ETHYLENE DIBROMIDE  (EDB)




        POSITION DOCUMENT 4
    OFFICE OF PESTICIDE PROGRAMS




U.S. ENVIRONMENTAL PROTECTION AGENCY




         September 27, 1983

-------
                    EXECUTIVE SUMMARY







This Position Document (-PD')-4- presents the Agency's final determi-



nation regarding the Rebuttable Presumption Against Registration



(RPAR) of the pesticidal uses of ethylene dibromide (BOB).



The RPAR was initiated by EPA in December, 1977, based on a



National Cancer Institute (NCI) study demonstrating the potential



of EDB to induce cancer in rats and mice.  The basis for the



RPAR was the presumed risk of cancer to man as a result of



pesticidal exposure to EDB.  The RPAR was further supported by



additional evidence submitted to the Agency showing EDB to



be mutagenic and capable of producing adverse reproductive



effects.





In December 1980, the Agency issued a Position Document 2/3



and Notice of Preliminary Determination Concluding the RPAR



which proposed several regulatory actions to reduce the



human health risks resulting from pesticidal use of EDB.



The proposed actions were based on the Agency's risk/benefit



analysis of each registered use of EDB.





EDB is registered as a fumigant for preplant application to



soil, post-harvest commodity fumigation, spot fumigation of



grain milling machinery and several other low-volume uses.



The major pesticidal use of EDB is preplant application to



soil, which comprises about 90% of domestic use.

-------
                              ii
This final determination regarding EDB is based on an evaluation
of the comments submitted in response to PD 2/3 by the EPA
Scientific Advisory Panel, the U.S. Department of Agriculture,
industry and the general public.  In reaching a final
determination, the Agency has reevaluated the health risks
and the economic benefits of EDB for the regulatory actions
proposed in PD 2/3, and has carried out a thorough scientific
and technical review of all comments received.

The Agency has now adequately developed evidence that the
preplant soil fumigant uses of BOB are likely to result in
leaching to groundwater, and, therefore, in the contamination
of human drinking water supplies.  The potential contribution
of continued soil fumigant use of EDB to the contamination
of drinking water poses unacceptable risks and results in an
imminent hazard.  Consequently, the Agency has determined
that it is necessary to immediately discontinue this use of
EDB through an emergency suspension order.

The uses of EDB in spot fumigation of grain mills and for
fumigation of stored grain pose a significant dietary cancer
risk to the general public and high risks to applicators and
millworkers.  The Agency has determined to cancel these uses.
The Agency is extremely concerned about the potential exposure
of the public from these uses of EDB and is gathering additional
information concerning these exposures.  When this additional
information has been evaluated, the Agency will consider the
need for emergency suspension of these uses.

-------
                             iai



The Agency has determined that the use of EDB as a quarantine




fumigant to ensure fruit fly disinfestat ion of food commodities




shipped out of quarantine areas poses substantial risks to




both occupationally exposed workers and to the general public.




Further, the Agency has concluded that the human health




risks of continued use of EDB as a quarantine fumigant exceed




its benefits and that this use must be cancelled.  At the




present time the Agency recognizes that there are no chemical




alternatives for this use and that other alternative technology,




including cold storage treatment and irradiation, are not




immediately available.  The effective date of this cancellation,




therefore, has been delayed until September 1, 1984 to allow




USDA/APHIS and the industry adequate time to change to alterna-




tive means of quarantine fruit fly disinfestation.






The Agency has determined that the use of EDB as a felled log




fumigant to control bark beetles poses high cancer risks to




applicators and provides little economic benefits.  The




Agency, therefore, has determined that this use should be




cancelled.






For the remaining uses:  subterranean and drywood termite




control, fumigation of beehive supers, vault fumigation of




stored clothing and furniture, and the USDA Japanese Beetle




Quarantine Program, the Agency has decided that the final



determination regarding these low-volume uses will remain




unchanged from the proposed action.  The Agency believes




there may be some exposure to applicators as a result of




these minor uses; however no data are available which




demonstrate that these exposures are significant.

-------
                            iv




These uses will be allowed to continue if certain use restric-




tions are implemented and additional data requirements are




fulfilled.  For control of termites, the current label directions




will be reinforced and expanded by allowing use only by




certified applicators.  For the other uses, protective clothing




and other low-cost measures, designed to reduce exposures to




EDB, will be required.  In addition, registrants of products




with each of the low-volume uses will be required to conduct




studies to determine the levels of EDB in the air to which




applicators are exposed, the number of applicators involved




in these uses, and the amounts of EDB applied annually.

-------
                  ACKNOWLEDGEMENTS






EPA ETHYLENE DIBROMIDE TEAM




Richard Johnson, Team Leader, RD




Anne Barton, Statistician, HED




Gary Burin, Toxicologist,  HED




Stuart Cohen, Chemist, HED




Charalinggayya Hiremath, Toxicologist, CAG




James Holder, Toxicologist, HED




Roger Holtorf, Economist,  BUD




George Ludvik, Senior Entomologist, BUD




Robert McLaughlin, Attorney, OGC




Marcia Mulkey, OGC




Neil Pelletier, Plant Pathologist, BUD




Joseph Reinert, Chemist, HED




Amy Rispin, Chemist, HED




Esther Saito, Chemist, HED



Carol Sakai, Reproductive  Biologist, REAG




Richard Schmidt, Chemist,  HED



Todd Thorslund, Biostatistician, CAG




Linda Vlier, Economist, BUD




John Worthington, Chemist, HED






OTHER ACKNOWLEDGEMENTS






Arvella Farmer, Secretary, RD




Barbara Moore, Secretary,  RD

-------
                       TABLE Of CONTENTS
Executive Summary




I.   Introduction	1




II.  Analysis of EDB 2/3 Comments on Exposure and Risks	6




    A.  Comments Relating to Exposure	6




       1.  Appl icator Exposure	7




              a.  Preplant Soil Fumigation	7



              b.  Spot Fumigation of Grain Milling Machinery	9




              c.  Quarantine Fumigation	11




              d.  Felled Log Fumigation	13




       2 .  Dietary Exposure	14




              a.  Preplant Soil Fumigation	14




              b.  Spot Fumigation of Milling Machinery	16




              c.  Quarantine Fumigation	17




       3 .   Other Exposure	18



    B.  Comments Relating to Hazard	19




       1 .  Oncogenicity	19




       2 .  Muhagenicity	27



       3.  Reproductive Effects	28




              a.  Ter Haar, 1981	28




              b.  Levine et al., 1981	30




              c.  Conclusion	32

-------
                                viii




III.   Revision and Amendments to PD 2/3,  Exposure and Associated




      Health Risk	33




    A.  Applicator Exposure	33




       1.   Preplant Soil Fumigation	33




       2.   Quarantine Fumigation	36




       3.   Felled Log Fumigation	43




    B.  Applicator Risk	44




    C.  Dietary Exposure from Quarantine  Fumigation	47




    D.  Contamination of wheat By-Products from Bulk Grain




        and Spot Treatments of Grain Milling Machinery by EDB	55




    E.  Groundwater Contamination 	59




    F.  Dietary Cancer Risk Equation 	69




    G.  Dietary Cancer Risk 	70



    H.  Groundwater Contamination Cancer  Risk 	79




IV.  Analysis of PD 2/3 Comments and Revisions Relating to Benefits.80




       1.   Preplant Soil Fumigation	80




       2.   Stored Grain Fumigation	86




              a.  Rates and Frequency of  Application	86




              b.  Volume used to Fumigate Grain	87




              c.  Conclusions for Grain Fumigation	88




       3.   Spot Fumigation of Grain Milling Machinery	88




              a.  Efficacy of Alternatives	89




              b.  Frequency of Application	90




              c.  Usage Estimates	90




              d.  Treatment Costs	93




              e.  Conclusions on Spot Fumigation	95




       4.   Quarantine Fumigation of Citrus, Tropical Fruits,  and




           Miscellaneous vegetables	96




       5.   Felled Log Fumigation	100




       6.   Minor Uses	101

-------
                                  IX




V.  Summary and Conclusions	102






   References and Citations




   Additions to References and Citations - Groundwater




                                         - Economics




                                         - Exposure









   Appendix I:   EDB Position Document 2/3




   Appendix II:  Comments Received in Response to PD 2/3




   Appendix III: SAP and USDA Reviews of EDB-RPAR

-------
                         List of Tables








 1.    Data on EDB Air Levels During Transfer and Application	34




 2.    EDB Respiratory Exposure During Soil Fumigation	36




 3.    EDB Inhalation Exposure of Workers at Citrus Fumigation




      Stations at a Citrus Warehouse	38




 4.    Summary Table of PD 4 Applicator Exposure Estimates Which




      Have Been Modified from PD 2/3	41




 5.    Estimated EDB Exposure to Felled Log Applicators	43




 6.    Revised Cancer Risk Due to Occupational Inhalation of EDB	46




 7.    Residue Levels in Edible Portion of Fruit (ppm)	49




 8.    Dietary Burden in California	52




 9.    Dietary Burden Comparison:  Citrus and Tropical Fruits	53




10.    EDB Groundwater Contamination	60




11.    Comparison of the PD 2/3 and PD 4 Dietary Burdens and




      Cancer Risks	72




12.    Dietary Burden - Ages Specific Estimates	75



13.    Incremental Dietary Cancer Risks for One Year Additional




      Exposure	78




14.    Cancer Risk for EDB Groundwater Contamination	79




15.    Annual Economic Impact of EDB Soil Fumigation Use




      Cancellation	82




16.    Summary of Final Regulatory Decisions on EDB	103

-------
I.    Introduction






Under the Federal Insecticide, Fungicide, and Rodenticide Act,




as amended (FIFRA) (7 U.S.C. Section 136 et seq.)  the Environmental




Protection Agency (EPA or the Agency) regulates pesticide




products.  Section 12(a)(l)(A) of FIFRA requires that all




pesticide products, unless exempted from regulation, must be




registered before they may be sold or distributed.   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 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."






Unreasonable adverse effects on the environment are defined




in FIFRA Section 2(bb) to mean "any unreasonable risk to man




or the environment, taking into account the economic, social




and environmental costs and benefits of the use of any pesticide."




In sum, any decision to register or continue to register a



pesticide must weigh both the risks and the benefits from




the use of the pesticide.






The Rebuttable Presumption Against Registration (RPAR)




process was designed to collect and evaluate benefits and





                                   1

-------
risk information regarding those pesticides which are suspected




of causing adverse health or environmental effects, and




facilitate regulatory decisions concerning them while allowing




interested groups to participate.   The regulation governing the




RPAR process is 40 CFR 162.11.






On December 14, 1977, the Agency issued a Notice of RPAR and




the RPAR Position Document 1 on EDB in the Federal Register




(42 F.R. 63134-63161).  The PD 1 explained the background and




provided supporting data for the presumption of hazard cited




in the RPAR Notice.  The presumption against the registration




of EDB was based on oncogenic, mutagenic and reproductive




effects in test animals.  The PD 1 provided registrants and




other interested parties the opportunity to submit rebuttal




comments.






The presumption against EDB could have been rebutted by




showing:  1) that the Agency's initial determination of hazard




was in error; or 2) by demonstrating that existing use




patterns and the resulting exposure to EDB did not induce




any significant adverse effects.  (See 40 CFR 162.11 (a)(4))




Respondents were also asked to submit evidence of the social,




economic, and environmental benefits from the uses of the




pesticide.  See 40 CFR 162.11 (a)(5)(iii).

-------
On December 10, 1980, the Agency issued "Pesticide Products




Containing Ethylene Dibromide (EDB); Preliminary Notice of




Determination Concluding the Rebuttable Presumption Against




Registration; Availability of Position Document 2/3."




45 FR:  81516-81524).  In the PD 2/3, the Agency concluded that




the presumptions of oncogenicity, mutagenicity, and reproductive




effects were not rebutted.   The PD 2/3 also presented the Agency's




evaluation of rebuttal comments received in responnse to the




original RPAR Notice.   It provided the Agency's analysis of




the risks and benefits associated with the use of EDB and it




presented the regulatory options that were considered for




purposes of risk reduction.  Finally, the PD 2/3 contained the




Agency's proposed regulatory approach to conclude the RPAR.  The




PD 2/3 is summarized below and included in its entirety as




Appendix I.




Summary of PD 2/3




    The Position Document (PD) 2/3 proposed several regulatory




actions to reduce the human health risks resulting from registered




pesticidal uses of ethylene dibromide (EDB).  These proposed



actions were based on the Agency's determination that certain




uses of EDB resulted in unreasonable adverse effects to man.




All of the registered uses of EDB were examined in this




document.  These include:  preplant fumigation of soil (to




control nematodes); fumigation of stored grains (to control




insects); spot fumigation of grain milling machinery (to control




insects); post-harvest fumigation of citrus, tropical fruits,




and vegetables under USDA Animal, Plant and Health Inspection




Service (APHIS) guarantine (to control fruit flies);  fumigation
  418-574 0-83-2

-------
of felled logs to control bark beetles); fumigation of stored



beehive supers and honeycombs (to control the greater wax moth)



vault-fumigation of stored furniture and clothing (to control



various pests); treatment of plants, soil and grass sod in



accordance with the USDA/APHIS Japanese Beetle Quarantine



Program; and fumigation beneath concrete slab foundations of



buildings and porches to control subterranean termites, and



spot fumigation of their galleries to control drywood termites.








    The first regulatory action the Agency proposed was to



allow registration to continue for preplant soil fumigation,



the largest volume use of EDB.  However, the Agency would permit



this use only if certain restrictions were implemented and



additional data requirements were fulfilled.  Specifically, the



Agency would require applicators to wear respirators and



protective clothing; reentry into a treated field would also be



prohibited for 24 hours following fumigation.  The Agency would



have required soil fumigant registrants to conduct studies to



determine if this use results in groundwater contamination.



When results from these studies were available, the Agency would



reexamine the adequacy of its regulatory measures for this use.






    Second, the Agency proposed to cancel the EDB registrations



for two other major uses; fumigation of stored grains and spot



fumigation of grain milling machinery. The bases for these



proposed cancellations was that the public health risks of



cancer, heritable genetic damage, and reproductive disorders



outweighed the economic benefits of these uses.  The Agency

-------
                                   5




concluded that residues of EDB contaminate the human diet as a




result of these registered uses.  Applicators, persons near a




treatment site and workers who handle treated commodities are




also exposed to significant levels of EDB via inhalation and




dermal contact.




    Third, the Agency proposed that on July 1, 1983, EDB regis-




trations for post-harvest fumigation of citrus, tropical fruits




and vegetables would be cancelled.  The interim period would




provide time for the development and implementation of




efficacious alternatives.




    Fourth, the Agency proposed to cancel the EDB registration




for fumigation of felled logs, a relatively low volume use.




The basis for this action is the Agency's determination that




the potential risks to applicators outweigh the economic bene-




fits of this use.



    Finally, for the remaining uses:  fumigation of beehive




supers, vault fumigation of stored clothing and furniture, and




the USDA's Japanese Beetle Quarantine Program, the Agency




proposed to allow these uses to continue only if certain use




restrictions were implemented and additional data requirements




were fulfilled.  Specifically, the Agency would require




protective clothing and other low-cost measures designed to




reduce applicator exposures to EDB.  In addition, the Agency




would require registrants of products with each of these uses




to conduct studies to determine the EDB air levels to which




applicators are exposed, the number of applicators involved in




these uses, and the amounts of EDB applied annually for these




uses.

-------
                                   6

1 *•   Analysis of PD 2/3 Comments on Exposure and Risk


The Agency has revised some aspects of its assessment of risks,

benefits and regulatory requirements in accordance with additional

information provided by the public, United States Department

of Agriculture (USDA), Science Advisory Panel (SAP) and other

sources.  Except as discussed below and in Section III, other

aspects of the analysis presented in the PD 2/3 are unchanged.


    A.   Comments Relating to Exposure


The Agency has received comments relating to exposure to EDB

as follows:


 Use pattern                      Cgmmenter

Preplant Soil Fumigation          (#24) Great Lakes Chemical Corp.
                                  (#34) Dow Chemical Co.

Spot Fumigation Grain Milling     (# 2) Millers' National Federation
    Machinery                     (#22) Ferguson Fumigants,  Inc.

Fumigation of Stored Grain        (#2) Millers'  National Federation

Quarantine Fumigation             (#23) Hawaii Papaya Industry
                                        Assoc.

                                  (#25 and #25A) Florida Dept.  of
                                       Citrus

Felled Log Fumigation             (#15 and #15A) Colorado State
                                                 Forest Service

-------
         1.    Applicator Exposure






              a.   Preplant Soil Fumigation






Dow Chemical Co., has submitted new exposure studies for this




use of EDB (Skoronski, 1979, and Martin, 1980).  These studies,




along with some clarifying information provided by Dow (Martin,




1982), have  been evaluated by the Agency.  The air levels of EDB




reported in  these studies are substantially higher than those




used by the  Agency for the PD 2/3 applicator exposure assessment.




The PD 4 applicator exposure assessment for this use has been




modified accordingly (see Section III).






Great Lakes  Chemical Co. (Comment #24) stated that it is



unrealistic  to assume that 100% of inhaled EDB is absorbed by




the lung; however, Great Lakes did not suggest or provide a




basis for choosing an alternative.  Great Lakes stated that a




respiratory  rate of 0.6 m^/hr would be a more appropriate




value to use than the 1.2 m^/hr value for moderate activity




used in PD 2/3, and that the NIOSH EDB Criteria Document (NIOSH,



1977) used this 0.6 m3/hr figure.  The document did use the




0.6 nvVhr value, but it is not stated what level of physical




activity this breathing rate corresponds.  The Agency consistently



uses a 1.2m3/hr value for males engaged in light to moderate




work (Hayes, 1975).

-------
Great Lakes offered "more realistic" estimates of annual and



lifetime durations of applicator exposures.  The Agency considers



these estimates more reliable (Weidman, 1981) than those used



in PD 2/3 and they have been used for the modified applicator



exposure estimates in this PD 4 (see Section III A).





Great Lakes stated that requiring applicators to wear respirators



is unnecessary since even the highest lifetime exposure calculated



in PD 2/3 (Table 16, p. 48) is less than the NIOSH recommended



standard of 130 ppb (NIOSH, 1977).  Great Lakes suggested that



the Agency consider as an alternative regulatory option that



respirators be required only "... when the product is exposed



to the environment during small spills, repairs, calibration,



transfers or sampling." The highest exposure in Table 16 (532.4



mg/yr for pineapple soil fumigation application) corresponds



to a lifetime exposure of about 300 mg/kg. This compares to



the NIOSH figure of 686 mg/kg which is stated to be "substantially



below that total dose known to induce adverse effects in



experimental animals."  This dose level was published prior to



findings of the cancer inducing potential of EDB.  The Agency



rejects the use of this exposure level in regulating a cancer



causing chemical, such as EDB.  It  is Agency policy that,



lacking valid scientific evidence supporting a threshold (below



which no risk of cancer exists), any exposures other than zero,



must be presumed to be attended by  a risk of cancer in humans.



The Agency has determined that the  risk posed to applicators,

-------
as well as the risk posed by contamination of ground water




aquifers (as discussed in Section III E and H of this document)




from this use of EDB, require emergency suspension of registra-




tions of this use.  The revised exposure assessment for this



use is discussed in detail in Section III of this document.






        b.  Spot Fumigation of Grain Milling Machinery






The Miller's National Federation (MNF) (Comment #2) commented




that the Agency's calculation of applicator exposure during




the spot fumigation of milling machinery did not consider the




case in which respirators are worn properly, or take into




account the rotation of applicators .  EPA, however, evaluated




this situation in the PD 2/3 (see case I, table 18, p.54 of PD




2/3).  The text describing this table (p. 53 of PD 2/3, second




paragraph)  contains a transcription error; this sentence should




read:  "Alternatively, with proper use of the respirator and



rotation of applicators,  an inhalation exposure of 4.3  - 12.5




mg/kg/yr is estimated."






MNF stated that the Agency incorrectly used an air level value




of 3.8 ppm (the limit of  detection) as a persistence level




for a full 8-hour day during normal operations, even though




air sampling was discontinued when the limit of detection




was reached.  The Agency  agrees that this approach leads to




a bias toward higher exposure,  but in the absence of more

-------
                                  10




precise data, this approach is considered to be a prudent and



reasonable approach.






MNF stated that the exposure ranges for applicators and workers



(p. 65 in PD 2/3) overlap, and that this overlap represents



"a highly improbable  situation".  MNF, however, did not point



out which assumptions or calculations are incorrect or



could lead to this "improbable situation."  The use of pro-



tective equipment by  applicators and not millworkers, and



the significantly longer exposure of millworkers provides



explanation for this  result.






