancellation.

     Option 3 would overcome the  disadvantages  of both Options 1 and 2.
The Agency would welcome  applications to  extend the use of granular
diallate as a replacement for  the emulsifiable  concentrate and EPA
considers a two-year time frame appropriate to  accomplish this.  At the
same time, the two-year time frame  permits the  production of adequate
supplies of granular diallate  to  meet expected  market demand.

     Also under Option 3  registrants  would be allowed to convert
emulsifiable concentrate  formulations to  granular and expand granular
registrations to crops for which  only the emulsifiable concentrate is
presently registered.
                                   V-l

-------
VI.   References

Albert,  R.E.,  1979a.   Diallate RPAR Rebuttal Response.   Carcinogen
      Assessment  Group,  EPA,  1979.

Albert,  R.E.,  1979b.   Risk Assessment of Avadex.   Carcinogen Assessment
      Group,  EPA,  1979.

Brusick,  D.J.  1977a.   Mutagenicity Evaluation of  CP15336,  BIO-77-166  (Final
      Report  July,  1977)  submitted to Monsanto,  Co,  Inc., by Litton
      Bionetics, Inc.,  Project No.  2683.  9pp.

Brusick,  D.J.  1977b.   Mutagenicity evaluation of  CP15336 in the mouse
      Lymphoma  assay, BIO-77-171  (Final Report,  August,  1977)  submitted to
      Monsanto  Co,  Inc.,  by Litton Bionetics, Inc.,  Project No. 2584.
      Submitted to  EPA  March 22,  1973.  EPA Accession No. 233353.

Carcinogenesis Bioassay  Data  Systems,  1975.   Preliminary Analysis Report
      for  Avadex.   Prepared for the National Cancer Institute by EG&G/Mason
      Research  Institute.

Development  Planning and Research Associates,  Inc.,  March,  1979.  Economic
      Analysis  of Effects of Restricting  Use of  Diallate on Sugarbeets,
      Flax, Lentils and Dry Peas;  to Environmental Protection Agency.

Gardner,  R., 1980.  Estimates  of  human exposure to diallate.  Memo  to James
      Wilson, SPRD  (TS-791), Project Manager.

Innes, J.R.M.,  B.M. Ulland, M.G. Valeno,  L.  Petrucelli, L.  Fishbein,
      E.R. Hart, A.J. Polxotta, R.R.  Bates,  H.L. Falk, J.J.  Gart, M. Klein,
      I.  Mitchell and J. Peters.  1969.  Bioassay of Pesticides and
      Industrial Chemicals  for  Tumorigenicity in Mice:   A Preliminary Note.
     J.  Nat. Cancer Inst.  42:1101.

Jensen,  J. 1980.   Applicator Exposure  to  Diallate While Using the Granular
     Formulation.   Memo to J. E. Wilson,  Jr., Project Manager for Diallate,
     April 23,  1980.

Keplinger, M.L. 1974.  Mutagenic Study with  Avadex  Technical in Albino
      Mice.  3TL-74-13.   Industrial Bio-Test  Laboratories,  Inc., August 20,
      1974.  1ST No. 622-05252.  Submitted to EPA  by Monsanto Company on
      October 14,  1976.  14 pp.

Keplinger, M.L. 1976a.   Two-year Chronic  Oral Toxicity  Study with Avadex
      Technical in Albino Rats BTL-74-12.  Industrial Bio-Test
      Laboratories, Inc.,  December 17,  1976.   1ST  No.  622-05250.  Submitted
      to EPA by Monsanto Company on February  4,  1977.  EPA  Accession No.
      228094.

Keplinger, M.L. 1976b.   Neurotoxicity  Study  with  Avadex in  Chickens,
      BTL-76-84.  Industrial Bio-Test Laboratories,  Inc., December 28,
      1976. %IBT No. 8580-09119.  Submitted to EPA by Monsanto Company on
      February  16,  1977.  EPA Accession No. 226094.
                                   VI-1

-------
Lewis, C.,  1980.  Additional information on use of granular diallate.
    Memo to James Wilson, SPRD Project Manager.

Mauer, I.,  1978.  Ph.D., Geneticist, Toxicology Branch, EPA, November 28,
     1978.   Diallate:  Preliminary Assessment of Mutagenicity, Regulatory
     Options, and Alternatives, Memorandum to Roger Gardner, Toxicology,
     EPA.

Monsanto Agricultural Products Co., 1977.  Rebuttal to the Rebuttable
     Presumption Against Registration and Continued Registration of
     Pesticide Products Containing Diallate.  £1A  [30000/15].

Pesticide Residues Committee, 1965.  NAS-NRC Report on *"No Residue" and
     "Zero Tolerance," June 1965.

Phillips, B.M. 1977.  Final Report to Monsanto Company 42-day Neurotoxicity
     Study with Diallate in Chickens. . (*1B(30000/15) attachment A)

Rinkus, S.J. and M.S. Legator, 1977.  Mutagenicity of Diallate.  Submitted
     to EPA (FRS/TSD).  July 6, 1977. 5 pp.

Sandhu, S.S., Ph.D., Biochemistry Branch, EPA, August 30, 1978.  Evaluation
     of the Data for the Mutagenicity of Diallate and Tr.iallate.
     Memorandum to William F. Durham, Ph.D., Director, Environmental
     Toxicology Division (HERL/RTP), EPA.

Selim, S. October 16, 1978.  Exposure Analysis for Diallate.  Proprietary
     Environmental Fare Branch, Hazard Evaluation Division, OPP, EPA.

Sikka, J.C. and P. Florczyk. 1978.  Mutagenic Activity of Thiocarbamate
     Pesticides in a Microfaial Test System.  J. Aqr. Food Chem. 26:146-.

Simmon, V.F., E.S. Riccio, and A. Griffih. 1978.  In vitro Microbiological
     Mutagenicity Assays of Diallate and Triallate (Final Report, April,
     1978).  Stanford Research Institute International, Project LSU-3493.
     Prepared for EPA (HERL/RTP) under Contract NO. 68-01-2458. 23 pp.

Sochard, M.R.  1980.  Editing of Diallate PD 2/3 Pursuant to Memo From J.
     Jensen, EFB, to J. E. Wilson, Jr., Project Manager for Diallate/
     May 14, 1980.

Spurrier E., 1979.  Memo of telephone conversation with R. Troast.

Ulland, 3., S.K. Weisburger, and J.H. Weisburger. 1973.  Chronic Toxicity
     and Carcinogenicity of Industrial Chemicals and Pesticides.
     Toxicol. Appl. Pharmacol. 25:446.

USDA/State Assessment Team on Diallate, 1977.  Assessment of the Meed for
     Diallate in Agriculture *10A[30000/15].

U.S. Department of Health Education and Welfare. 1969.  Report of the
     Secretary's Commission on Pesticides and Their Relationship to
     Environmental Health. U.S. Government Printing Office, Washington,
     D.C.

                              71-2

-------