MNF urged the Agency  to consider requiring respirators only



during the high exposure times (opening or opening and start-up)



to reduce risk.  The  Agency fully considered the use of



protective equipment  in this use, but concluded any realistic



assessment of the effectiveness of such a restriction would



still leave significant applicator exposure.  This exposure



combined with the dietary exposure to the general public



compelled the conclusion that this use presents unreasonable



adverse effects.





Ferguson Fumigants, Inc. (Comment #22) reemphasized their



claim that applicator exposure is "nil" since an essentially



closed system is employed.  No new data were provided to



support this comment, and the available information, instead,



demonstrates significant exposure potential.

-------
                               11




              c.  Quarantine Fumigation






Florida Dept. of Citrus (FDC) (Comments #25 and #25A)



challenged the data on levels of EDB used to calculate



exposure to fork-lift truck operators at port warehouses.



The Agency acknowledges that the data used were limited.



However, the Agency believes that the difference in the levels



reported is probably the result of the length of time the



fruit has been in storage in the warehouse, and continues to



believe the levels used in the PD 2/3 were appropriate.





FDC claims that the Agency based its calculations of average



inhalation exposure to warehouse workers on two exaggerated



assumptions.  First, that workers were exposed for 155 days a



year.  The Agency agrees with FDC that fumigation stations do



not operate 155 days per year.  The Agency, however, calculated



the applicator exposure at the fumigation station in the  PD



2/3 at 112 days per year (the average number of operating days



at Wahneta during the 1978-1980 seasons).  Because bleed-off



from the cartons and fruit continues beyond the day of treatment,

-------
                                  12




the Agency believes that the exposure period of 155 days during



the period from October through May is not an exaggerated



assumption for warehouse workers who handle the fruit after



treatment.





Second, FDC states that work at the citrus fumigation stations and



port warehouses can hardly be classified as "heavy activity"



because such work does not involve much exertion or heavy



physical activity.  The Agency has now used a breathing rate of



1.2 m^/hr, the applicable rate for males engaged in "light



to moderate activity" in calculating exposure (see Section



III), instead of 1.8 m3/hr as used in the PD 2/3.





The Hawaii Papaya Industry Association (Comment #23) submitted



two studies of air levels of EDB in papaya fumigation plants.  The



NIOSH study submitted has been reviewed, but cannot be used as



the basis for a worker exposure assessment, because of significant



deficiencies, i.e., inadeguate method of analysis, sample size,



and protocol in the design of the study (Day, 1981).  The



second study submitted by the Hawaii Papaya Industry Association



(Hertlein and Hagadone, 1981) has been reviewed by the Agency



and can be used for the calculation of exposure to workers at



papaya fumigation sites.  This study is the basis for the



exposure analysis presented in Section III.





The Agency has also received a summary of a monitoring study



conducted by NIOSH at Hilo, Hawaii.  The values in this study



are slightly lower than those found by Hertlein and Hagadone.

-------
                                  13




The reduction appears to have occurred because the Papaya industry



now employs charcoal filtration in refrigeration rooms and



fumigation chambers.  However, these results are within an



order of magnitude of the Hertlein and Hagadone results and



would not result in a substantial change in exposure estimates.





              d.   Felled Log Fumigation






In PD 2/3, the Agency stated that no data were available on



applicator exposures during felled log fumigation.  In response



to the PD 2/3, the Colorado State Forest Service (CSFS) (Comment



#15A) submitted two exposure studies reported in a 1973 USDA/Forest



Service Document (Undi,  1973).  These studies have been accepted



and are the basis for the exposure assessment presented in



Section III.






CSFS stated that the use concentration of EDB is 4.5%, not 23%



as stated in PD 2/3.  The Agency has confirmed that the typical



concentration of EDB used against the mountain pine beetle



(the main target pest of the forestry use of EDB)  is a 5:1



volume/volume dilution of a 23% formulation.  This corresponds



to a 4.76% weight/weight use concentration.  This  figure has



been used for the exposure assessment in Section III.

-------
                                  14




         2.    Dietary Exposure






              a.    Preplant Soil Fumigation






Great Lakes  Chemical Corp. (Comment #24) stated that the




analytical method in a California Department of Food and Agri-




culture (CDFA) study (Maddy et al., 1979) has a gualitative




limit of detection (the minimum concentration which produces a




signal which can be unambiguously recognized above background)




of 0.1 ppb.  Great Lakes contended that this value is the proper




one to use in the dietary exposure calculations, rather than




the quantitative detection limit (the lowest signal which can




be used to reproducibly quantify the amount of chemical present)




of 1 ppb, which was used  in PD 2/3.  However, scientists familiar




with the CDFA method (Communications, J.C. Reinert, T. Jackson,



D. Rains) indicate that for some crops analyzed, the qualitative




limit of detection (for a 4:1 signal: noise ratio) is well




above 0.1 ppb.






Thus, the Agency believes that recalculating the dietary exposure




for preplant soil fumigation is unwarranted.  This is because




choice of a  proper residue level is uncertain and, furthermore,




any change would result in a change of less than an order of




magnitude in dietary exposure.  As noted by Great Lakes, the




estimated dietary exposure is hypothetical, and any actual




residues of EDB in crops, would probably be at concentrations

-------
                                  15




below this level.  The Agency estimation of dietary exposure




from preplant soil fumigation should thus be considered a worst




case projection.




Very recently, CDFA reported finding detectable residues of




EDB in carrots grown in EDB-fumigated soil (personal communi-




cation, R. Johnson and K. Maddy, CDFA, September, 1983).




These results have not yet been evaluated by the Agency.






Dow Chemical Co. (Comment #34)  has submitted a series of




published papers and a theoretical treatise on the fate of EDB




in soil in order to preclude the need to conduct field dissipation




and leaching studies.  Dow's arguments, while well presented,




are not persuaive regarding potential contamination of ground-




water as a result of this EDB use.  Indeed, recently developed




groundwater monitoring and tests of soil movement of EDB demonstrate




that continued use of EDB as a soil fumigant can be expected to




contaminate groundwater sources.

-------
                                  16




              b.   Spot Fumigation of Milling Machinery





The Millers' National Federation (MNF) (Comment #2) stated that



the Agency's PD 2/3 estimate of dietary exposure was based on



incomplete and contradictory data which was derived from samples



without defined histories.  MNF argued that the Agency could



not reach a conclusion regarding the origin of EDB residues



(whether they come from the spot fumigation use or bulk grain



fumigation).





Ferguson Fumigants, Inc. (Comment 122) suggests that the



dietary exposure to EDB which results from the spot machinery



use is insignificant when compared to total dietary burden.



Although the Agency agrees that it is currently unable to



calculate the precise proportion of the total dietary burden



from grain products attributable to spot fumigation, it does



have data showing that this use may contribute significantly



to this dietary burden (see discussion in III D).





The SAP expressed its great concern over the potential human



dietary exposure to EDB residues in finished bakery products



as a result of the spot fumigation of milling machinery and



stored grain fumigation, and recommended that, because of the



extremely large population potentially at risk, these uses



should be cancelled until such time as convincing  evidence



exists that these uses present no unreasonable risk to consumers.

-------
                                  17




                   c.   Quarantine Fumigation






The Florida Department of Citrus (FDC) (Comments #25 and 25A)



stated that the fresh citrus food factor used in PD 2/3 (3.8%)




was too high.  Based on information contained in a 1979 report




from the Florida Crop and Livestock Reporting Service, FDC




calculated a fresh citrus food factor of 2.2%.  The Agency




agrees that the 3.8% value used in PD 2/3 was too high.  The




Agency has recalculated this value for fresh citrus only




(Reinert, 1982b) and obtained a value of 2.8%.  This value is




close to that proposed by FDC and has been used in the PD 4




dietary exposure assessment (see Section III).






FDC claimed that the EDB residue level in citrus used in the




PD 2/3 was too high.  It was estimated from a derived residue



loss equation and used for the dietary exposure assessment.




FDC presented actual analyses of EDB residues in the edible



portion of citrus at various times after fumigation (King et




al. , 1980).  The results for the edible portion of grapefruits




(avg. residue level = 0.027 ppm) and oranges (0.029 ppm) seven




days after fumigation are in good agreement with the most




recent data of Maddy (Maddy, 1981a), which have been used in




the PD 4 dietary exposure assessment (see Section III).  Very




recent data submitted by USDA (USDA, September, 1983) are




discussed in Section III.

-------
                              18




          3.      Other Exposure





FDC suggested that, although the potential exposure of



populations in private residences located a short distance



from fumigation stations is low, increasing the exhaust stack



height from 10 to 30 meters could aid in reducing ground level



EDB concentrations by increasing atmospheric dispersion.  The



Agency agrees that EDB exposure to the general population



should be reduced to the lowest level feasible in areas sur-



rounding citrus fumigation stations.

-------
                                   19




B.    Comments Relating to Hazards






         1.    Oncogenicity






The Millers National Federation (MNF) (Comment #2) and the




Occidental Chemical Company (Comment #28) both stated that the




route of administration in the 1977 NCI study, intubation, was




inappropriate because human exposures would be primarily by




inhalation.   The Agency's Carcinogen Assessment Group (CAG)




recognizes the difficulties in determining the human inhalation




dose from a rodent intubation study.  However, two inhalation




studies in rodents have since been completed (Wong et al.,




1979 and NCI, 1980)and the risk estimate for inhalation  is now




based on a rat inhalation study.






The Florida Department of Citrus (FDC) (Comments #25 and #25A)




stated in their comments that results from bioassays cannot




properly be used to calculate risk estimates for humans, since




it has not been demonstrated that humans metabolize a substance




in the same way as the animal species used in the experiment.




The CAG agrees that there is no known evidence which demonstrates




that humans metabolize EDB in the same manner as do experimental



animals. However, there is also no evidence demonstrating that




humans do not metabolize EDB in the same manner as do experimental




animal.  In the absence of information concerning the metabolism




of the compound in humans, it is prudent to assume that animals



are adequate models for predicting human effects.  Because EDB




is a direct-acting alkylating agent and induces carcinomas in




both rats and mice, the assumption that the effects will
 418-574 O - 83 - 3

-------
                               20




be the same for humans is particularly reasonable.






FDC comments that the one-hit model for assessing mutagenic



and carcinogenic potential is unsatisfactory.  The FDC also



stated their opinion that the risk estimates presented in the



PD 2/3 are probably exaggerated since the one-hit, no-threshold



model is the most conservative method of calculation.  FDC



suggests that the threshold concept is more valid and practical.



The dominant scientific view of carcinogenicity is that most



agents that cause cancer also cause irreversible DNA damage.



There is substantial evidence from mutagenicity studies that



the biological response characteristic of mutagenesis is



associated with a linear nonthreshold dose-response relationship,



and this model is appropriate.  This is particularly true at



the lower end of the dose-response curve which represents



typical human exposures.  The linear non-threshold model is



also consistent with the available epidemiological cancer



studies.  (e.g., radiation-induced leukemia, breast and thyroid



cancer, and liver cancer induced by aflatoxin in the diet).






There is also some evidence from animal experiments consistent



with this model (e.g., liver tumors induced in mice by 2-ace-



tylaminofluorene).

-------
                              21




Because it has the best scientific basis of any of the




current mathematical extrapolation models, the linear




nonthreshold model has been adopted as the primary basis




for risk extrapolation to low levels of the dose-response




relationship.  The risk estimates made with this model should




be regarded as conservative, representing the most plausible




upper limit for risk, i.e., the true limit is not likely to




be higher than the estimate, but it could be lower.   Therefore




the Agency agrees that this model may be conservative.  The




risk estimate presented in the PD 2/3, based on this procedure,




gave a plausible upper limit for risk.






The CAG has now adopted a new procedure for estimating the




cancer risks for EDB dietary exposure.  The one-hit model




with "Weibull" timing, which is conceptually similar to, but




operationally more systematic than the one-hit model is now




used to estimate cancer risk for EDB, to adjust for the




short duration of the study (CAG, 1982a).






Other models, such as the multi-stage model, have also been




suggested for use in extrapolating EDB cancer risks (USDA




draft report, 1983).  However, this model, as proposed by




USDA, did not take into account the extremely rapid onset of




tumor or the short duration of the study (Burin, 1983).




Furthermore, the complete model, proposed by USDA, is not



yet available for detailed scientific review.  The Agency




considers the new CAG model to be the best model currently




available for determination of EDB dietary cancer risk.

-------
                                22




FDC also commented that a major fault of the NCI dietary



studies was a failure to select the proper dose of EDB.  The



dose was adjusted during the study and was finally calculated



as a time-weighted average dose.  This resulted in a poor



survival rate.  In response, the CAG considers that the



adjustment of dose during the course of the study makes the



calculation of effective dose uncertain.  However, it does not



negate the positive carcinogenic response to EDB observed in



the studies.  The FDC further commented that the evidence



regarding oncogenicity falls short of establishing a connection



between cancer and EDB residues in food.  They note that intubation



(introducing EDB directly into the stomach of laboratory animals)



is recognized by EPA as an inappropriate methodology for assessing



risk from foods ingested orally.  They also point out that the



scientific community, in making risk assessments from food



additives, has rejected such studies for this reason.






The CAG, however, has compared tumor responses where dibromoch-



loropropane was administered via both gavage and diet, and



found no significant difference.  And EDB and DBCP are:



1) structurally related halogenated aliphatic hydrocarbons,



2) mutagenic, and 3) induce a carcinogenic response in the



stomachs of rats and mice.  The CAG, therefore, regards the use



of an intubation study to assess the carcinogenic risk of EDB



residues in food as appropriate.  In addition, the risk estimates



have been adjusted to reflect the small degree of difference



in effects between dietary and gavage administration seen in



the DBCP studies.

-------
                               23




Dow Chemical Co. (Comment #34) commented that the Agency's




wholesale rejection of the Dow rat inhalation study (PD 2/3,




pp. 12-13) for risk assessment is unfortunate.  Dow added that




this is the only study that examines effects after exposures




are discontinued — a situation analogous to the infrequent and




intermittent exposures experienced by workers.  The Dow Chemical




Co. rat inhalation study (Nitschke, et al.  , 1980) was reviewed




by the CAG.






The Agency rejects these comments.  The CAG concluded that the




study is inadequate to assess carcinogenicity in rats because




1) the duration of exposure was short (a maximum of 13 weeks),




and 2) the period of observation post exposure (88 to 89 days)




was too short to determine whether it demonstrated development




of a carcinogenic response.






The FDC argued that two studies on occupational exposures to



EDB from the Octel Company were valid.  The comment claims




that EPA has disregarded any epidemiology studies that show




no effects.

-------
                                24




The Agency disregarded the Octel studies, however, not because




they showed no effects, but because both embodied serious




flaws as noted in the PD 2/3.






The Ethyl Corporation, (Comment #19) discussed the Octel




mortality survey on 117 persons occupationally exposed to EDB




where employment was known to exceed four years.   Ethyl noted




that the Agency position was that,  since no monitoring




or exposure levels of EDB in the plant are discussed, the entire




study is useless.  Ethyl Corporation asked that the EPA position




on this study be reconsidered,  in view of what Ethyl characterized




as the considerable importance of this study, as evidence of




human exposures to significant levels of EDB over a period of




several years.  since receipt of EDB exposure estimates from




Ethyl Corporation, the Agency has reviewed the validity of the




Octel mortality survey (CAG, 1982d).  In addition to the key




argument against the study's validity, (the lack of monitoring




data on exposure to EDB, discussed in the PD 2/3), other criticisms




apply.  They are as follows:






    o Small study population.  The study population consisted




      of 117 employees.  A group this size is not large enough




      to adequately detect the presence of a cancer risk.




      Age-specific cancer mortality rates for this study population

-------
                              25




  were derived based on only 4 observed deaths in the




  entire group of 117 compared against the person years




  at risk (broken down by age category over the time period




  1940 to 1977).  These rates were then compared to rates




  for similar categories in Southwest England in 1961 and




  1970.  It was calculated that 8.4 cancer deaths could be




  expected to occur in the 117 group members during the




  followup period.  This is an insufficient size of a study




  group, statistically, at least a 2-fold risk of cancer




  generally must be found at a power equal to 0.8, in




  order to detect a statistical difference at probability




  of 0.05.






o Invalid cohort selection.  The author decided to drop 84




  people from the study for whom he had no further information




  and another 29 because incomplete information indicated




  a length of employment of less than 4 years.  Together,




  these two groups represent 47% of the original list of




  214 names of individuals who had been reported to have




  worked there since 1940.  Because almost half of the




  potentially exposed members were arbitrarily excluded




  from the study, the potential for bias is considerable.




  Furthermore, the criteria for exclusion of people from




  the study group were loosely defined.

-------
                                26




    o Duration of EDB production.   Although the author notes



      that this plant has produced  bromine from 1940  to 1973,



      nowhere is it mentioned when  the production of  EDB began,



      although it is stated that EDB production ceased



      in 1970.  Because some workers may have terminated



      employment before or shortly  after EDB production began,



      this raises the question of  whether all of the  117 "selected"



      employees were exposed to EDB for  4 years, the minimum



      duration acceptable for this  study.





In short, the methodological inadequacies of this study preclude



its use in an evaluation of the carcinogenicity of EDB to man.





The major flaws in the second Octel study, "Dibromoethane:   A



survey of employee records at Almwch Factory" (Turner, 1976)



are listed below:



    o Invalid age comparison.  The  author compared a  crude



      mortality rate by cause, derived from his study cohort



      during the period 1960 to 1975, with an age-specific



      rate of a control population  of males ages 45 to 64,



      residing in Gwynedd, Wales in 1974.  The average age of



      the cohort population was 40; the average age of the



      control group was much higher.  Therefore, because of



      the differences in the ages  of the groups, the  comparison

-------
                                27




      of the two groups could not be expected to demonstrate




      any excess mortality in the study group.  Details of




      this calculation are addressed in the CAG memo addressing




      comments to EDB risk (CAG, 1982b).






    o Invalid person-years calculation.  Twenty-eight deaths




      in the study group were noted during the period 1960 to




      1975, yet person-years were counted as  if those who died




      lived to the end of 1975.  This biased downward the




      death rates of the study group.




    o Small study group.  As in the first Octel study, the




      size of the population is not statistically large enough




      to assure that it would detect the presence of a significant




      cancer risk.






In summary, the two studies described above are not considered




to be adequate to evaluate the carcinogenic potential of EDB.






         2.   Mutagenicity






The Agency did not receive any comments regarding the




presumption of mutagenicity.   The Agency concludes that the




qualitative mutagenic criteria are fully established for EDB




In addition, the mutagenic properties of EDB support the Agency's




position that EDB is a direct acting carcinogenic agent.




Since the Agency issued the PD 2/3, the state-of-the-art in

-------
                               28




mutagenic risk assessment has developed to the point where



one can attempt to estimate risk to humans exposed to many



chemical mutagens (Banbury Report 1979).  The Agency is working



toward the goal of basing regulatory decisions on quantitative



mutagenic risk assessments.  Until the Agency has such a capa-



bility, our regulatory position is to reduce exposure to mutagens



to the lowest levels practicable.








         3.   Reproductive Effects



Two unpublished studies addressing the adverse reproductive



effects of EDB were submitted in response to the PD 2/3.



These studies are discussed below.



              a.   Ter Haar, 1981



Ter Haar (1981) compared the frequency distribution of sperm



concentrations in workers exposed to EDB with the frequency



distribution in clinical studies of fertile men (Smith and



Steinberger, 1977, Nelson and Bunge, 1974).  He reported that



these frequency distributions did not differ significantly.






The Agency concludes that there are a number of inadequacies



in the Ter Haar (1981) study.  Some of the more significant



ones are:






    o  The criteria for determining adverse reproductive function



       are not defined.

-------
                               29




    o No attempt was made to determine if sperm concentration



      from workers exposed to EDB accurately reflected true



      daily sperm production.






    o There was no matching control group, and there was no



      adequate comparison made between exposed and non-exposed



      workers.





Because of these and other inadequacies, this study cannot be



accepted as useful evidence for the lack of adverse reproductive



effects in workers exposed to EDB.

-------
                               30




              b.   Levine et al., 1981



Levine and his coworkers attempted to perform an epidemtological



study of fertility in male workers exposed to EDB.  They compared



the average number of children born to the spouses of married



workers with the average number of children born to all females



in the U.S., as reported by the National Center for Health



Statistics.





In this study, it was assumed that the reproductive health of



workers exposed to EDB was not adversely affected because the



average number of children born to the spouses of the workers



was similar to averages reported in the national statistics.






The major deficiencies in this study are:



    o The most significant deficiency is the lack of sensitivity



      for detecting adverse reproductive effects.





    o The investigators did not evaluate a large enough



      population of unexposed workers to make a comparison of



      fertility.  Because of the lack of an adequately matched



      control population, this study cannot discern whether



      potential differences are the result of exposure to EDB



      or other confounding variables.

-------
                                 31




    o The investigators did not consider such factors as the




      desire of workers to have children or use of contraceptives




      to control fertility.  An observation that an individual




      has produced offspring does not exclude the possibility




      of serious reproductive consequences such as subtle




      decrease of fertility, problems in sexual performance or




      reduced libido.






The deficiencies in this study preclude its use to determine




a measureable effect of EDB exposure on fertility.  Because of




these deficiencies, this study cannot be accepted as evidence



for the lack of adverse reproductive effects of workers exposed




to EDB.






An additional comment relates to both studies.  Neither study




examined the possibility of adverse reproductive effects in




female workers.  To adequately address the possibility of




adverse effects in all workers, it is also necessary to evaluate




reproductive ability in both sexes, and evaluate pregnancy outcomes




of women exposed to EDB.

-------
                               32




                         c.  Conclusion



Two recent studies (Ter Haar 1981; Levine et al., 1981) have



been sumbitted in response to the Agency's presumption in



the PD 2/3 of adverse reproductive effects resulting from



exposure to EDB.  These studies attempt to show that there is



not a significant difference between fertility of workers



exposed to EDB and fertility of the general population.  Because



they contain numerous deficiencies, these reports do not rebut



the presumption of reproductive effects as described in PD



2/3.  The Agency's detailed analysis of these studies is



contained in "REAG Response to Industry Rebuttals Concerning



The Reproductive Risks of EDB" (REAG, 1982).

-------
                               33




III.   Revisions and Amendments to PD 2/3; Exposures and




       Associated Health Risks




    A.   Applicator Exposure






Subsequent to the publication of the PD 2/3, the Agency received




substantive additional information regarding four applicator




exposure situations.  Therefore, the Agency has revised the




original estimates of applicator exposure for preplant soil




fumigation and quarantine fumigation of citrus, and completed




an exposure analysis for fumigation of felled logs and the




post harvest fumigation of papayas.






         1.   Preplant Soil Fumigation






RDB inhalation exposure of soil fumigation applicators has been




reassessed based on two factors (Reinert 1982 c).   First,  use




information, more reliable than that used in the PD 2/3, was




provided to the Agency in the comments of Great Lakes Chemical




Corp. (Comment #24).  Second, new exposure studies for this




use pattern were provided by Dow Chemical Co. in their comments




on the PD 2/3 (Comment #34).






The Agency estimates that the use of certain specified types of




respirators during all phases of application would reduce



inhalation exposure by about 90%, at the peak efficiency of




the respirators, assuming no breakdowns, malfunctions, or



improper use of the equipment (dermal exposure is assumed to




be insignificant compared to inhalation for this use).

-------
                                34

Based on the new data on average air levels encountered during

transfer and application, use of respirators solely during the

transfer of the pesticide could possibly reduce exposure by

approximately 80%.  The additional data have been combined

with the data used in the PD 2/3 and new averages determined

as shown in Table 1.


 Table 1.  Data on EPS Air Levels During Transfer and Application


              PD 2/3 avg.          New data           Revised exposure

               (mg/m^)             (mg/ra^)               (mg/m^)


Application    0.81*               14.6, 10.8              1.9
(open cab)                          3.8, 0.54


Transfer       3.03**              484, 238, 36.1        114.0
(open system)                      27.7, 7.7
*   Average of 21 data points

**  Average of 2 data points
Recently, additional applicator exposure data related to the soil

fumigation use of EDB has been made available to the Agency (CDFA,

1983).  The levels of EDB measured in the air during application

are consistent with values in Table 1.  Air levels measured while

loading the chemical are considerably lower than those in Table

1.  This is expected because closed mixing and loading systems

-------
                             35




were employed in the CDFA study.






Potential dermal exposure to the hands was measured  in  this




study.  The results are consistent with  the assumption  that




potential dermal exposure will be insignificant compared to




potential inhalation exposure  for this use pattern.






The revised respiratory exposure estimates for applicators




engaged in soil fumigation using open transfer systems  and




open cabs are shown in Table 2.
   418-574 0-83-4

-------
                                   36

          Table 2.   EPS Respiratory Exposure During Soil Fumigation
 Use Site
      Duration of
 annual exposure (hrs)
 Fruit trees
 Tobacco
 Peanuts
 Pineapples***

 Misc. vegetables
   and fruits
 Transfer

    0.8
    1 .8
    1.5
24.5 - 43.0

  2.6-7.8
Application

     7.2
    16.2
    13.5
220.5 - 387

  23.4-70.2
                 Annual
                 inhalation
            exposure (mg/year)**
     130
     280
     240
303.3 - 532.4

   410-1200
Sweet potatoes
Irish potatoes
Cotton
Citrus
8
8.1
2
0.8
72
72.9
18
7.2
1300
1300
310
130
 *   data from Great Lakes Chemical Corp., Table II of exhibit 2 of their
     rebuttal package.

 **  dermal exposure is assumed to be insignificant compared to inhalation
     exposure.

***  prom PD 2/3, where it was noted that this may be an overestimate of
     exposure for this crop,  because different agricultural practices are
     employed in the fumigation of pineapple fields.

     Sample calculation:  using as an example the use site "fruit
     trees"; 1.2 m^/hr (7.2 hours app./year x 1.9 mg/m3 + 0.8
     hours trans./year x  114  mg/m3) = 125.8 mg/year.
           2.   Quarantine Fumigation
     The Agency has received information from the Florida Department

     of Citrus demonstrating that work at citrus fumigation stations

     and warehouses should not be classified as "heavy activity."

     The breathing rate for "moderate activity" (1.2 m5/hr) has

-------
                                37




therefore been used to revise the applicator and warehouse




worker respiratory exposure calculations.  The Agency has also




received additional worker monitoring data from the Occupational




Safety and Health Administration (OSHA, 1981) on exposure to




truckers and fumigation station personnel.  This information




has been used by the Agency to revise exposure estimates for




citrus fumigation.






The additional data have been used to revise the Agency's




estimates of respiratory exposure to workers involved in



quarantine fumigation.  Two types of new data are utilized in




the Agency's revision of respiratory exposure.  The breathing




rate was reduced from 1.8 ra^/hr. for heavy labor (listed in




PD 2/3) to 1.2 m^/hr. for light to moderate labor.  The results




are lower estimates of exposure to EDB per year.  The second




kind of new data is EDB worker monitoring data now available




which were not available for the PD 2/3 assessment.  The revised




exposure estimates are presented in Table 3.






The Agency has received information from the Papaya Administrative




Committee regarding the flow/work practices of papaya shippers




in treating fruit with EDB (Comment #23).  The Agency previously




had no information on the treatment of fruit in Hawaii.   The




Papaya Administrative Committee states that in Hawaii, fruits



are delivered to the packing houses in wooden field bins.   The




fruits remain in these containers through fumigation until




they are hand-packed.  After hot water treatment at approximately




120° for 20 minutes for partial fruit fly disinfestation and



disease control, the fruits are dried and fumigated according




to APHIS requirements.  Following EDB fumigation,  fruits are

-------
        Table 3.  Revised EDB Inhalation Exposure of Workers at Citrus Fumigation Stations and  at a Citrus  warehouse
Location Workers Concentration
of EDB (mg/m3) Number Breathing Hrs/dayV
Avg. of samples Rate (m3/hr)V
Wahneta - Indoor operators
Ft. Pierce - Indoor operators
Wahneta - Outdoor operators
Ft. Pierce - Outdoor operators
Wahneta - Truckers/Station
Personnel
Ft. Pierce - Truckers/Station
Personnel
OSHA £/ Truckers/Station
Personnel
Wahneta - Truckers
OSHA6/ Truckers (during
coupling and uncoupling)
OSHA6/ Truckers (on road)
Tampa - Forklift operators,
Laborer (inside)
Tampa - Forklift operators,
Laborer (outside)
Tampa - "Stickmen"
OSHA6/ "Stickmen"
0.207
0.607
0.406
0.280
0.048V

1.092

0.18

1.93
4.26

0.07
8.99
2.727/
4.75V
2.36
1.65V
2.00V
4.76
8.57
14
1
19
2
3

1

9

2
5

5
1
unknown
5
3
unknown
17
2
6
1.2
1.2
1.2
1.2
1.2

1.2

1.2

1.2
1.2

1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
8
8
8
8
8

8

8

0.08
0.08

8
8
8
8
'8
8
8
8
8
Ceys/yrV
112
112
112
112
112

112

112

112£/
112£/

112
155V
155?/
155V
155V
155V
155V
155£/
155V
Avg. exposure to
EDB (mg/kg/yr)4/
3.2
9.4
6.2
4.4
0.8

1.4

2.8

0.2
0.6

1.1
192
58
101
50
35
42
101— r
182
1.  Assuming light activity (Specter, 1956).
2.  Hrs./day are variable depending on work load during year.  Average 8  hr.  work day is assumed  here.
3.  Data from APHIS (1977)
4.  Assuming a 70 kg worker.
5.  Including "non-detectable"  samples at level of  detection (0.008  m3).
6.  OSHA survey 1981 draft report
7.  From Florida Citrus Rebuttal
8.  Assuming trucker returns once each day
9.  Assuming 22 working days a  month over the period of October 11—through  May 14.
                                                                                                                                   CO
                                                                                                                                   oo

-------
                               39

typically stored until packing time.  The fruits are then

culled, sized and handpacked into cardboard cartons.


The Agency has received information from the Hawaii

Papaya Industry Association regarding exposure to applicators

fumigating papaya with EDB (Hertlein and Hagadone,  1981).  EDB

concentrations were measured in the ambient air of  the fumigation

area.  EDB levels ranged from 0.1 to 3.6 ppm; the average

concentration was 0.86 ppm or 6.6 mg/m3.  The Agency has

calculated an applicator exposure of 10 mg/kg/yr using the

following assumptions.


    o  The EDB concentration in the ambient air is  representative
       of applicator exposure.

    o  Applicators perform "light" to "moderate" work at a
       respiratory rate of 1.2 m^/hr.

    o  Applicators are exposed for 10 minutes per load of
       papayas fumigated.

    o  An applicator fumigates 530 loads per year.

    o  An applicator weighs 70 kg.

Sample calculation:

6.6 mg/m3xl.2m3/hr x 88 hr/yr x 1/70 kg = 10 mg/kg/yr


The person assigned to the job as applicator at papaya fumigation

station is very often also assigned to operate a fork lift to

move bins of fruit from one operation to another.  Workers

removing papaya from fumigation chambers would be exposed to

air concentrations of EDB from 0.07 to 1.6 ppm; the average

concentration was 0.54 ppm or 4.2 mg/m3.  Workers removing

papaya from refrigators after fumigation would be exposed to

air concentrations of EDB ranging from 0.02 to 0.53 ppm; the

average concentration was 0.36 ppm or 2.8 mg/m3 (Hertlein and

-------
                              40




Hagadone, 1981).  The Agency has calculated worker exposure of



8 mg/kg/yr and 6 mg/kg/yr, while moving bins of fruit from the



fumigation chamber and refrigators, respectively (Ludvik,



1982; Saito, 1982).





Fruit sorters and packers are exposed to air concentrations of



EDB ranging from 0.09 to 0.22 ppm;  the average concentration



was 0.14 ppm or 1.1 mg/rn^.  Assuming a 60 kg woman works an 8



hour day, 5 days per week and 48 weeks per year, she would



be exposed to 42 mg/kg/yr of EDB.   If the woman works an 8 hour



day, 3 days per week and 50 weeks per year, she would be exposed



to 26 mg/kg/year (Ludvik, 1982; Saito, 1982).





A summary of all applicator exposure estimates is presented in



Table 4.

-------
          Table 4.  Sunroary Table of FD 4 Applicator Exposure Estimates Compared with Exposures fron FD 2/3

                                                          mg/kg/year I/
Wbrkers at Citrus Fumigation
Stations In Fla. 3_/ (inhalation)

     W  - indoor operators 4/
     FP - indoor operators
     W  - outdoor operators
     FP - outdoor operators

     W  - truckers/station personnel
     FP - truckers/station personnel
     OSHA - truckers/station personnel
     W - truckers
     OSHA - truckers (during coupling)
     OSHA - truckers (on road)

     T - forklift operators/laborers (inside)
     OSHA - forklift operators/laborers (inside)
     T - forklift operators/laborers (outside)
     OSHA - forklift operators/laborers (outside)

     T - "stickmen"
     OSHA - "stickmen"

Hawaiian Papaya

     Applicator

     Packer/Sorter

     Indoor laborer
Felled log 2/
Applicators
                              PD 2/3
                            4.8
                            14
                             9.4
                             6.5

                             1.1
                             2.1

                             0.4
                           287

                            75.3


                           151.8
                            PD 2/3
     PD 4
dermal

inhalation
   3.2
    9.4
    6.2
    4.4

    0.8
    1.4
    2.8
    0.2
    0.6
    1.1

  192
  101
   50
   42

  101
  182
   10

26-42

   14

    PD 4
      69

      24

-------
     Table 4 Continued.  Summary Table of PD 4 Applicator Exposure Estimates Compared with Exposures from FP 2/3

                                                          mg/kg/year I/

Soil Fumigation:                                               pp 2/3                      FD 4
Applicators (inhalation)

     tree site                                                 0.01                       1.9
     tabacco                                                  0.18 - 0.44                 4.0
     peanuts                                                  0.21                        3.4
     Misc. fruits                                             0.18 - 0.92                 5.9 - 17
      & vegetables
     sweet potatoes                                           0.034 - 0.053              19
     Irish potatoes                                           0.07 - 0.12                19
     cotton                                                   0.35 - 1.8                  4.4
     citrus (preplant)                                        0.44 - 0.56                 1.9
     pineapples                                               4.3 - 7.6                   4.3 - 7.6
     citrus (barrier)                                         7.4                         7.4

Spot treatment                                                                                                             >U

     applicator                                               4.3 - 59.3                  4.3 -59.3

     mill worker                                              9.4-10.9                  9.4 - 10.9
I/   body weight = 60 kg for packers/sorters for papaya, 70 kg otherwise.

2/   for this use pattern, the number of years an individual would work as a fumigator is estimated to be
     1.25 years/lifetime (4).

3_/   exposure stidies for workers involved in citrus fumigation in California have been carried out (18,32),
     and these studies have been evaluated (33, 34).  However, the use information (annual and lifetime durations
     of exposure) necessary for exposure assessment is unavailable.

4/   W = Wahneta; PP = Ft. Pierce; T = Tampa

-------
                               43

         3 .    Felled Log Fumigation


In the PD 2/3, the Agency reported that there were no data

available on EDB exposure to felled log applicators.  The

Agency has since obtained data on EDB air concentration during

felled log application (USDA, 1973), use rate information

(Day, 1981a  and EPA, 1981), and surrogate data on dermal

exposure to 2,4,5-T resulting from a similar application method

(Lavy et al. , 1980).   The Agency has estimated EDB exposure

to felled log applicators based on these data.  The exposure

estimates assume a breathing rate of 1.2 m^/hr. for moderate

activity, and that the average applicator weighs 70 kg.  The results

are summarized in Table 5 below.

       Table 5.  Estimated EDB Exposure to Felled Log Applicator

                      mg/hr.  mg/day   mg/yr.   mg/kg/lifetime*

         Dermal       75      300      4800      86
         Respiratory  26      104      1664      30
* 1.25 years/lifetime spent in application (Ludvik, 1982).  In using
these surrogate data to estimate EDB exposures, the Agency recognizes
that questions have been raised regarding whether the Lavy patch data
adequately represent the amount of dermal exposure to the applicators,
However, at this time,  the Lavy data are the best data available to
the Agency which can be extrapolated for this particular use pattern.
Therefore, the Agency is using these data with the understanding that
any EDB exposure assessment, based on these data, is subject to the
same qualifications and limitations as the original Lavy data.

-------
                               44




    B.    Applicator Risk






The Agency's Carcinogen Assessment Group has estimated the




cancer risks associated with occupational exposure to EDB in




quarantine fumigation of citrus, quarantine fumigation of



Hawaiian papayas, felled log fumigation, soil fumigation, and




spot fumigation of grain milling machinery.  Risk figures were



derived using the original linear non-threshold model discussed




in the PD 2/3.  These risk figures are based on a rat inhalation



study (McGaughy, 1982) and are presented in Table 6.






The Agency calculated risk figures for applicator by assuming




that the major route of applicator exposure will be  via in-




halation.  For the felled log treatment the Agency does assume



that dermal exposure could be a significant route of exposure.




In addition to the exposure received on the job, workers will




also be exposed to a similar amount of EDB in their diet as will




the general public.  Some workers, particularly those involved in



soil fumigation, may also be exposed to contaminated drinking water.






Proper use of respirators, operating at peak efficiency, during all



phases of application could reduce inhalation exposure by about 90%,




and could result in similar reduction in increased cancer risk.




It should be noted that this estimate assumes that that the



respirator is in good repair, properly fitted, and worn at all




times that an exposure potential exists.  The latter consideration

-------
                              45



is particularly important because some tasks e.g. packing and



sorting papayas and soil injection, could result in low levels



of EDB for approximately 8 hours per day.  Physiological and



psychological limitations make wearing a respirator for extended



periods highly unlikely (NIOSH,  1976).

-------
                              Table 6.  Revised Cancer Risk Due to Occupational Inhalation of EDB
Exposed group
 Range of exposure
(mg/kg/day)	
Lifetime probability
estimates
                            No. of workers
                            exposed	
Soil injectioh (6)
  f armer/applicator
Quarantine fumigation (6)
  (Florida citrus)

  station indoor
  operator

  station outdoor
  operator

  trucker/station
  personnel

  warehouse indoor
  laborer

  warehouse outdoor
  laborer

  warehouse stickman

  Quarantine Fumigation (6)
     (Hawaii papaya)

     applicator
 3.0 x IP"3 - 3.0 x 10~2     3.5 x 10~3 - 3.5 x 10~2      14,000
 5.0 x ID"3 - 1.5x 1(T2


 6.9 x ID"3 - 9.7 x 10~3


 3.1 x 10-" - 4.3 x ID"3


 1.6 x ID"1 - 3.0 x ID"1


 6.6 x 10-2 - 7.8 x 10~2


 1.6 x lO-1 - 2.8 x 10"1




 1.5 x 10-2
5.9 x 10-3 - 1.7 x 10~2        nl/


8.1 x 10-3 - 1.1 x ID"2        10V


3.6 x 10~4 - 5.0 x lO"3        lo£/


1.7 x 10-* - 3.0 x HT1        4-i-


7.4 x 10~2 _ 8.7 x 10-2        4_10


1.7 x 10-1 - 2.8 x 10-!        4-10




1.7 x ID"2
                                                                     as
sorter/packer
indoor laborer
Spot fumigation (4) (6)
applicator
millworker
Felled log treatment
Inhalation
dermal
total
4.1 x 10-2 _
9.1 x 10~3 -
6.6 x 10-3 _
1.5 x 10-2 -
1.2 x 10-3
3.4 x 10-3
6.9 x 10-2
1.3 x 10-2
9.1 x iO-2
1.7 x 10-2


4.7 x lO"2 - 7.8 x 10"2
1.0 x 10-2- 1.4 x iQ-2
7.7 x IO-3 - 1.0 x MT1
1.7 x 10~2 - 2.0 x lO"2
1.4 x ID"3
8.4 x 10-2
9.5 x 1Q-* (5)
unknown
unknown
2400 - 6000
16,000
unknown

(1)  Assumed that 31 federal and state employees are evenly distributed.  (Mittelman 1980)
(2)  Per warehouse, Seasonal variation.  Assumed that 12-38 laborers per warehouse are evenly distributed  (Mittelman 1980)
(3)  Ludvik 1982c
(4)  Figures have not been revised - no additional data received since publication of EDB PD 2/3.
(5)  Based on an inhalation model, assuming 100% absorption and 1.25 years of application exposure/lifetime.
(6)  Exposure is without proper use of respirator.

-------
                              47




    C.  Dietary Exposure from Quarantine Fumigation






The Agency has modified the PD 2/3 estimates of the dietary




exposure resulting from fumigation of citrus and tropical




fruit in three significant ways.  First, the citrus food factor




pertaining to the consumption of fresh fruit has been used,




rather than the factor for all citrus (both fresh and processed).




This change has been made because EDB is used only on fruit




destined for the fresh market.  Second, residue data on the




edible portions of citrus and tropical fruit has been used,




rather than data on the whole fruit.  Use of whole fruit in




estimating exposure tends to exaggerate true dietary exposure,




because EDB residues tend to concentrate in the peel and seeds




- portions of. the fruit which are not typically consumed.




Some consumers of treated citrus do consume portions of the




citrus peal; dietary exposures to those persons would be




underestimated.






Further, in response to a recommendation by the FIFRA



Scientific Advisory Panel, the Agency has used additional data




that were not previously available for residues of EDB in




citrus and tropical fruit.  A recent study by the California




Department of Food and Agriculture (CDFA) has been used (Maddy




et al., 1981a).  In this study, samples of fumigated fruit




originating in Florida, Texas and Mexico were randomly taken from



wholesale and retail markets in California and analyzed for




EDB.  Normal commercial practices, such as typical time intervals



from fumigation to the store shelves, were assumed.  Third,

-------
                                48




more accurate estimates of the percent of fumigated fresh




fruit on the grocery shelves in various geographical areas




have been used rather than assuming a uniform national average




treatment of 2.6% of fresh citrus and 72% of fresh tropical




fruit.  This change has been made because certain states, such




as California, require fumigation of most incoming citrus and




tropical fruit, while other states do not require fumigation




of any fruit.  Thus, California consumption of fumigated fruit




exceeds the national average.  We expect dietary exposures,




calculated for California, to also be representative of other




locales which require fumigation.  These areas include Texas,




Arizona and New Mexico.




In addition, data collected by USDA's Agricultural Marketing




Service indicate that most fresh citrus imported from Mexico




(all of which is treated with EDB), is shipped to the Southwestern




and Western U.S.  Although data specifically identifying the




amount of citrus shipped to the individual states is unavailable,




the Agency believes that this contribution would increase dietary




exposures in the Southwestern and Western U.S.






If such data were available, the estimate of the total volume




of EDB treated citrus consumed  in California would be larger,




i.e., it would include both domestically produced EDB-treated




citrus and imported EDB-treated citrus.  The USDA data further




indicates that a small volume of the citrus shipped from




Florida to California is ultimately shipped to Hawaii and



consumed there.  Although exact figures are not available,




this will result in a slight overestimate of the California-




consumed citrus treated with EDB.  The Agency does not expect




the proportion of EDB-treated citrus consumed in Hawaii to

-------
                                49

differ significantly from the national average (Zygadlo,

1982b).   A summary of the data collected by CFDA prior to

1982 on residues of EDB in the edible portions of citrus and

tropical fruits is shown in Table 7.


  Table 7.  Residue Levels in Edible Portion of Fruit (ppm) \
                        Number of           Range of
                    Samples Analyzed   Values Observed^/ Mean 4/
Grapefruit
Limes
Papaya V
Mangos
Lemons
Oranges
24
10
35
27
3
6
ND
ND
ND
0.002 -

ND
0.326
0.01
0.102
0.27
ND
0.24
0.051
0.0013
0.011
0.057
ND
0.048
(1)  Source:  Maddy, K.T. jrt aj. . ,  CDFA Report HS-959, Oct. 21, 1981b.
(2)  Limit of detection of analytical method is 2 ppb.
(3)  Wholesale and retail samples  combined.
(4)  Non-detectable samples were assigned a value of zero.

-------
                               50




More recent data (CFDA, June 2, 1983) has shown residues of EDB




in fruit can be even higher:






Commodity       Origin   Date Sampled   EDB Residues (PPM)

Grapefruit
Grapefruit
Oranges

Texas
Texas
Texas

2/2/82
2/4/82
1/23/82
Pulp
0.28
0.21
0.88
These data indicate that there may be significant seasonal




variations in the levels of EDB residues in fumigated citrus.




Three composite samples, shown above, (2 grapefruit and one



orange which were fumigated in Texas and collected by CDFA in




California in January and February, 1982) showed signficantly




higher levels of EDB than the CDFA samples used for the dietary




exposure assessment. The residue levels used in the dietary




risk assessment in this document were obtained from fruit



which were collected in September and October of 1981.






The average lower temperature in the winter months are likely




to be responsible, at least in part for the observed difference




in residue levels (Maddy, K.T.  personal communication to J.C.




Reinert, EPA, August 12, 1983).  In addition, the Agency received




a preliminary report from the Agricultural Research Service,




USDA and the Florida Department of Citrus on factors affecting




residue levels in grapefruit fumigated with EDB (USDA, 1983).




These reports demonstrated that at the current treatment rate




both time and temperature are factors in reducing EDB residue




levels in the edible portion of grapefruit.

-------
                                51




Although elevated storage temperatures resulted in lowered EDB



residues, temperature was not as significant as time in reducing



EDB levels (USDA, 1983).  Because this report was a controlled



study, not a monitoring study of fruit residues in commerce - it



is not useful for estimating dietary levels.  The results, however,



are not inconsistent with the exposure data for California used



in this document.)






Arizona, California, New Mexico and Texas require fumigation of



all fresh citrus shipments from Florida.  Additionally, Arizona,



California and New Mexico require fumigation of fresh citrus



shipments from Texas.  However, because of the success Texas has



had with their Mexican fruit fly erradication program, only



about 20% of citrus shipped from Texas to these states had to be



fumigated this past growing season.  Dietary exposure estimates



for domestic citrus were developed for California based on



1982 historical Texas and Florida citrus fruit shipment data



(Zygadlo, 1982a).  This information indicated that 17% of fresh



oranges, 35% of fresh grapefruit and 4.4% of other fresh citrus



consumed in California were fumigated with EDB.  Since the 1982



data were developed, the California dietary exposure estimates



may need to be reduced due to the success of the Mexican fruit



fly certification program in Texas.  Information submitted to



the Agency from CDFA indicate that, a smaller percentage of



citrus consumed in California may be fumigated, than indicated



in the 1982 estimates, but, CDFA estimates have not yet been



confirmed.  The dietary burden to California consumers is shown



in Table 8 using the original 1982 estimates.
  418-574 0-83-5

-------
                                  52
      Table 8.  EDB Dietary Burden from Quarantine Citrus Fumigation
                             in California
% of diet
mean residue fresh
Fruit level (ppm) fruit
Oranges 0.048
Grapefruit 0.051
Lemons and limes^/ 0.001
Other citrus 0.025

1.35
0.82
0.17
0.48

% of fresh dietary bur
fruit (mg/kg body
treated 5/ weight/day) ^
17 2.75
35 3.66
4.4 1.86
4.4V 1.32
Total 6.54
x 10~6
x ID"6
x 10-9
x 10~7
x 10"6
                                                                     4/
      (1) Assumed to be the same as for "miscellaneous citrus"
      (2) The Agency has conflicting information on the percent of
          lemons and limes which are fumigated with EDB.  Their
          estimated contribution to the dietary burden is insignificant
          regardless .

      (3) Sample calculation using "oranges" as an example:  Dietary
          burden = (0.048 mg/kg diet x (1.5 kg food consumed/day)  x
          (1/60 kg) x   0.0135 [fraction of diet which is fresh
          oranges] x 0.17 [fraction of consumed fruit which  has been
          treated] = 2.75 mg . EDB/kg . b. w./day .

      (4) If Mexican Citrus is included,  the addition to dietary
          exposure (Reinert, 1982, DB  = 0.411 x 10~6 mg/kg/day)
          to the above total (6.54 x 10~6 mg/kg/day) yields  a
          grant total  of 6.95 x 10~6 mg/kg/day.

      (5) Based on the 1982 estimate discussed on the previous page.
In states which do not reguire incoming domestic shipments of

fruit to be fumigated with EDB, dietary exposure to EDB will

still exist from consumption of fumigated imported Mexican  .

citrus and imported tropical fruits.  Mexican citrus accounts

for 1.38% of the overall fresh market in the U.S.   Fumigated

-------
                              53

tropical fruits account for 72% of the fresh tropical fruit

market in the U.S.  In the absence of specific data, the Agency

will assume that this fruit is uniformly distributed across

the country.


Table 9 compares the PD 2/3 and PD 4 estimates of the total

dietary burden which results from the use of EDB as a post-

harvest fumigant of citrus.
          Table 9. Dietary Burden Comparison;  Citrus*
                    (mg/kg body weight/day)

              PD 2/3                   1.2 x in"5

              PD 4 (non-CA             4.1 x 10~7
                    residents)

              PD 4 (CA residents)      6.9 x 1CT6
            *Including the contribution from Mexican citrus
            and using the 1982 estimates of percentage of
            fresh fruit treated.

The calculations below present the dietary burden estimates

for a "tropical fruit eaters".  It is assumed that "tropical

fruit eaters" consume 12 fruits/year.  These individuals

constitute a small portion of the U.S. population.  Since the

average EDB residue contamination has been determined to be

0.031 ppm (=0.031 mg EDB/kg fruit, the average residue level

for the all papaya and orange samples) (Reinert,  1982) then the

dietary burden for tropical fruit (mangos and papayas) can be

estimated as follows:

-------
                               54

	Example Dietary Burden Calculation for Tropical Fruits

                     1 yr.         0.25 kg.*          0.031 mg
DB = 12 fruits/yr. x 365 days  x      fruit      x       kg.

     x _1
       60 kg b.w.

DB = 4.25 x 10~6 mg EDB/kg/b.w./day

* 0.25 kg, the average weight of the edible portion of
papayas and mangos.

-------
                                55




D.  Contamination of Wheat By-Products from Bulk Grain and




    Spot Treatment of Grain Milling Machinery with EDB






    Grain Fumigation



    On February 27, 1980 a re-evaluation of the total dietary




    burden of EDB with respect to the post-harvest grain uses was




    submitted to Special Pesticide Review Division (J.W. Holder,




    RGB ro R. Johnson SPRD).   In that review it was estimated




    that the "probable" residue level for EDB in wheat bread




    derived from grain fumigated after harvest with EDB would be




    0.07 ppb (.008 ppb from on-farm and 0.062 ppb from off-farm




    uses).  It was also estimated that the "realistic worst




    case" level which assumed that the grain would be treated



    at the maximum rate, not  mixed with untreated wheat (no




    dilution), and minimum commercial transit times (minimum




    opportunity for dissipation of  residues)  would be 31 ppb EDB




    (7.2 ppb from on-farm and 23.4  ppb from off-farm uses).  A




    third estimate using a mass-balance approach yielded an




    estimated contamination level of 87 ppb.






    Since the above modeling  estimates were made, the results of  an




    investigation of actual residues in wheat flour from commerce,




    and biscuits produced from that flour has been published (D.




    Rains and J. Holder, J.O.A.C 64, (5), 1981, pp 1252-1254).



    The flour analyzed in the investigation were geographically




    representative of cities  throughout the continental U.S.




    All but one of the twenty-two flour samples contained detectable




    levels of EDB (more than  0.5 ppb).  The reported levels in




    flour ranged from ND to 4,200 ppb.  Twenty of the twenty

-------
                              56




two biscuit samples baked from the flour also contained



detectable levels of EDB.  The levels reported to have survived




the baking process ranged from ND to 260 ppb and averaged



35.7 ppb.






If these data are indicative of the actual residues in U.S.




finished wheat products, they indicate that the "probable"




case underestimated EDB levels, and that actual levels may




approach the "realistic worst case" level of 31 ppb.






Therefore, it is now the Agency's judgement that the "realistic




worst case" level is the more appropriate estimate for the




dietary burden from bulk grain uses, rather than the "probable



case" estimate used in the PD 2/3. Thus, 31 ppb EDB in grain



products is equivalent to a dietary burden of 8.0x10"^ mg




EDB/ks/day for a 60 kg person consuming 1500g of total food




per day.






Spot Fumigation




As discussed previously in PD 1 and PD 2/3, the available




residue data reflecting the levels of EDB in grain products



processed in mills spot-treated with EDB are of limited




value.  All of the available studies involved wheat with




significant levels of EDB before milling.

-------
                                 57




However, even though these data are not adequate to determine



with certainty the residue levels expected in much of the



flour milled during the four weeks between treatments, they do



show that spot fumigation does contribute significant residue



levels of EDB based on a time-weighted average over the period



of the month between treatments.






In the PD 2/3, the estimate for the residue levels contributed



by spot fumigation was based on a 1978 Litton Bionetics study



for Ferguson Fumigants, Inc.  The estimate of 68 ppb in flour



(based on a time-weighted-average over four weeks) made no



correction for the levels of EDB in the incoming wheat grain



because the exact portion of these EDB residues that transferred



to the flour (and thus to wheat by-products) was unknown, and



because several investigators have reported that the majority



of these are retained by other milling fractions (such as



bran, offal, shorts, germ, etc.).  However, even if  it is



assumed that all the EDB residues detected in the incoming



wheat were concentrated in the finished flour, the available



data still indicate that the spot fumigation contributed significant



residues of EDB to the flour.  It has been reported that EDB



levels in the incoming grain ranged between 30 to 50 ppb.



Assuming that all the EDB was concentrated in the finished



flour, the resulting levels would be 43 to 71 ppb in flour.



If these calculated background levels are subtracted from the



levels actually reported for flour (which ranged from 54 to



1,410 ppb), the calculated time-weighted-average residue level



would be 34 ppb.  Thus, the 34 ppb appears to be the minumum



net increase, over time between fumigations, that results in



wheat flour from spot fumigation.

-------
                               58




Accordingly, it continues to be the Agency's judgement that



the spot fumigation use contributes significantly to the



dietary burden of EDB, and that 3 ppb in finished baked products



(based on 68 ppb in flour) is a reasonable estimate of the



expected dietary burden.  Thus, 3 ppb EDB in grain products



is eguivalent to a dietary burden of 5.8 x 10~^mg EDB/kg.



b.w./day for a 60kg person consuming 1500g of food/day.



This is calculated from the worst case (J. Holder,1980)



DB = 7.7 x 10~6 times 3/4, where 3/4 of the mills are



estimated to use EDB for spot fumigation use.






The Agency is continuing to monitor for residues of EDB in



grain, flour, and finished baked goods.  This monitoring is



being undertaken in coordinated interagency efforts with



USDA and the Food and Drug Administration.  In addition to



flour and baked goods, milk, and meats including beef, poultry



and pork are being sampled.  The Agency will continue to



evaluate these data to establish a more complete understanding



of the extent of hazard to public health resulting from



fumigation of stored grains and spot fumigation of grain



milling machinery.  If the extent of the hazard posed by



either or both of these uses of EDB becomes more clearly



delineated and defensible, the Agency will consider emergency



suspensions for these uses.

-------
                               59

                 E. Ground Water Contamination



     As of September 16, 1983, BOB has been found in water

from approximately 114 different wells in four states.  In all

cases, EDB use as a soil fumigant is an extremely likely

cause or is at least a possible cause of the contamination.

A summary of the monitoring data follows.  A discussion of

the persistence and transport of EDB follows after the monitoring

summary, along with a discussion of the possible significance of

gasoline leakage to the contamination in ground water.


1.   Monitoring Results

     In 1980, the EPA helped design and sponsor a ground water

contamination study by the California Department of Food and

Agriculture.  The object of the study was to determine the

extent and mechanism of ground water contamination by four

pesticides in certain areas of California.   EDB was one of

the chemicals selected, and the draft report*  indicated that

two wells out of over two hundred sampled** were contaminated
*  The final report is being written in three volumes which are
   presently in various stages of peer review.

** Not all of the wells sampled were near EDB use sites.

-------
                                 60

with 0.1 ppb and 0.2 ppb EDB (Zalkin, et al., in progress).

More importantly, EDB was found at  concentrations between

0.1 and 31 ppb in the soil at depths greater than 20 feet,

moving down to ground water.  Since these preliminary reports

were received in spring, 1983,  EDB  has been found in ground

water in 16 different counties  in the states of California,

Florida, Georgia, and Hawaii.  The  levels reported vary

between 0.02 ppb and ca. 300 ppb, and are typically between

0.05 and 5 ppb.   The results are summarized in Table 10,  and

a state-by-state discussion follows.


                  Table 10  EDB Ground Water
               Contamination Known  or Suspected
                to be Due to Agricultural Use
States
California
Florida
Georgia
Hawaii
Total
No. of Counties
5
9
1
1
16
No. of Positives
(approx. )
18
86
6
L 4
114
Typical Positives
(ppb)
0.1-2.0
1.0-15.0
0.1-10.0
0.02-0.1
0.05 - 5.0
     In general, the detectable levels of EDB cited above

have been confirmed by two gas chromatography (gc)  methods

and/or gc-mass spectrometry when possible.

a.   California

     The monitoring has been done by the Department of Food

and Agriculture, the State Water Resources Control  Board,

-------
                                 61
and the Department of Health Services, with some assistance
from EPA.  The 15 to 20 positives have been in the San Joaquin
Valley area.  Approximately two small community drinking water
supplies have been contaminated.  Probably 250 to 300 water
samples have been negative for EDB over the last four years,
in California, but for many of these negatives:
     1) it is not known whether EDB had been used near
        (ca. 1/4 mile) the wells;
     2) the number of years of EDB use have often not been
        known;
     3) and, the minimum detection limit at the time of sampling
        was higher, making it more difficult to detect low
        levels of EDB than it now is.
State agriculture and health officials have concluded that
in most, if not all, cases contamination is likely due to
EDB use as a soil fumigant.  (A single positive in Solano County
has not been included in this assessment because it may have been
due to poor industrial practice.)

b.  Florida

     Two different areas of Florida have been sampled since
mid-1983 and both areas contain ground water contaminated
with EDB.  Those areas are an eight county citrus growing
region in central Florida, and the soybeans/peanuts area of
Jackson County, in the panhandle.
     As of early September, 1983,  86 out of 334 wells sampled
contain detectable levels (> 20 parts per trillion)  of EDB; 75
of these 86 positives contain EDB concentrations greater than
the level 0.1 ppb (Bigler, 1983),  which the state has selected
as an interim action level.  The contamination in the central
Florida area was most likely due to citrus buffer zone applications

-------
                               62
done by the state agriculture department;  the panhandle
contamination was most likely due to EDB soil fumigation of
soybean and/or peanut fields.
     The Florida monitoring program is a collaborative effort
between the Departments of Environmental Regulation,  Health
and Rehabilitative Services, and Agriculture.  They have been
sampling wells within 300 ft. of use sites, and intend to
sample a total of 1000 or more wells.  All sampling and
analysis has been done since approximately mid-summer, 1983.
As a result of this monitoring program, on September 16 the
State of Florida temporarily suspended the soil fumigation
use of EDB.
c.  Georgia

     EPA scientists were notified in June, 1982, by EDB
registrants that three wells were contaminated with EDB at
levels as high as 100 ppb (an irrigation well) in Seminole
County.  After the registrants failed to conduct any further
monitoring, the EPA and the USGS, in cooperation with the
county extension agent, initiated and completed a reconnaisance
study of the area in late summer, 1983.  Nineteen wells and
one surface water sample were analyzed, and six wells were
found to be contaminated with between 0.03 and 11.8 ppb EDB.
In addition, soil cores from one hole drilled to. the producing
aquifer contained EDB below 20 feet.  The pesticide was
known to be used throughout most of the study site; apparently,
the uses began after EPA suspended DBCP use  in the fall of 1979,

-------
                               63
The report (Jovanovich and Cohen)  is in review.  Conclusions
in this report are that the contamination did not result
from a single point source and was due to agriucultural use
of EDB.
d.  Hawaii
     Based on the preliminary results from California, in June,
1983, the Agency requested that the State of Hawaii monitor
soil and ground water for EDB.  Staff from the Departments
of Health and Agriculture, Castle  and Cooke (Dole), and others,
immediately began an active monitoring program (Wong, et
al., in progress).

     As of September 1983, most,  if not all, community water
supplies within one half mile of  EDB use sites have been
sampled and analyzed for EDB contamination.  Thus over 100
different drinking water wells have been analyzed for EDB
between June and September.  To date, one site has contained
EDB contaminated ground water - the four wells in the Waipahu
well field in south cental Oahu.   The levels in all four
wells range between 0.02 and 0.10  ppb.   EDB has been used on
pineapple fields near the Waipahu  wells.

     Soil cores were taken from five Oahu pineapple fields
from depths as great as 60 feet.   Three holes were drilled in
each field, for a total of 15 holes.  EPA has received partial
data on at least 9  of those holes.   The data analyzed thus far
show that EDB is present in soil  cores down to the 20 to 30
ft.  range.  However, a closer examination of the data indicates
that the soil core samples were most likely contaminated

-------
                               64
during drilling.  EPA scientists cannot say with certainty
what portion of the positive results was due to slow leaching
through the soil profile or what protion was due to contamina-
tion during the drilling process.  However, it is likely
that accidental contamination was a significant contributing
factor.  Thus the soil core results neither support nor
refute the hypothesis that the four wells were contaminated
through soil fumigation and leaching to ground water.  A
brief description of south central Oahu's volcanic soil
follows, along with a concluding statement about the problem.

     Most of the organic matter, which tends to bind neutral
organic compounds such as EDB, is confined to the top few
feet of soil, with a heavy amount of clay (ca.  80%) through-
out much of the soil profile (Green, 1983).  The flow rate
of water through the top few feet can increase significantly
during a rainstorm, and is relatively fast, while the under-
lying inert clays and saprolite have slower, steadier hydraulic
conductivity (Green, 1983; Eyre, 1983).  This material between
the fresh basalt and the top soil has been found to have
practically zero adsorptive capacity for DBCP and certain
herbicides (Green, 1983).

     Based on this description and the persistence and
transport discussion below, a picture emerges of the probable
fate of EDB in a typical volcanic soil.  Within several weeks

-------
                               65
of application, the pesticide leaches down into and diffuses
throughout the top couple of feet of soil.  During and after
this time, degradation may occur, to a possibly significant
extent.  There is likely a significant amount of sorption
to the soil, which is decreased during and/or after a rainfall
event.  This region of EDB sorption could serve as a "reservoir1
of EDB which could then leach in low and nonsteady amounts
into and through the underlying low organic clays, saprolite,
and fissured basalt.  Little if any sorption would occur in
this zone.  Therefore, no "reservoirs'  of EDB could be built
up (as in the top soil) and the EDB would be present in the
water phase.  From this point, leaching to the aquifer could
occur quickly with significant dilution occurring along the
way.  Therefore intermittent contamination of Oahu's basalt
aquifer will likely occur in unpredictable, although probably
low concentrations if continued EDB use is allowed.  The
term "low" in this context means concentrations not exceeding
a couple of hundred parts per trillion.

2.  Environmental Persistence and Transport Studies
     EDB has a water solubility of approximately 4500 ppm and
a room temperature vapor pressure of 11 torr.  Thus it would
tend to be mobile in both air and water.   It diffuses in
soils (Thomason and McKenry, 1974) and does not appear to
sorb well to soils relative to other neutral organic chemicals

-------
                               66

(e.g., Chiou, et al., 1979).   Therefore EDB is very mobile

in the subsurface environment.

     Not much is known aDout  EDB's persistence, but the Agency

expects it to be stable to photolysis in sunlight.   EPA

recently began a hydrolysis study (Jungclauss, et al., in

progress).  Preliminary indications are that its half-life

at 20°C (a likely ground water temperature in a southern

climate) is greater than one  year under neutral conditions*

and it may be as long as a couple of dozen years.  The study

should be completed in November, 1983.   Cohen, et al.  (1983)

have stated that one of the criteria for determining that a

pesticide may be a potential  ground water contaminant  is a

hydrolysis half-life of greater than ca. 6 months;  thus EDB

meets this criterion.


     There are two indications that EDB can degrade in soils.

Castro and Belser (1968) demonstrated that EDB was  degraded

to ethylene and bromide ions  under optimal conditions  in a

California agricultural soil.  However, even in the same

soil, the halflives in different experiments varied tremendously
* A likely hydrolysis product is vinyl bromide, a possible
  carcinogen.

-------
                               67
from ca.  2 weeks to ca. 18 weeks.  According to Cohen, et.
al., (1983) the soil half-life criterion for potential ground
water contamination is a half-life of greater than ca. 2-3
weeks.   Thus EDB meets this criterion as well.  Also, high
levels of bromide ions have been detected in the perch water
aquifer under the site of a several hundred gallon EDB spill in
1977 on Oahu (Mink, 1981).  This indicates breakdown of EDB
to an unknown extent in a three year period.

3.  Leakage from Underground Fuel Tanks
     Underground fuel storage tanks are known to leak, and
this leakage can be predicted (Rogers, 1983).  EDB has been
used for  several decades as a lead scavenger in gasoline and
aviation  fuel.   EDB constitutes approximately 0.03% by weight
of leaded gasoline, assuming 1.1 gm lead/gallon of gas, and
17.86% and 63% are the percentages of EDB and tetraethyl
lead in the tetraethyl lead mixture, respectively.  Thus,
there is  some possibility that EDB could contaminate ground
water through leakage of underground fuel storage tanks.
     418-574 0-83-6

-------
                               68
     This potential for EDB to contaminate ground water from
gasoline losses should be examined on a relative basis.  The
Agency estimates that over 20 million pounds of the pesticide
EDB are applied to soil in the U.S. annually.  To have an
equivalent loading to soil from losses of gasoline there
would have to be approximately 10 billion gallons lost
directly to the soil every year.  This rough calculation
is an overestimate which ignores surface and subsurface
attenuation of EDB.  However, even if the EDB applied to the
surface were attenuated 10 times as much as the EDB which could
leak from subsurface gasoline storage tanks, the equivalent
loading to the soil would still require a nationwide annual
leakage of 1 billion gallons of gasoline.

     Although gasoline is one of the more frequently reported
contaminants of wells, EPA has not seen any data so far which
demonstrates that the contamination described in section 1
above was due to this phenomenon.  EPA scientists would
expect that any wells contaminated by this mechanism would
also contain ethylene dichloride (EDC), which is also a lead
scavenger in gasoline (Johns, 1976).  Thus EDC would be a
good chemical marker for this contamination source.

-------
                              69
    F.  Dietary Cancer Risk Equation
The CAG has revised the estimates of cancer risks associated
with EDB residues in the diet.  This revision was made to
utilize all of the information collected in the experiment,
particularly the effect for partial lifetime exposure at a
given dose (Weisburger, 1977).  The model developed by the
CAG is essentially the one-hit model with "Weibull" timing
(CAG, 1982).   This model fits the observed data well, it is
consistent with the biological hypothesis of direct acting
genotoxic agents, and it permits a direct estimate of the
effect of partial lifetime exposure to be made.  The model
takes the form:
            P(t,d) = l-e~x
where:         x = (1.02 x 10~13) x d x Wm1/3[(t-s)7'6-(t-f)7•6]
              wm = average weight of the individual from age
                   f to s (in kg)
              s = age at start of exposure (years)
              f = age at end of exposure (years)
              d = mg/kg/day exposure from age s to f
              t = age at end of observation period (years).
For a lifetime exposure, Wm = 60kg., s = 0,  f = t = 70 yrs.
of age, the above equation has the form P(d)  =  1 -e'^ld^
It should be  noted that this lifetime risk equation is
different from the risk equation used for the PD 2/3.

-------
                                70
G.  Dietary Cancer Risks
    1)   Dietary Risks to the Average Consumer
The dietary burdens (DB) for EDB uses and the cancer risks for
the average consumer estimated  in PD 2/3 are reproduced in
Table 11.  The dietary burden for soil incorporation (food
only, not from leaching into ground water) remains the same
as PD 2/3 with the resultant risk being different due to the
use of a different and improved model to estimate risk.  The
estimate of bulk grain dietary  burden has been increased
significantly (for discussion of dietary burdens see Section
D) , and therefore, the risks for the average U.S. consumer
from this use also increases significantly.  The estimate
for additional cancer risk die  to bulk grain fumigation is
now 3.3 x 10~3.  This estimated risk from the bulk grain
use constitutes a majority (approximate 85%) of the known
dietary risks from EDB.

The spot use of EDB on grain milling machinery has been
estimated to have the same time-weighted average EDB residues
(from treatment to treatment) in wheat flour and baked wheat
products as in PD 2/3.  The cancer risk estimate for spot
fumigation has been increased to 2.4 x 10~4.  States not
reguiring guarantine EDB fumigations still have contributions
from EDB-treated  imported fruit, e.g.  from Mexico.  It was
assumed these imported fruits could be uniformly distributed for
fresh market consumption throughout the U.S.  The tropical fruit.

-------
                              71
are assumed to be 100% treated, and distributed throughout
the U.S.  The dietary burden for tropical fruit is also
estimated (Table 11, footnote 3).  Both of these contributions
of EDB are less than those in states (e.g.  California),
which do require quarantine fumigation of fruit from Florida.
It is estimated that the risk to an average individual (60
kg, consuming 1.5 kg food/day)  in those states requiring EDB
quarantine fumigation would be 2.4 x 10~~4 from fresh citrus
consumption.
     The drinking water cancer risk are summarized in the
following section (Section H).

-------
          Table 11.   Comparison of the  FD 2/3  and  FO I  4    Dietary Burdens  and Cancer  Risks  from ED3

                                          to the Average U.S.  Consumer
Fumigation
  Use
   EDB FD 2/3          ECB Lifetime
   Dietary Burden      Cancer Risk
(mg.EDB/kg. b.w./day)    (FD 2/3)
                   EDB  PD 4            EDB Lifetime
                   Dietary Burden      Cancer Risk V
                 (mg.EDB/kg.  b.w./day)    (PD 4)
1.  Soil Incorporation -     2.60 x lO"7         5.6 x 10~5          2.60  x 10~7  V       1.1  x KT5
2.  Vfrieat Grain

    (bulk fumigation)
   1.81  x 10~7         2.6 x 10~6
                   8.00  x  10~5  2/      3.3 x ID'3
3.  Spot  (wheat grain -
     milling machinery)
   3.29 x 10-6
5.1  x 10-5           5.80  x  10~6  5/     2.4 x 10~4
4.  Fruit  (quarantine
      fumigation, bulk)

    in states not
    requiring fumigation

                citrus -

              tropical -
                                           4.11  x 10-7

                                           4.25  x KT6  V
                                        1.7 x 10-5

                                        1.7 x 10~4
                                                                                           -3
                                                                                           CO
    in states
    requiring fumigation
                citrus -     	
    in entire U.S.
    population
                citrus -     1.09 x 10~5

              tropical -     1.70 x 10~6
                       2.1  x lO-

                       3.2  x 10-5
                                           6.8  x 10-6
                                         (See table  8  & 9)
                                                                                         2.8 x
  Footnotes to Table 11 on next page

-------
                                            Footnotes for Table  11

(1)   No residues found at a detection  limit  of 1  ppb EDB.  Assumption  that EDB  residues can occur at 1 ppb.
     CB = (1  x 10"9,  residue  level)  x  (1.5 x 106  mg food  consumed/day)  x  ( 0.1038,  food factor  for wheat) x
           (1/60 kg,  body weight for "average U.S.  consuner")  =   2.6 x 10~6 mg/kg/day.

(2)   Dietary  burden increased significantly  from  the PD 2/3  (Table  22,  Probable Case Estimates  of DB)
     See text for rationale for estimating EDB dietary burdens in PD 4  in accordance with  the Realistic worst
     Case proposed in the report by  J.  Holder, 2/27/80.


(3)   Dietary  burden for a "tropical  fruit eater":   DB =  12  fruits/yr  x 1 yr/365 days x 0.25 kg /fruit x
     0.031 mg  EDB/kg  fruit  x 1/60  kg  body weight.  Assumed  a  rate of 12 fruits/yr and an average between
     a papaya and a mangoe  is 0.25 kg/fruit.

(4)   Lifetime risks for an average person (eats 1500 g food/day and weighs 60  kg  ) for a  lifetime of ex-
     posure and observation for cancer were  calculated by the  equation P = 1 - e~41d  where d  = dose =
     exposure in mg EDB/kg/day

(5)   The esimated spot contribution  has been changed from the  probable estimate  (3.2 x 10~6) to the realistic
     worst case (7.6  x 10 ~6) which  is estimated  in the J. Holder report  dated  2/27/80.  This realistic
     worst case has been adjusted by 3/4 in  accordance with  the  most recent use information in  Chapter IV, PD 4.
     The adjustment ( 0.75 x  7.6 x 10~6) makes the  dietary burden estimate for  spot = 5.8  x 10~6 mg/kg/day.

-------
                                  74

2.)   Dietary Burdens from EDB at Various Ages and Age-Related

     Variation in Risks to BOB due to one Year of Additional EPS

     Exposure


 The estimated dietary burdens (DB)  at ages of 2, 16, 30,  and 50

 years are presented in Table 12 with assumptions and example

 calculations presented in the footnotes.  The total DB (and DB

 from each fumigation use) decreases with age (i.e., the DB is

 highest for the young*)


 The estimates of dietary burdens in Table 12 are used to calculate

 estimates of cancer risks for one-year of additional exposure to

 EDB with observation for cancer for the rest of a normal (70

 years) lifetime. It can be seen in Table 13 that these incremental

 risks decrease with age.  Males were used for the calculation in

 Table 13.  Female risks are approximately 20% less than the male

 risks.
 * These age-specific estimated dietary burdens reflect the variation
 with age of body weights and daily food consumption (both in
 total grams/day and in percentage of each type of food consumed
 per day).  The later, variation of food factors with age, are
 not known at this time, but are being estimated by the Agency
 at this time and will be available at a later time
 (Chaisson, EPA).

-------
                               Table 12:  Dietary Burdens (DB)  Resulting  from ECB  Fumigation Uses



                                                   (Age  Specific Estimates)



                                                        (mg  EfB/kg/day)

Fumigation Use
Soil Incorporation
(food only)
Grain (wheat, bulk fumigation)
Spot (grain milling machinery)
Fruit Quarantine
( in states not requiring fumigation)
citrus-
tropical-
Fruit Quarantine
( in states requiring citrus
fumigation)
Estimated
2 years old
8.84 x 10-7
2.61 x 10-4
8.38 x IIT6


1.40 x 10~6
negligible

1.44 x 10~5

Dietary Burdens
from EDB
16 years old
3.91 x
1.21 x
3.70 x


6.17 x
4.18 x

8.64 x

10
10-4
10-6


10-7
10-6

i
-------
                                                    Footnotes for Table 12


1.  Dietary Burdens (CB) are at the ages indicated and are calculated in PD 2/3 for 60 kg individual eating
    1500 grams food day.  CB's presented in PD 2/3 for 60 kg individual eating 1500 grains food/day.  CB's
    presented in Table 	 were converted to EB's here for different age groups by mulitplying by the appropriate
    correction factors for different food consumption and weight.  For instance, for a two-year old male
    weighing 14*&-kg and eating 1232g/day:

    soil fumigation DB                                                             Soil Fumigation CB
     (in PD 2/3)                               conversion factors                      (PD 4)

    2.6 x 1CT7 mg/kg/day     x          60 kg	   x      1232 g /day      =   8.84 x 10~7 mg/kg/day
                                       1500 g/day                14.5 kg
        "average person"
     weighing 60 kg, eating 1500g./day                                              CB for 2 yr. old

    This type of conversion was applied to soil incorporation, spot, and citrus quarantine fumigation
    uses where the contamination levels (in ppb of ED3) were reconsidered to be the same as in PD 2/3 and
    are carried over here (in Table 12) in the PD 4 for different age-group consumers.

    Data used in the conversions:
                                 2 yr old     16 yr old     30 yr old     50 yr old
       body weight (kgs.) -         14.5          60.94         73             77

    daily food consumption -       1232          2288           2208          2250
         (grams/day)

      conversion factors -         3.398          1.502         1.227         1.169

    Female consumption and body weights accounted for a 20% (approx.) decrease in DB, and thus a 20%
    decrease in the female risks.  The examples shown in Table 12 are based on male body weight
    and food consumption.



2.  Bulk wheat grain fumigation was re-estimated to be higher than in the PD 2/3 estimate of 0.07 ppb
    and is now estimated to be 31 ppb ED3 (see text for rationale).

    The DB is calculated here is for a 16 yr. old male, assuming 100% of the commodity is contaminated
    at the 31 ppb level, and wheat products are 10.36% of the diet.

                D3 = 31xlO~9 x 2288 g/day x 1000 mg/g  x 0.1036      =  1.206 x 10~4 mg ECB/kg bw/day
                                 60.94 kg body weight

-------
                               Footnotes for Table 12 (Cont'd)


3.  The dietary burden for a "tropical food eater" is used for the quarantine use of EDB on tropical
    fruit to to be shipped to the continental U.S.
    It was assimed that a reasonable level of consumption is 12 tropical fruits per year.
    Thus, the DB can be calculated, for a 16 yr.  old male in the U.S.:

               DB  =  12 fruits/yr x  1 yr/365days  x  0.25 kg /fruit  x 0.031 mg  EDB/kg fruit
                                         60.94  kg body weight for a 16 yr old male


          where  0.25 kg is the average weight of papayas and managers and mangos on per
                 fruit basis 0.031 mg EDB/kg fruit  =  31 ppb ECB average contmination level
                 12 fruits/yr.  is assumed for the 16, 30, and 50 yr. old people and 2 year
                 old children were assumed to eat negligible amounts of tropical fruit.

4.  The total DB's frcm each use are summed from the Agency's current understanding of probable
    ECB contaminating sources.  Sources of possible contamination include meat and
    milk.  Meat and milk could be contaminated by livestock being fed on the farm, or in
    fattening feedlots, ECB contaminated grains.   The Agency does not have data on these
    conodities, nor can any reasonable estimates be made at this time due to a lack of information.

-------
                              Table 13: Incremental Risks from One Year o£ Additional


               Dietary Exposure to BOB at Various Ages with Observation for Cancer for a Lifetime £/
Fumigation Use
Soil Fumigation -
(food only)
Wheat Grain -
(bulk)
Spot (Fumigation
milling machinery)-
Fruit Quarantine
( in states not
requiring fumigation
Citrus -
Tropical -
Fruit Quarantine
( in states requiring
fumigation)
citrus -
Beginning and Ending Age of One Year of Additional
2 to 3 yrs. old 16 to 17 yrs. old 30 to 31 yrs. old
2.0 x KT6 3.0 x 10~7 3.5 x 10~8
5.8 x 10~4 9.4 x 10~5 1.0 x 10~5

1.9 x KT4 2.9 x 10~6 3.3 x KT7


3.1 x 10~6 4.8 x 10-7 5.6 x 10~8
negligable 3.3 x 10~6 3.9 x lO"7

3.2 x 10~5 6.7 x UT6 7.8 x Kr7
Exposure to ECB
50 to 51 yrs. old
3.3 x 10-10
9.7 x 10-8

3.1 x 10~9


5.2 x 10-10
3.6 x 10-9

7.3 x 10~9
                                                                                                                                    -3
                                                                                                                                    00
1.)  The risk model used was that described in IV F.  Risks for women were ..-     ,U:d to be approximately 20% less
than males because of differences in dietary burden.  Male risks are present.  .  u this table.

-------
                                79
H.  Ground Water Cancer Risk
The model, discussed in Section III F, was used to estimate

the cancer risks below.


The detection limits for EDB vary in different laboratories

between 0.02 parts per billion (ppb) and 0.10 ppb.  Therefore,

the Agency has included these detection limits in Table 14

below, in addition to other observed contamination levels.
Table 14.    Cancer Risks for EDB Ground Water
             Contamination
Concentration
0.02
0.1
1.0
5.0
100
ppb
ppb
ppb

ppb
Lifetime Cancer
3 x 10-5
1.5 x 10-4
1.5 x 10-3
7.5 x 10-3
1.5 x 10-1
Risk*





* EPA assumes consumption of 2 liters of water per day for an
  adult, a 60 kg lifetime average weight, and a 70 year lifetime,
  This 2 liter assumption includes water consumed per se,
  as well as water used in reconstituting powdered drinks, etc.
  Cancer risks, as presented in this table, are the additional
  risks estimated for a lifetime of exposure.

-------
IV.  Analysis of PD 2/3 Comments and Revisions Relating t.o Benefits

Detailed analyses of the benefits derived from registered uses of
EDB are presented in the PD 2/3.  Benefit analyses for the major
soil nematicide sites have been substantially revised since
issuance of the PD 2/3.  The following discussion is a summary
of the results of the updated analyses, as well as a summary
of comments which were received in response to the PD 2/3
addressing the benefits derived from the use of EDB and the
Agency's response to these comments.
         1.   Preplant Soil Fumigation

Ethylene dibromide is applied as a soil fumigant on numerous
agricultural crops for nematode control.  Use as a soil fumigant
accounted for over 90 percent of the total pesticidal use of
EDB in 1978.  Since 1978 use of EDB as a soil fumigant has
increased significantly.  In 1978, the Agency estimated that
cancellation of this use could result in adverse economic
impacts of over $20.5 million annually (1978 nominal dollars).
The 1983 analysis estimated that cancellation would result in
adverse economic impacts at the user level from $26.4-$42.8
million (1983 nominal dollars) annually with additional effects
at the market and consumer level.

When the original benefit analyses, used as input to the EDB PD
2/3 were made, the soil nematicide market was unstable due to
regulatory action anticipated on dibromochloropropane (DBCP).
The PD 2/3 benefits analyses incorporated assumptions which
were subject to uncertainty concerning how EDB would be used
in the absence of DBCP.  In addition, since issuance of the PD
2/3, EDB was registered.for use as a soil nematicide on soybeans.
This registration now accounts for approximately half of total
EDB pesticidal use.
                                 80

-------
                                81




Due to new concerns over EDB contamination of groundwater and



substantive changes in EDB's soil nematicide market since the



PD 2/3, new benefit analyses on the major soil nematicide sites



were conducted during the summer of 1983.  The analyses were



undertaken to identify the locations and volumes of the current



use of EDB as a soil nematicide, and to assess the current



(1983) benefits of this use.  New benefit analyses were completed



for soybeans, California vegetables, cotton, peanuts,  tobacco,



sweet potatoes, white potatoes, and pineapples.  The results



of these analyses are summarized in Table 15.






Currently, approximately 1.0 million acres are treated for



nematode control with nearly 23 million pounds of EDB active



ingredient annually.  The major use sites are soybeans (12.8



million pounds active ingredient), cotton (4.5 million pounds



active ingredient), and peanuts (2.2 million pounds active



ingredient) with lesser volumes applied to other crops.  Most



of the use is located in the southeastern United States.





Soybean, peanut, and pineapple growers are expected to experience



the largest economic impacts.  Aggregate annual impacts at the



user level range from $26.4-$42.8 million annually dependent in



the selection and efficacy of available alternatives.   The primary



substitutes for EDB soil nematicide formulations are expected to



be D-D, Telone®, Nemacur®, Mocap®, Furadan®, and Temik®.

-------
                  Table 15.   Annual Economic Impacts of the Cancellation of EC8 As a Soil Nematicide on
                                                     Major Use Sites
                                                                              Economic  Impacts
Crops


Soybeans





California
Vegetables I/



Tomatoes


Beans^/


Carrots


Other produce
(melon,
watermelon.
squash, bell
peppers, as-
paragus, head
lettuce, cucumber
peas, cauliflower
pimpkin, parsnip)
Acres Treated
000 's acres

532.8










8.7


11.4


3.7


2.3








Active Ingredient
000 's pounds

12,787.0










177.3


173.6


68.2


70.5








User


User impacts of $7.0 million
due change in control costs.
Price of soybean oil may
increase by $0.01-30.03 per
pound. No loss in production
is anticipated.
User impacts resulting from
increased nematode control
costs. No loss in production
anticipated.

Increased control costs of
alternatives range from
$0.3-$1.1 million
Increased control costs of
alternatives range from $0.5-
$1.5 million
Increased control costs of
alternatives range from
$0.5-$1.5 million
Increased control costs of
alternatives range from
$0.1-$0.3 million






Market Consumer

Welfare losses to
consumers of about $7
million in second year
after cancellation, with
decreasing losses in
later years (declining
to $0 in 5-7 years) .
Increased production
costs may cause some
users to grow different
crops. Consumer impacts
would be negligible.


















                                                                                                                                    oo
                                                                                                                                    CO
I/ ECB use was recently suspended by CDFA in the countries of Fresno, Kern,  Merced and Stanislaus.
   These counties apply a significant proportion of the EEB used in California.   The decrease is
   not reflected in the estimates given in this table.
2/ Represents use on both green lima beans and dry lima beans.

-------
.c.
O>
 I
Ul
                                           Table 15.   Annual  Economic  Impacts  of  the Cancellation  of  EDB As  a Soil  Nematicide on
                                                                          Major Use Sites (Cont'd)
                           Crops
                           Cotton
                          Peanuts
                          Tobacco
                          (flue-cured)
                          Sweet Potatoes
                          White Potatoes
                                           Acres Treated
                                             000's acres
                                             185.0
                                             134.5
68.1
 9.1
                                              22.5
              Active Ingredient
               OOP's pound

               4,500.0
               2,151.8
913.6
                                                           160.2-236.4
            960.0-1080.0
                                                                                                        Economic   Impacts
                                                                                          User
                  S3.0 million user impacts.
                  Primarily southeast region.
                  Impacts result from increased
                  production costs.  No loss
                  in production is anticipated.
                  Change in control costs of $4.1
                  million.   Change in value of
                  production of S3.8 million due
                  to planting delay and reduced
                  efficacy.
                  Aggregate increase in treatment
                  cost $2.02 million.  No loss
                  in production anticipated.
                  $0.3-$0.5 million user impacts
                  resulting from increased control
                  costs.  No loss in production
                  anticipated.
                                    User impacts from increased
                                    control costs $1.6-$2.2
                                    million.  No loss in
                                    production anticipated.
                                                       Market/Consumer
An additional cost for
nematode control could
cause growers to change
crops.  This would most
likely occur in the south-
east region.  Negligible
consumer impacts.
                                                                                                                     None.
                                                                                                                     None.
                                                                                                                     None.
                                                      Negligible.
                                                                                                                00
                                                                                                                oo

-------
                 Table 15.  Annual Economic Impacts of the Cancellation of EDB As a Soil Nematicide on
                                                   Major Use Sites (Cont'd)
                                                                            Economic  Impacts
   Crops
Pineapples
Total
               Acres Treated
000's acres

   4.0
   982.1
                Active Ingredient
000's pound

    810.0
22,772,2-22,968.4
                                                              User
Assuming newly
registered alternative
is efficacious, user
impacts will range from
$3.6-85.0 million
annually in increased
nematode control
costs and losses in value
of production.  In
addition, up front
capital investiments
of $3.2-$6.9 million
will be necessitated.

If newly registered alternative
is not efficacious, user impacts
of $15.8 million annually in
increased nematode control
costs and losses in value
of production are anticipated.

In either case, since pineapple
companies are vertically in-
tegrated, additional losses
in profit may be incurred
at processing and marketing
levels due to reduced production
and increased average costs per
unit.
                                                           Market/Consumer
Due to the oligopolistic
nature of pineapple
industry, any market/
consumer impacts
cannot be quantified;
however, substantial
losses in production
could result in prices
increases at the consumer
level, if alternative are
not adequately effective.
                                                                                                                                  CO
$26.4-$42.8 million in increased
nematode control costs and
losses in value of production.
At $26.4 million, capital
investments of $6.9 million
would be incurred by pineapple
growers.
Sources:  See Additions to Bibliography (Economics)

-------
                                85




The cancellation of EDB for use on soybeans would result in



increased nematode control costs of $7.0 million annually for



soybean growers.  The 1981 value of U.S. soybean production was



approximately $13 billion.  Cancellation of EDB for use on peanuts



would result in increases in control costs of $4.1 million and



reduced value of production estimated at $3.8 million annually.



1981 value of U.S. peanut production was approximately $1.0



billion.






Data submitted to the Agency indicates that the magnitude of



the economic impacts are most likely dependent on the efficacy



of Nemacur®, a newly registered systemic nematicide.   if Nemacur®



performs  adequately, pineapple growers would experience annual



losses due to increased nematode control costs and losses in



value of  production of $3.6-$5.0 million annually while incurring



capital costs of $3.2-$6.9 million.  If Nemacur® is not



efficacious, economic impacts could increase to $15.8 million



annually  in losses in value of production and increased nematode



control costs.   These impacts are large relative to the value



of pineapple production; 1981 value of Hawaiian pineapple



production was estimated at approximately $90 million.





Telone II®, a preplant chisel injected nematicide, can be



applied to Hawaii pineapple for control of the root-knot



nematode, Meloidogyne javanica (Treub).  EDB is applied for



control of the reniform nematodes, Rotylenchulus reniformis



(Linford).  If Telone II®, without Nemacur®, provided nematode



control equivalent to EDB, growers would experience increased



nematode  control costs of approximately $0.5 million annually.

-------
                                86




However, EPA has not received documentation that Telone II®




alone can achieve comparable efficacy to EDB.






         2•    Stored Grain Fumigation






The conclusions reached following the benefits analysis of EDB




fumigation of stored grain in the PD 2/3 concluded that alternatives




were available.  EDB for this use would probably be replaced by




the carbon tetrachloride/carbon disulfide (CT/CD) and/or aluminum




phosphide.  Other potential alternatives were identified but are




not anticipated to capture a significant portion of the liquid




grain fumigant market in the near future.  Carbon dioxide has since



been registered for this use,






Hopes Consulting, Inc. (Hopes) submitted a detailed evaluation




(Comment #26) of the benefits derived from the use of EDB to




control insect pests of stored grain.  Additional information




was also provided by the Douglas Chemical Co.  (Douglas) and




by the U.S.  Department of Agriculture (USDA).  The comments




and additional information are discussed topically.






   a.   Revised Estimate of Rate and Frequency of Application




Using methodology similar to that used in the Agency's




Preliminary Benefit Analysis, Hopes estimated the economic




impact of cancelling EDB.  Hopes estimated an increase in




user costs of more than $56 million as compared to the




Agency's estimate of a possible slight decrease in user

-------
                                 87




costs.  Differences in rates and frequencies of application




assumed in the PD 2/3 and by Hopes account for the large




variance in estimated user impact.  The Agency's calculations




were based on rates and frequencies of application from




product label data and State and Federal pesticide use




recommendations for EDB and CT/CD.  Hopes used an application




rate for CT/CD which is double that of EDB and also estimated




that CT/CD would be used twice as often as EDB (e.g., four




times annually versus twice annually for EDB).  Douglas, a




major producer and distributor, indicated that currently, only




one annual EDB treatment occurs at a typical rate of about 2




gal/1000 bushels of grain.  These data more closely approximate




the Agency's original estimates and indicate that usage




rates for EDB may have declined slightly.  The Agency believes




that the Douglas data approximate current label information




and are more indicative of current use than are the Hopes data.






   b.   Additional Estimates of Volume for Grain Fumigation



Hopes estimated approximately 4.56 million gallons of EDB




formulations were used to treat stored grain.  The Agency




estimated approximately 1 million gallons were used.  The




USDA provided preliminary results of a study titled "An




Interim Analysis of EDB Use in On-Farm Storage."  This



analysis indicates 57,492 pounds of EDB were used on farms.




Douglas estimated on-farm use of the active ingredient EDB,




to be 202,000 pounds in 1981 and that liquid grain fumigants



containing EDB approximates 375,000 to 425,000 gallons annually,

-------
   c.   Conclusions On Stored Grain Fumigation Use





Based on available use recommendations and label data, the



Agency believes that Hopes has overestimated the rate and



frequency of applications of both EDB and CT/CD.  As a result



of overestimated usage, Hopes has also overestimated the



benefits attributed to EDB usage.



The Agency believes current usage of EDB and CT/CD is probably



less than typical usage levels in the mid to late 1970's when



the PD 2/3 benefit analysis was conducted.





         3.   Spot Fumigation of Grain Milling Machinery





EDB has been used for more than 20 years as a spot fumigant to



control insect pests that infest flour milling equipment.  The



only alternative spot fumigant is a mixture of 75 percent



ethylene dichloride and 25 percent carbon tetrachloride; however,



the dosage required for this pesticide is so large that the



milling industry does not consider it to be a viable broad scale



alternative.  The overall results of using this substitute



include the application of a much greater quantity of pesticide,



a higher cost for fumigation, greater labor costs, and more



down time for the mill.






The Agency also determined that the most likely alternative to



EDB for spot fumigation is general fumigation of the entire mill



spaces.  Methyl bromide (MB)  and aluminum phosphide (AlP) are

-------
                                89



currently the two most commonly used general space fumigants.  The




Agency projected in the PD 2/3 that at least 3-5 additional general




fumigations per year would be required to obtain control.  Based




on a comparison of current practices versus the use of alternatives,




the PD 2/3 indicated an increase in costs of $2.9 to $6.0 million




annually for use of 3 and 5 additional general space fumigations




respectively.  The Agency has received numerous comments




regarding the viability, efficacy, and costs of alternatives




to EDB for controlling insects which infest flour milling



machinery.  Commenters provided substantive data for four




broad issues: a) EDB alternatives and their efficacy, b) EDB




usage estimates, c) frequency of application,  and d)  EDB spot




treatment costs versus the costs of general fumigation.  The



comments are discussed below and analyzed in detail in a separate




document titled "Analysis of Comments on Proposed Action on




EDB Use in Spot Fumigation of Milling Equipment" (Holtorf and




Ludvik, 1982).






              a.   Efficacy of Alternatives






The Agency's assessment of the efficacy of methyl bromide




(MB)  or aluminum phosphide (AlP) as alternatives to EDB was




questioned by Ferguson Fumigants Inc. (Comment #22)  and by




USDA.  Ferguson argued that neither MB nor AlP are pupacidal




and that only EDB alone is efficacious in controlling tolerant




life stages such as the egg and especially the pupal  stage.




However, the Agency is aware of a number of studies  which




present data that demonstrate the pupacidal activity  of MB or

-------
                               90




AlP.  These studies include:  Kazmaier and Fuller (1959),




Krohne and Lindgren (1958), Lindgren et al., (1958), and




Munro (1969).  The Kazmaier and Fuller study also indicates MB




and AlP are ovacidal (effective against egg stage).   Therefore,




the Agency rejects the contention that EDB is the only efficacious




compound for controlling all life stages.






              b.   Usage Estimates




Ferguson Fumigants Inc., the Miller's National Federation, and




the USDA all submitted data which indicate that the overall




quantity of EDB usage in spot fumigation may be less than that



estimated by the Agency in the PD 2/3.  The analysis of benefits




presented in PD 2/3 was based on the assumption that




all mills used EDB spot treatments.  Although the usage




estimates provided by the commenters were not consistent, the




new data indicate that approximately 75 percent of the wheat




flour milled in the U.S. is produced in mills treated with




EDB.






Because the Agency's original assessment was incorrect, the



overall benefits of EDB use in flour milling are less than




were estimated in the PD 2/3.  Consequently, because fewer




mills than previously estimated use EDB, the relative costs of




substituting general space fumigation for spot treatments were




also overestimated in the PD 2/3.






              c.   Frequency of Application






The single, most often discussed issue is that of the frequency




of general fumigations necessary to maintain insect control.

-------
                                91




Most comments  indicated that 4 or  5 general  fumigations would




not be adequate to maintain effective control.  Commenters




stated that general  fumigation would have to be practiced




on the same schedule as spot fumigation or about 12 to 14




fumigations annually.  One commenter (Comment #44) indicated




that general fumigation may be practiced 8-10 times annually.




It was indicated in  the preliminary benefit analysis, however,




that frequent  (i.e.  12-14 per year) general fumigations would




be prohibitively expensive, and that the industry would probably




adopt a strategy which would utilize fewer general fumigations.






The Agency received comments that  even 4 or 5 general fumigations




would be prohibitively expensive (Comment #43), and that general




fumigation would not be conducted  even 4 or 5 times per year




to replace spot fumigation (Comment #1A).   Though arguing




that 4 or 5 general space fumigations would not substitute for




spot fumigations,  the commenters did not provide the Agency




with data to support their claim that general fumigation is



ineffective.






The Miller's National Federation (MNF)  surveyed 27  member firms




which produce approximately 80% of  the  wheat flour  milled in




the United States  (Henwood, 1982).   Their  survey,  in addition




to a previous MNF  survey on total  usage  of  EDB,  provides the




most definitive source of  information available regarding the




frequency of  application of EDB in  the  milling  industry.   The




Henwood study indicates that,  of the mills  surveyed,  those




treated with EDB produce about  75%  of  the daily capacity of the

-------
                                92



industry and that each mill is treated with EDB an average of



about 10 times per year.






In summary, the comments submitted to the Agency indicating



that current usage of EDB for spot fumigation is less than



estimated in PD 2/3; that EDB is used less frequently than



indicated in PD 2/3; and that the amount of flour exposed to



treatment is less than indicated in the PD 2/3.  The Agency



recognizes that general fumigations will require more diligence



in regard to sealing mills and maintaining duration of exposure



to obtain control of insect pests in flour mills.  The Agency



does not believe that the milling industry will be required to make



12-14 general space fumigations annually rather than 4-6 general



fumigations as indicated in the preliminary analysis.  The Agency



has concluded that the information submitted by commenters does



not adequately support their contention that alternatives to



EDB would not be used, or the Agency's preliminary benefits



analysis failed to consider appropriate facts.

-------
                                 93



              d.    Treatment Costs




Three respondents provided mill-specific information on




comparative costs of spot vs. general fumigation in their




mills.  The Ohio Department of Agriculture (Comment #14)  noted




that the cost of a complete general mill fumigation is about




$5,400 compared to a spot fumigation cost of $600.   These costs




did not include the charge for "down time".






The Hawaiian Flour Mills, Inc., (HFM) (Comment #29) provided a




detailed breakdown of the costs of general and spot fumigation.




HFM indicated that their fumigation costs would increase  from




$8,774 annually using EDB to $24,079 annually for general




fumigations using aluminum phosphide and in addiiton they




would lose of 4 to 5 production days per year.  HFM did not




provide data regarding the frequency that EDB was currently




being used in their mills nor were data submitted on how




frequently general fumigation or other controls would be  used




if EDB were cancelled.  North Dakota Mills and Elevator (Comment




#27) indicated that each general fumigation costs approximately




$4,800 while each spot fumigation costs approximately $500.






The National Pest Control Association Inc. (NPCA) (Comment #44)




submitted estimates of the impact of cancelling EDB in wheat,




rye, and corn mills.  NPCA estimated that use of general  mill




fumigation in the wheat and rye milling industry would increase




treatment costs by about $12.8 million annually ($43,000  per mill)




if 8 general fumigations replaced EDB spot fumigations.  For corn




mills, NPCA estimated an increase of $2.6 million annually




($28,000 per mill) again, basing their estimates on 8 annual

-------
                                94




general fumigations.  Ferguson Fumigants, Inc. (Comment #33)




provided a 1978 analysis of the cost of replacing EDB with




general space fumigation.  They estimated that the cost to the




milling industry of using methyl bromide would be approximately




$24.1 million annually as compared with about $1 million for




an EDB program.  This analysis assumed that general fumigation




would replace spot treatment on a one-to-one basis; therefore,




13.7 general fumigations per year would be required throughout




the entire milling industry.






The NPCA 1981 estimate of $15.4 million (using 8 general




fumigations) is about double the Agency's preliminary 1978




estimate of approximately 7.8 million (using 5 general space




fumigations including the one normally scheduled).  The Ferguson




Fumigants Inc.   estimate of $24.0 million assumed 13.7 general




fumigations.  The NPCA and Ferguson Fumigants estimates appear




to be based on the cost of using general fumigation throughout




the entire milling industry.  However, EDB usage is apparently




not as widespread as previously believed and could be 25% to




50% less than was estimated in the PD 2/3 (i.e., 50% to 75% of




the mills use EDB).  These estimates of the cost of space




fumigation are therefore overestimated to the extent that they




rely on the assumption that cancellation of EDB would impact




the entire milling industry.

-------
                                 95



Although data are not available  to estimate with precision the




frequency of usage of general fumigation in place of spot




treatments, The NPCA 1981 estimate (based on 8 general fumigations)




of $15.4 million, certainly is not unreasonable and shows relatively




close agreement with the PD 2/3  estimate.  Allowing for inflation



of input factors during the 1977 to 1981 period.  The estimated




impacts presented in the PD 2/3  would be higher and could be




expected to converge with the NPCA estimates, especially if the




PD 2/3 estimates were recomputed based on 8 general fumigations.




The Agency does not believe that the milling industry, on average,




will adopt a schedule of 13.7 general fumigations as suggested




by Ferguson Fumigants Inc.  The  high cost and added inconvenience




of general fumigation will force millers to limit the number of




general fumigations to a schedule which is less frequent than




is currently practiced with the  spot materials.






              e.   Conclusions on Spot Fumigation






Comments submitted in response to the PD 2/3 generally agreed




with the Agency's preliminary assessment of benefits that




cancellation of EDB use in spot  fumigation will increase the



cost of insect control in the milling industry.  The comments




stressed that in comparison to whole mill fumigation (with MB




or AlP), EDB spot treatment is much more convenient, more




effective in certain situations  and is much less expensive.




The comments did not provide new data, information, or arguments




which could persuade the Agency  to alter its preliminary findings.

-------
                                96




   4.   Quarantine Fumigation of Citrus ,  Tropical Fruits ,  and




              Miscellaneous Fruits and Vegetables






The U.S. Animal and Plant Health Inspection Service (APHIS)




and the states of Texas, Hawaii, Louisiana, California and




Arizona require the use of EDB in their fruit fly quarantine




programs.  Prior to March, 1982, the Japanese government  also




required EDB fumigation as the only approved method of treating




all fresh citrus and tropical fruit imported from the U.S.  In




PD 2/3, the Agency estimated that major economic impacts  would




occur if EDB were cancelled for these uses.  Depending on the




acceptability of cold storage and/or irradiation as viable




technological alternatives, the magnitude of adverse economic




impacts may be reduced.  Several comments pertaining to the



estimates of benefits were submitted.  Most of the comments




claimed an increased need for EDB due to recent infestations




of the Mediterranean and Oriental fruit flies in California.




A second major concern  is the viability of gamma irradiation




as an alternative to EDB post-harvest fumigation.






Several commenters (Comments #3, ttlOA, #20) argued that




the 1981 Mediterranean  fruit fly infestations in California




have increased the need for, and thus the benefits derived




from the use of EDB fumigation.  Due to the uncertainty associated




with predicting the extent and/or intensity of future medfly




infestations, the Agency has not modified the original estimates




of benefits.  In March, 1982, the Agency was informed that the




Japanese government has approved the use of cold storage, for




those commodities adaptable to the procedure (oranges and




grapefruit)  infested with the Mediterranean fruit fly.

-------
                                97




The USDA/APHIS is currently in the process of developing cold




storage procedures for treatment of the Carribean fruit fly




(Helms, Personal Communications, 1982a and b).  As of August



1983 experimental data indicated that Carribean fruit fly




eradication is technologically feasible using cold storage for




shipments to Japan (Vlier, personal communication).  Although




results indicate technological feasiblity, it is uncertain




whether Japan will accept the cold storage quarantine method




for the Carribean fruit fly.






With regard to interstate citrus shipments and quarantine



regulations, some progress has been made in reducing the need of




EDB treatments to Texas citrus.  Over the past 2 harvest seasons



(1981/1982, 1982/1983) APHIS has implemented a Mexican fruit fly




certification program in Texas.  The program measures fruit fly




populations and determines whether those populations are low




enough to allow interstate shipment of citrus without quarantine




treatment.  Theoretically, the certification program will reduce




the need for quarantine treatment; however, since fly population




levels can increase, not all fruit in all Texas locations and




harvest periods can be certified as pest free.  Therefore,




alternative controls for EDB fumigation must be available as




back-up to the certification program to avoid adverse economic




impacts.






The U.S. Department of Agriculture (USDA), the Hawaii Papaya




Industry Association (HPIA), the Hawaii State Department of




Agriculture (HA), and the California Citrus Quality Council




(CCQ) all indicated that they do not consider gamma irradiation

-------
                                98




to be a feasible alternative by July of 1983.   However, the




Pood and Drug Administration has published an Advance Notice




of Proposed Rulemaking 46 FR, 18992-3) which initiates the




process to establish criteria for evaluating petitions for




irradiated foods.  In addition, FDA has authorized the use of




irradiation in emergency situations to control Mediterranean




fruit fly in infested fruit (Hayes, 1981).  The Agency has now




extended the phase out of the EDB quarantine use until



September 1, 1984, to provide additional time to implement



alternatives.






The Hawaii Papaya Industry Association (Comment #23)  indicated




that approximately 91 percent of the Hawaiian papayas were




marketed as fresh fruit with approximately 74 percent being




shipped to the mainland U.S.  and to Japan.  The HPIA contends




that a cancellation of EDB, without suitable alternatives,




would totally destroy the Hawaiian papaya industry.   The Agency




acknowledges that a lack of acceptable control measures would




have serious negative impacts on the papaya industry.  However,



if FDA approves the use of irradiation for papaya and if irradiation




is acceptable to Japan, the economic impacts on the Hawaiian




papaya industry are expected to be less substantial.






The Hawaii State Department of Agriculture indicated that




currently, papayas are grown and fumigated on at least three



different islands and that construction of irradiation facilities

-------
                                 99



at each location would be economically unsound.  The Department




also indicated that a central irradiation facility would increase




shipping costs and drastically  increase the amount of handling.




Although the Agency agrees that  such circumstances could result




in economic impacts, the State  of Hawaii did not submit data




which would allow the Agency to  quantify these concerns.




The Agency has further determined that papayas are fumigated



on only two islands, Hawaii and  Kauai, and that most of the




fumigation occurs on Hawaii.






The Agency is aware that several commercial enterprises are in




the process of designing and building gamma irradiation facilities




in several locations in California and Florida.  In light of




the pending FDA rulemaking and  the proposed construction of




gamma irradiation facilities, the Agency believes that gamma




irradiation should be considered as a potential alternative




to EDB fumigation in the future.






The USDA asserted that a cancellation of EDB would weaken the




trade positions with Mexico, Haiti, Israel, and Morocco.  USDA




contends that suitable alternatives to EDB are not available




for citrus and tropical fruits  originating in these countries.




However, the USDA did not provide data demonstrating that




cancellation of EDB would adversely affect those trade positions




with Mexico, Haiti, Israel, or  Morocco.  EPA,  however,




concludes that these exporting  nations will have available the




same alternatives for EDB use as United States growers.
418-574 O - 83 - 8

-------
                                100



                   5.    Felled Log Fumigation






In PD 2/3, the Agency estimated 68,000 logs annually in Colorado



and North Dakota were fumigated with EDB to control several



species of bark beetles.   When the PD 2/3 Benefits Analysis



was performed, endosulfan and lindane were registered alternatives,



but the Agency did not consider lindane as a viable alternative



due to its status as an RPAR chemical.  Although endosulfan was



considered less effective than EDB or lindane, the Agency



concluded its use for bark beetle control could reduce treatment



costs as much as $36,000  annually (1978 nominal dollars).



The Forest Service, Larimer County, Colorado, provided data on



the benefits of using EDB on felled logs.  The process, from



felling through treatment, costs $10 to $100 per tree and



cannot be conducted on a  scale large enough to effectively



reduce bark beetle populations.  In addition, the Forest



Service indicated that control costs in conjunction with adverse



health effects have resulted in a decision to suspend further



use of EDB in bark beetle control projects in Larimer County.






The U.S. Department of Agriculture responded to the PD 2/3 by



noting that endosulfan is not specifically registered to control



the mountain pine beetle and that the Department was not aware

-------
                                101




of efficacy data comparing endosulfan with EDB.   The USDA




further stated that lindane is actually the compound of choice,




as an alternative to EDB for treatment of the mountain pine beetle,






The Agency did not receive sufficient additional information




that would allow an improved estimate of the benefits derived




from the application of EDB to felled logs.  The Agency, however,




is reevaluating its preliminary decision to cancel the use of



lindane on felled logs, and is likely that this  preliminary




decision will be reversed.  Thus, lindane will probably be




available as an alternative to EDB.  In sum, the economic



impacts of an EDB cancellation would be minimal  for this use.






         6 .   Minor Uses




The Agency did not receive substantive data for other uses




of EDB.  Therefore, the Agency has no basis to change the




original assessments as contained in PD 2/3.  Consequently,




the PD 2/3 estimates of benefits resulting from the following




uses of EDB are not revised:  termite control, fumigation of




stored beehive supers and honeycombs, fumigation of stored




clothing, furniture, and quarantined plants, soil and grass sod.

-------
                                102



V.  Summary and Conclusions






The Agency has determined that EDB is a potent animal carcinogen,



a mutagen to a variety of plant and animal cells, and is capable



of producing adverse reproductive effects in animals. EDB is



potentially capable of producing these adverse health effects



in humans at measurable levels in the human diet, in the



ambient air of application sites, and in the ambient air of



facilities where treated commodities are handled and stored.



The Agency has carefully weighed the human health risks against



the economic benefits for each use of EDB.  The Agency has



determined that no remedial changes can be made in the



registration of EDB for most uses to avoid unreasonable adverse



effects on human health.






The Agency further has determined that continued registra-



tion of EDB as a soil fumigant poses an imminent hazard during



the period in which administrative hearings would delay the



the cancellation of these registrations.  Continued use of the



chemical during these hearings would contribute additionally



to an already serious hazard.  Emergency suspension is the



only mechanism available to the Agency to cease immediately



the use of a pesticide.  The Agency has also determined that



an emergency exists resulting from the soil fumigant use of



EDB such that an emergency order should be issued immediately



suspending these registrations.

-------
                                103

The Agency has further determined that the registrations for

use as a soil fumigant, grain fumigant, spot fumigant and

felled log fumigant should all be cancelled.  The Agency,

however, has determined that the cancellation of EDB products

used for fumigating citrus, tropical fruit and vegetables

should not be made effective until September 1, 1984, in order

to allow time for alternatives for this use to become available

on a commercial scale.  The Agency has decided to continue

registration of EDB products used for the minor uses of

termite control, fumigation of beehive supers and honeycombs,

vault fumigation and Japanese beetle control, only if specified

changes are made on the labels to reduce the risks presented

by these uses.


A summary of the Agency's final regulatory decision on EDB is

presented in Table 16, and the details of the regulatory decision

are presented in the Federal Register Notice announcing the

availability of this document.


    TABLE 16—SUMMARY OF FINAL REGULATORY DECISION ON EDB

Soil Fumigation                Cancellation effective 30 days
                               after this Notice; Emergency Sus-
                               pension effective immediately.

Stored Grain Fumigation        Cancellation effective 30 days
                               after this Notice.

Quarantine Fumigation of       Cancellation effective September 1,
  Citrus, Tropical Fruits      1984.
  and Vegetables

Spot Fumigation of Grain       Cancellation effective 30 days
  Milling Machinery            after this Notice.

-------
                                104
Felled Log Fumigation
Termite Control, Beehive
  Supers, Vault Fumiga-
  tion, Japanese Beetle
  Control
Cancellation effective 30 days
after this Notice.

Continued registration with label
changes imposing use restrictions;
require submission of monitoring
and use data.

-------
                             References and Citations

 Aafjes,  J.H.r J.M.  Mais, and Schenck.  1980.  Fertility of Rats with .Artificial
    Qligozcospermia.  J.  Reprod. Fertil. 58(2):  345-352.

 Adams, C.F.   1975.   Nutritive Value of American foods, in; Agriculture Handbook
    No. 456.  p. 104. CSDA. Wash., D.C.

 Amann, R.P.   1981.   A Critical Review of Methods for Evaluation of
    Spermatogenesis  from  Seminal Characteristics.  J. Androl.

 Amir,  D.  1973. The Sites of the Sperraicidal Action of Ethylene Dibromide in
    Bulls.  J.  Reprod.  Pert.  35(3):   519-525.

 Amir,  D.,  and  E.  Ben-David.   1973.  The Pattern of Structural Changes Induced
    in  Bull Spermatozoa by Oral or Injected Etnylene Dibronide (EDB),  Ann.
    Biol.  Anim. Biochem.  Biophys. 13(2 )r  165-170.

 Anir,  D.,  and  ¥.  Lavon.   1976.  Changes in Total Mitrogen, Liponroteins and
    Aaino  Acids in Epididytnal and Ejaculated Spermatozoa of Bulls Treated
    Orally with Ethylene  Dibrcmide.  J.  Reprod. Pert. 47(1):   71-76.

 Banbury Report 1  Assessing Chemical Mutagens:  The' Risk to Humans.  1979.
    V.K. McElheny  and S.  Abrahanson, ed.,  Cold Spring Harbor Laboratory, po.. 319—
    328.

 Barak, A»V.  ard P.V. Harein.   1931..  ^predictable Economic Thresholds of Farm
    Stared Grain in  Ilimesota-   Univ..  Minn., St. Paul.  !!M.

 Barfield,  A.r  et  al. 1979..  Pregnancies Associated with Sperm Concentrations
   Below 10 miTTion/fal in Clinical  Studies of Potential Male  Contraceptive
   Method, ffonthyl  Eepot Hedkoxyprogesterone Acetate and Testosterone  Esters.
   Contraception  20(2)t   121-127.

 Bierer, B.W,,  and C.I*. Vlckers.   1959.  The Effect on  Bjg  Size and Production
   of Fungicide-treated  Fumigated Grains  Fed to Hens.   J.A. \fet.  Med.  Ass.
   134(10)j  452-453.

 Bcndi, A., E.  QlonucJcL,  and M.  Calderon.   1955.  Problems  Connected with
   Ethylene Dibrcmide Fumigation of Cereals.   II.   Feeding Experiments with
   Laying  Hens,   J. Sci.  Food  Agric.  6:   600-612.

 Brem, H., et al.  1974.  The Mutagenicity and ENA-modifying Effect of
   Haloalkanes.   Cane. Ttes. 34:  2576-2579.

Buselmaier, W., et al. 1972.  Mutagenicity Studies with Pesticides in  the Host-
   inediated Assay and the  Ebminant-lethal Test in  the  Jfeuse  (In German).
   Bid.  Zbl.  91s  311-325.

CfiG 1982a.  Transmittal of Cements on  the  Draft Ethylene  Dibrcmide Position
   Document 4.  Menorandon to J. Panetta  from R.E. McG.ughy,  CAG.  29 March
   1982.

    1932b.  Rebuttal Analysis  for the Carcinogenicity  of Ethylene  Dibrcmide.
   Seport dated 15 March 1982, 9 pp.

                                      105

-------
                                      106

Caylor, J.F., and C.K.. Laurent.   1°60.   The Effect of a Qrain Fumigant on Egq
   Siae of White leghorn Hens.   Raul try Sci. 39t  216-219.

Cohen, S. (reportinq Man'q.)  Ihdated.   Study of (Voundwater Contamination by
   Pesticides in California.   EPA Proj. No. HED 80-5.  Environ.  Prot.  Agency.
   Wash. , D.C.

Cotton, R.T.  1956.  Pests of Stored Qrain and Grain Products.   Burgess Publ.
   Co. , Minneapolis, MN.  po.  94&112.

Courtens, J. , D. Amir, and J.  Eurand.   1980.  Abnormal Spermioqenesis  in Bulls
   Treated with Ethyiene Dibrcmide:  An Ultrastructural and Ultra  Cytochemical
   Study.  J. DLtrastruc. Res.  •»!:   103-115.

nay, ^.R. l<»81(a).  Data Evaluation Record and Exposure Analvsis.   Evaluation
   of untitlefl study information from Capt. n. Strichnan, rFAP.  .15 Sept. PI.
   Environ.  Rrot. Aqehe,r.  T-Zash.,  D.C.
Day, H.P. 1^81 fb).  Data E«;aluation Ftecord and Fxposure Analysis,  * Dec.
            Rrot. ^72ncy.  T-feish.,  D.f*.
Day, H.P. 19^1 (c).  Data Evaluation  Psccrd and Ficposure Estinate.  .*> Pec.
   Environ, ftrot. Agency.  Wash.,  D.C.
r-yr !!.?.. If1?!^).  Data Evaluation  :?^ccr's airr rxpcsirs .^nalvsis.  Evsl'uation
  'of >rrr«« Health P'aluation anri r'ech.  Assist.  Pntv "o. "^ ^"-7..  June
       .  Environ. Rrot. Aqency.  T-fesh., D.r.
ri^cmnaso, n.J. 1982. letter to J.T».  Miles with attached ARA Tech Suocort,
   Region VTII request  for  EDB monitorina data.  U.S. Dept. Labor,  rccup.
   Safety Health Admin. Billings,
Edwards r ^. , Jackson, and  A.R.  Jones*  1970.  Studies with Alleviating  Esters —
   II.  A Chemical Interpretation throuqh Metabolic Studies of the
   An ti fertility Effects of  Ethyiene  Dinethanesulphonate and Ethyiene
   Dibromide.  Biochem. F»iarmacol.  19:
El jack, A. H. , and P.. Hrudka.   19"79.   Pattern and Dynamics of Tteratospermia
   Induced in Earns by Parental  Treatnent with Ethyiene Dibromide.  J.
   Oltrastruc. Res. 67:   124-134.

EPA.  19^7.  Rebuttable Presumption Against Fteqistration and Continued
   Reaistration of Pesticide Products  Containina Ethyiene Dibrcmide  (EDB).
   Fed. Reg. 42(240):  63134-63161. Dec. 14. Environ. Prot. Aaency. Wash.,  D.C.
?PA.  10
-------
                                      107

Freund, .M., and B. Carol.  1964.   Factors Affecting Heamocytcmeter Counts of
   Sperm Concentration in Human Semen.   J. Reprcd. Fertil. 8:  149.

Fuller, H., and G. Morris*  1963.  The  Comparative Toxicity of Ethvlene
   Dibrcmide when Fed as Fumiaated Grain and V*en Administered in Single Daily
   Doses.  Poultry Sci^ 42:   508-514.

Glass, R.  1978.  Chapter 19: Infertility in "Reproductive Endocrinology"
   edited by S.C. Yen, and R.B» Jaffe.   W.B. Saunders Ccmpanv, Philadelphia,
   PA. pp. 398-420.

Gomes, W.R. 1977. Chapter 22: Pharmacological Agents and Male Fertility in "The
   Testis" edited by A.D. Johnson  and W.R. Gomes, volume IV: Advances in
   Physiology, Biochemistry,  and Function, Arrademic Press, New York pp. 605-628.

Gray, P.H. (Sect'v. SAP> 1°«1.  Review  of Prelinin^rv Notice of rte termination
   Concluding the Rebuttable  Presumption *aainst Registration fRPAR) of
   Pesticide Products Containing Fthv.lene Dibromi^e (EDR). ^op.   Prientific
   ArHrisory Panel.  Fnviron.  Prot. Aaency. Wash., D.C.
Oriffith, J. , R. **;ath, and F. Da^rido.   I0"7**,   ^pematoa^nesis in Agricultural
   t'ferkers Potentially Exposed to  Hthylene Dibrcrai^e (FTP).   Environ,
   Agency.

Hayes, A.A.,  1931. tetter  to  Sbelhy ^  °re'.-cr  rr-3q?r^ira FT-^ rositicn en
   irradiation of producei 1" Seoterrher.
Hayes, W.J. 19^5. Chapter 5:  Toxicology of Pesticides, The Williams and
   Wilkins Co», Baltimore*

Henwcod, R.M. 1982.  Letter to Roger  C.  Hoi tor f Tregarding survey of HUB usage
   in spot fumiqationl 30 March.

Bertlein, P. and M.R. Hagadone.   19B1.   Rsport on Investigation of F.thylene
   Dibrcraide Ftroiqation Chambers  in the  State of Hawaii for the Paoaya
   Administrative Committee, Honolulu, Hawaii.

Seller, C.G. , and Y. Clermont.  19P4.  Kinetics of the Germinal Epithelium in
   Man.  Recent Progr. Hormone Res. 20:   545.

Holder, J.W.  1979.  FJER Levels in Conmodities  as a Result of ETB Fumigation.
   Memorandum:  4 Dec. to M. Williams.   Environ. Prot. Aqency. Wash., D.C.

Holder, J.W.  1980.  Re-evaluation of the TTotal Dietary Burden of Ethvlene
   Dibromide Used as a Soil, Grain, Grain Machinery, and Fruit Fumiqant.
   unpub. report.  Environ. Prot. Aqencv.   Wash.,  D.C.

Holtorf, R.C. 1982.  Evaluation of PT> 2/? Comments Regarding fVain Ptoraae Uses
   of ETR.  memorandum:  15 Jan.  to J. Panetta.  Environ.  Prot. Agency. Wash. ,
   D.C.

Holtorf, R.C. and G.F. Ludvik.  19S1.  Grain Furaiaants:  An Overview of Their
   Significance .to U.S. Agriculture and  Commerce and Their Pesticide
   Regulatory "implications.  Benefits ft  Field Studies Div., Off. Pesticide
   Programs, Environ. Prot. Agency.   Wash., D.C.

-------
                                      108

Hoi tor f» R.C. and G.P. Ludvik.  1982.  Ethylene  Dibromide — Analysis of
   Comments on Rroposed Action on EUR Use  in  Spot Fumigation of Milling
   Equipment,  memorandum: 19 Jan. with  attachment to J. Panetta. Environ.
   Prot. Agency, Wash., D.C.

ftrudka, F., and A.R. EL jack.  19*79.  The Effect of Ethylene Difcrcmide on
   Differentiation of the Acrosome, Nucleus,  and Transicent Nuclear Appendages
   in Ram Sparmatdds.  J. Ultrastruc. Res. 67:   135-151.

Indust. Health Enqin. Asscc. , Inc. 19"**.   Fumiqation Air Monitoring for
   Ethylene Dibronide. by G.W. Wiutson.  IREA  Proj. 29 Mar. 1978.

Jensen, J.J&  1981 (a).  Data Evaluation Record  of:  Industrial Ryaiene Survey
   for Airborne Ethylene Dibromide Curina  Soil  Fumigation, Hcrostead, Florida,
   Octobsr 1980 by How Chemical Co.  Fnviron. Prot. Aqenrv. Wash., n.C.
Jensen, J.K.  19*^1  ^b).  Data  FValuation  Pecord of:  Industrial FVoiene Survey
   fior Airborne Ethylene Dibrcmi'*e  Dirincr coil Funiaation at the ncnesteari,
   Dade County, Florida, Mb\'.,  19"79. by Hbt7 Chem. Cb.  Fn^iron Prot. Aaency.
       ., D.C.
Kazmaier, H.E. and P.G. FuUcrr.  1950.   Ethylene rMbrcnide:  ?*ethvl Prcnide
   Mixtures as Pumiqants Against the Confused  Flour Beetle.  J. Eton.
   Entctiol. ?2r
    , J.P_r etal. 19?/);  An Plectrcn -C?T3ture Cas Orniotcrrrarhic ."ether" for
   Dstermnation of ^sidtras of  l,2-ri.hrcro9thare in Fcwirrstsr? ^rsnjfruit. J.
   Agric. Food Chen* 23 i
Wrohne, H. E. and D.L. Linrtaren.  1958.   Susceptibilitv of Life Stages of
   Sitophilus orv7:a to Various Fumiqants.   J.  Econ. Entcrool. 51: 157-8.

Lavy, '".L., et:_al. 1930.  Hbcposure  Me=jsurenvents of Applicators Snraying  f2,4,5-
   *!richloroohenoxy)acetic Acid  in  the  Forest.   J.  Agric. Food Chem. 28:  626-
   f29.

Levin, R., et^l. 1981.  Analysis of ^productive Experience at Ethyl's
   Magnolia, Arkansas Plant.  Report submitted  by Ethyl Corparation, Baton
   Rouge, Louisiana.

Lindgren, D.L. , L.E. Vincent, and R.  G.  Strong. 1958.  Studies on Hydrogen
   Phosphide as a Fumigant.  J.  Econ. Entcmol.   51: 900-3.

Loshiavo, S.R. 1%0.  Effects of Low Coses of Ethylene nibrcmide on Some Staaes
   of the Confused Flour Beetle, Tribolium confusunu  J. Econ. Entcmol.
   53: •'62-7.

Ludvik, G. F. 19Rlfa>.  Mountain Pine *5etle Control with EDR.  Memorandum:  8
   Ifec. to F.R. Day  Environ. Prot.  Aqency.   Wash., D.C.

Ludvik, G. F. 1981 (b).  Mountain Pine Peetle Control with FDR.  Mworandum:  16
   Dae. to J. Reinert.  Environ. Prot.  Agency.   t-2ash., D.C.
Ludvik, G. F, 1932 fa).  EDR Fumicjation of Papayas — Pac*sr and Annlicator
   Tines.  Memorandum:  to E. Saito.  25  Jan.  Environ. Prot. Agency. Wash. , n.C.

-------
                                      109

 Ludvik, G. F.  1982 (b).   PD 4 fbr EHf*.  ^morandum:  11  Fteb.  to U  Zygadlo.
    Ehviron.  Irot. Agency;  Washington, D.C.

 Ludvik, ft* F.  1982 (c).   ElB — Applicator Exposure as a Consequence of Papaya
    Fumigation.   Memorandum:  16 Feb. to E. Saito.  Environ.  Hrot.  Agency.
    Wash., D.C.

 Luttner, M.  1980 (a).  Revised ECBt  Citrus; Memorandum:  to  L. Zygadlo. 30
    July.  Environ.  Prot. Agency, Wash., D.C.

 Luttner, M.A-  1980 (b).   EEP Citrus Analysis.  Memorandum:  22  July to L.
    Zygadlo.  Environ. Prot. Agency. %sh., D.C.

 MadeV, K. T. et^al.  lers, in
           15 so.  MS  Cal. D?pt. Peer* "^ric.  Pacrsrento. Cal.
Maddv, 'C.T. etr_f?l_.  l°°l.fh^   Potential
   Osratcrs  Inr/o-b-fed  in the Uhloadinrr and Loadina of  Fruit  F-jniacited  with
    ecd, J.  l°^9.   F'zrther' n-«r=rr*.T3ricr!? en the ^le cf V'rrriccr^l3  ir;
   Infertility.   Fsrtil.  Ftsril.  2^:  s-''c.
f*acteod, J. , and  P..  Z.  Cold^ 1^53.  1*ie "ale Factor  in  Fertility and
   Infertility, VI.   Semen Cualitv anrJ Certain Other Factors  in  Relation to
   Ease of Conception.   Fertil.   Steril. 4:  10.
MacLeod, J. , and Y. Warn.   I0"7?.   Msle Fertility Fbtential  in ^rr^s of Seran
   Ouality: a Peview of the Past,  a ^tudy of the Rresent.   Fertil.  Steril.
   31 (2):   103-llfi.

Maddy, K.T. et jil.  19S2.   Rscent Studies and FV/aluations by the  California
   Cepartrent of Foort  and  Agriculture on ETS '.^rLth Ehrohasis  on nse  in
   Fumigation of Fruit.  Perort HS-956, Calif. Cent. Food Aqric. Sacramento,
   Calif.

Martin, R.n. 19PO.  Industrial  Hvgiene Survey for Ethylene  Dibrcmide Curing
   Soil Fumiaation, Homestead,  Cade Co. Florida. How Chem.  Co.,  "idland,  Mich.

Martin, R.H. 1982.  Letter (s Jan) to J. Jensen regarding Industrial Hvaiene
   cun'eys  to suotnrt  reqistration of Ere. fVTith attachments).   How Chem. Co.,
   Midland, Mich.
         , W.H.   1932.  Cost  Stonary of Disinfestation of  the  Florida Citrus
   Crop  throuah  taradiation.   U.S.  Cerenrtnent o^ Energy.

Mitte.lman, A.  19PO.  Mon-Dietary Fxrosure Analvsis  fbr  FTP  Uses and
   Environmental  Fate of FTB.   Unpub. report. 53 pp. Environ.  Prot. Agency.
   Wash., DoC.

-------
                                      no

Munro, H.A.U. 1969.  Manual of  Fumigation for Insect Control.  ESO Agricult.
   Studies No. 79.  FAD/UN Rone, p?  35,  294 and 295.

NCI 1980.  Bioassay of 1,2-dibrcmoethane for Possible Carcinogenicity.  Public
   Health Service, NIH, EHHS  Publication No. 80-1766.

Nelson, C.M., and R.G. Bunge. 1974.   Semen Analysis:  Evidence for Changing
   Parameters of Male Fertility Potential.  Fertil.  Steril. 25(6):  503.

NIOSH, 1977.  Criteria for a  Recommended Standard, Occupational Exposure to
   Ethylene Dibronide.  Stanford Research Institute, Menlo Park, California,
   NTIS SPB-276 621.

Nitschke, R.J., ejtal. 1980.  D-Week Repeated Inhalation Study on Ethylene
   Dibronide (EDB) in Male and  Femalw Rats.  Daw Chemical USA. 11 February 1980,

Pederson, J.R. 1979.  Sthylene  Dibrcnide as a Spot Fumigant in Flour Mills.
   •fercorandum:  to R.C. Holtorf USEPA from DPRA Feb. 23.

Sains, D.M. and J.t7. Holder.  1981.   Ethylene Dibrcnide Residues in Biscuits and
   Ccrr.ercial Flour.  J. ^ssoc. Off;. ,v«iL. Chsn. 54: 1252-1254.

R.E.A.G. 1S"32. Tas^cnse to Industry  "abuttals Concerning the reproductive
   2is/3 of 2T3.-  "srcrarr^ur!:   12 J^n to A. Garten. attachrr«r.t. rnvircn.
Pseves, ?.. A., ^t al. 1531.  :^:r.itcrir.g  Daring  Finigaticn C'.anrer
   Using Echylene 'Dibrcnide  (EDB).   Public.  HS-909,  11 pp.  Calif. Dent.
   Agric.  Sacramento, Calif.

Reinert, J. 1982(a).  EDB PD4.  Jfemorandum:  to  J.  Panetta. 4 Jan.  Environ.
   Protect. Agency.  Wash.,  D.C»

Seinert, J. 1932(b).  EDB PD 4:   Dietary Exposure  from Post Harvest Fruit
   Fumigation,  iremorandun:  to J. Panetta.   12  Jan.   Environ. Prot. Agency.
   Wash., D.C.

Reinert, J. 1982(c).  Memorandums   to J.  Panetta.  26 Jan Environ. Prot. Agency.
  Hash., D.C.

Saito, E.C. 1982(a).  Data Evaluation Record.   Evaluation of:  f-tanitoring
   During Fumigation Chamber Operations  Using Ethylene Dibrcmide (EDB).
   Environ. Prot. Agency. Wash.,  D.C.

Saito, E.C. 1982(b).  Data Evaluation Record.   Evaluation of:  Potential
   Inhalation Exposure of Forklift Operators in the  Unloading and Loading of
   Fruit Fumigated with Ethylene Dibromide  (EDB).  Environ. Prot. Agency,
   Wash., D.C,

Saito, E.C. 1982(c).  Data Evaluation Record.   Feb.  4.  Environ. Prot. Agency.
   Wash., D.C.

Saito, E.C. 1932(d).  EDB PD 4:   Inhalation  Exposure to Workers Resulting from
   EDB Fumigation of Papayas.  Memorandum:   1 Mar  to J. Panetta, with
   attachment.  Environ. Prot. Agency.   Wash.,  D.C.

-------
                                      Ill
Schmitt, R.D.  1977.   Food  Factors.   Memorandum: to O.E. Paynter 4 Sept.
    Environ.' Prot. Agency.

Short, R.D. , et  al»  1978.   Toxicity  Studies of Selected Chemicals, Task V:
    The Reproductive  Tbxicity of Ethylene  Ditromide Inhaled by **ale and Female
    Rats.  Environ* Rrot. Aaency,  Office of Toxic Substances, Contract No:   68-
    01-3242.

Skoronski, B.J.  1979.  Industrial  Hyqiene  Survey for Airborne Ethylene
    Dibromide During  Soil Fumiqation  at the Homestead, Dade Co., Florida. Cow
    Chem. Co., Midland, Mich.
Smith, K.D. , and EL Steinberger,  19^7.  mat is Oligospermia? In "The Testis  in
   Normal and Infertile  *ien," edited  by P.  ^oen and H.R. Manken, Raven
   Press, New York.

Stanford Research  Institute.  19T7.  Criteria fbr a Peccmmended Standard
   Occupational Exposure to Ethyiene  Pibrcmirte. prepared for !!IOSF. Vat1 1
   Tech. Inform. Serv. Publ.  P3-27fi 621, 2^S pp. Stanford Ftes. Instit., ^nlo
   Park, Calif.

Ter Haar. 1978. Rebuttal conment  submitted- in response to tf.S. Fnvironmental
   Protection Agency Positicn Dscunrant 1 on ethylene dibromide. Sperm counts
   Studies on 59 employees enplcved at Ftnyl forporation, Mannolia, Arkansas.
Ter Haar, lQftl. Ributtal ccnnient  submitted in response to U.S^ Fnvironnental
   Protection Money  Ftsiticn  n7curx:nt  2/1 on ethylene diJbrcnicie.  Discussion
   of ''aqnolia  (Ethyl) Spem Count  Stud'/.

Todd, J.W. and  R.R Terry.  1931. Citrus  Summary 1QPO. in:  Florida 2qric.
   Statistics p 44. Fla. Crop  Livest'k.  Ifepccting Serv., Orlando, Fla.

Turner, D. 1976.  Appendix  IT  fsee  abo-je).  Dibromoethane-A '  Survey of ftnlvch
   Records (See Ter Haar, G. 19"7R).

Dndi, S.I. 1973. Background Documents en Use of Ethylene Dibromide.  Memorandum
   with attachments,  Feb.  21.   Forest Service, U.S.  Cept. Agricult.

DSDA  1973.  U.S. Forest Service  Background Document on ECR.

Weidemann, G.J. 1981. Conments on Use Elated Issues in the Great Lakes
   Response to  the EDB PD  2/3.  Memorandum: to J.  Rsinert. 31 Dsc. Environ.
   Prot. Agency. Wash., D.C.

Jfeisburger, E.K. 1Q77.  Carcinogenicity Studies on Halogenated Hydrocarbons.
   Remarks prepared for Conf.  Compar. Metabol. Toxic.  Vinyl-chloride-related
   compounds May 2. 24 pp.

White, G.L. , and J. Lybarger,  1978,  Health Hazard Evaluation He termination
   Report "77-n9-481.  U.S. national Inst. fbr Occupational Safety and Health.

Whorton, D. , et al. 1979.  Testicular Function in FPCP Exposed Pesticide
   Markers.  J. CCCUP. ttsd. 21(3):

-------
                                       112

Whorton, D.M., and T.H. Milby.  1980.  Recovery of Testicular Function Among
   D8CP Workers. J. Occup. Med. 22(3):  177-179.

Wong, O. 1977. A Retrospective Study of Inferred Fertility of Married Men
   Occupationally Exposed to Ethylene Dibrcroide (EDB).  Report submitted by
   Ethyl Corporation, Baton Rouge, Louisiana.

Wong, L.C.K. et aU 1979.  Study of Carcinogenicity and Tbxicity of Inhaled 1,2-
   Dibrcnamethane in Rats treated with Disulfiranu  Midw. Res. Inst. Final
   Rpt. No. 210-76-0131.  100 pp.

Wyrobek, A.J., and W.R. Bruce.  1978.  Chapter 11:  The Induction of Sperm-
   stage Abnormalities in Mice and Humans, in "Chemical Mutagens,* \fol. 5
   Corporation, pp. 257-285.

Zygadlo, L.V. 1981(a). Feasibility Study of Irradiation Facilities.
   Memorandum: 8 Dec. to J. Panetta.  Environ. Prot. Agency.  Wash., D.C.

Zygadlo, L.V. 198Kb). Certified EDB Fumigation Centers in California
   Memorandum: 17 Dec. to J. Reinert. Environ. Prot. Agency.  Wash., D.C.

Zygadlo, L.V. 1982(a). Consumption of EDB Treated Citrus  in California.
   Memorandum to J. Reinert. 8 Jan. Environ. Prot. Agency.  Wash., D.C.

Zygadlo, L.V. 1982(b). Consumption of EDB Treated Miscellaneous Citrus Fruit
   in California.  Memorandum: to J. Reinert. 11 Jan. Environ. Prot. Agency.
   Wash., D.C.

Zygadlo, L.V. 1982(c). Analysis of Benefits Comments Submitted to the Agency  in
   Response to the EDB PD 2/3.  Memorandum: 19 Jan. to J. Panetta. Environ.
   Prot. Agency.  Wash., D.C.

-------
                                113
          Additions to References and Citations

                      Ground Water
Bigler, W. 1983.  Personal communication to s. Cohen.   Epidemiology
Program, Department of Health & Rehabilitative Services,  Jackson-
ville, Fla.

Castro, C.E., and N.O. Belser.  1968.  Biodehalogenation.  Reductive
Dehalogenation of the Biocides Ethylene Dibromide, 1,2Dibromo-3-
chloropropane, and 2,3-Dibromobutane in Soil.  Environ.   Sci.
Technol.,  2_, 779-783.

Chiou, C.T., L.J. Peters, and V.H. Freed. 1979.  A Physical Concept
of Soil-Water Equilibria for Nonionic Organic Compounds.  Science,
206, 831-832.

Cohen, S.Z., S.M. Creeger, R.F. Carsel, and C.G.  Enfield.  1983.
Potential  for Pesticide Contamination of Ground Water  from
Agricultural Uses.  American Chemical Society 186th National
Meeting, Washington, D.C. PEST 89 (full paper in progress).

Eyre, P. August 24, 1983.  Letter to S. Cohen, EPA.  Water
Resouces Div., USGS, Honolulu, Hawaii.

Green, R.  1983.  Personal Communication.  Department of  Agronomy
and Soil Science, University of Hawaii, Manoa, Oahu, Hawaii.
(Also based on some collaborative work with c. Liu, U.  Hi.).

Johns, R.  1976.  Air Pollution Assessment of Ethylene  Dibromide.
The Mitre  Corp., EPA Contract No. 68-02-1495.

Jovanovich,  A.P., and S.Z. Cohen.  In review.  Monitoring Ground
Water in Georgia for Ethylene Dibromide (EDB) A Preliminary
Reconnaissance in Seminole County, Georgia.  Hazard Evaluation
Division,  EPA.

Jungclauss,  G., R. Blair, and S.Z. Cohen.  In progress.
Kinetics and Mechanism of Hydrolysis of 1,2-Dibromoethane.
Midwest Research Institute and EPA.

Mink, J.F. 1981.  DBCP and EDB in Soil and  Water at Kunia, Oahu,
Hawaii.  (In addition to several personal communications to S.
Cohen in 1980-1981.)  For Del Monte Corp.,  48 pages and  appendices.

Wong L., P.  Eyre, J. Mink, A. Hepton, J. Hylin, K. Chenchin,
T. Arizumi,  and others contributing to a project analyzing
soil and groundwater in Oahu for EDB during the summer of
1983.  Department of Health and Agriculture, USGS, and Dole,
Del Monte  and Maui Pineapple Cos.

Zalkin, F.,  R.J. Oshima, R. Sava, and M. Wilkerson.  In progress.
Pesticide  Movement to Groundwater .  (Volumes I and II are in
draft form;  volume III not completed.)   California Department of
Food and Agriculture,  Sacramento, California.

-------
                                  114

            Additions to References and Citations

                            Economics


 Allison, J.R. and DD, Ofiara. 1983.  "Preliminary Benefit
              Analysis of EDB Uses as  a Nematicide on Peanuts"
              Department of Agricultural Economics, University  of
              Georgia.  Prepared under Cooperative Agreement with
              U.S. Environmental Protection Agency   #CR809369.

Daugherty, L.S. and D.L. Kittock. 1983.  "Preliminary Benefit
              Analysis of EDB Use on Cotton".   Pesticide Benefits
              Task Group Department of Agriculture Economics, University
              of Arizona.  Prepared under cooperative agreement with
              U.S. Environmental Protection Agency CR #809298.

 Gait, D. , J. Howard, R. Adamchak, and L. Lipper. 1983.   "The Potential
              Economic Impact of a Cancellation of Ethylene Dibromide
              (EDB) Use as a Soil and  PostHarvest Fumigant on
              Thirty-Two California Crops." Cooperative  Extension
              and Giannini Foundation  of Agricultural Economics
              University of California.  Prepared under  cooperative
              agreement with U.S.  Environmental Protection Agency
              CR #810033.

 Lenahan, J.R. 1983.  "Abbreviated Benefit Analysis of Ethylene
              Dibromide Use on White Potatoes". EAB/BUD/OPP  U.S.
              Environmental Protection Agency.

 Lenahan, R.J. 1983.  "Abbreviated Benefit Analysis of Ethylene
              Dibromide Use on Sweet Potatoes."  EAB/BUD/OPP, U.S.
              Environmental Protection Agency.

 Norton, G.W. and E.G. Rajotte. 1983.   "Economic Analysis of
              Ethylene Dibromide Use on Tobacco,"Virginia Polytechnic
              University and State University.   Prepared under
              cooperative agreement with U.S.  Environmental Protection
              Agency, CR #809256 .

 Vlier, L. 1983.  "Preliminary Benefit Analysis of EDB Use on
              Pineapples."  EAB/BUD/OPP U.S. Environmental
              Protection Agency.

 Zacharias, T.P., A.H. Grube and E.R.  Swanson.  1983.  "Abbreviated
              Benefit Analysis: Ethylene Dibromide (EDB) Use for
              Nematode Control of Soybeans." Department  of Agriculture
              Economics, University of Illinois, Prepared under
              cooperative agreement with U.S.  Environmental Protection
              Agency, CR #809315.

-------
                                  115

              Additions to References and Citations
Burin, G.J
Assessment
with EDB".
, and  Holder, W.J
 of  September 15,
 TB/HED/OPP U.S.
 Exposure

 1983 "Comments on USDA Draft Risk
1983 for Post-Harvest Fumigation of Citrus
Environmental Protection Agency
CDFA. June 2, 1983. Memorandum to K.T. Maddy from J. Lowe, entitled
"Commodity Samples Collected in 1982 for EDB Analysis"., 1 page.

CDFA. 1983. K.T. Maddy, 1982. Complexities of the Decision-Making
Process in Determining Acceptable Hazards of Ethylene Dibromide
(EDB) as a Pesticide in California HS-1038 Cal. Dept. Food
Agric., Sacramento, Calif.
NIOSH. 1976. A Guide to Industrial Respiratory
Publication No. (NIOSH) 76-189.
                                     Protection.  HEW
OSHA. 1981. Draft of Trip Report, entitled, "Fumigation Procedures
Utilizing Ethylene Dibromide on Florida Citrus". 18 pages plus attach-
ment (unpublished).

USDA. 1983. A Preliminary Report of Research on the Effect of Ethylene
Dibromide Citrus Fumigation on Fruit Fly Control and Residue Levels
Conducted by Agricultural Research Service, USDA and Florida Department
of Agriculture and Consumer Services.

USDA. 1983. Draft Document: Quantitative Risk Assessment for
Ethylene Dibromide in it Use as a Post-Harvest Fumigant For Citrus
Fruit. Conducted by Dynamac Corporation.
     418-574 0-83-9

-------
                    APPENDIX I
ETHYLENE DIBROMIDE:   POSITION DOCUMENT 2/3
       OFFICE OF PESTICIDE PROGRAMS
   U.S. ENVIRONMENTAL  PROTECTION AGENCY
              DECEMBER,  1980

                      117

-------
                           EXECUTIVE  SUMMARY

    This Position Document (?D)  2/3  proposes  several  regulatory actions to
reduce the human health  risks  resulting  from  registered pesticidal uses of
ethylene dibromide  (EDB).   These proposed  actions  are based on the Agency's
determination  that  certain uses  of EDB result in unreasonable adverse effects
to man.  The Agency requests registrants and  interested parties to review the
proposed actions and  the evaluations upon  which they  are based, and to provide
comments or any additional relevant  information.   All of the registered uses of
EDB have been  examined in  this document.   These include:   preplant fumigation
of soil (to control nematodes);  fumigation of stored  grains (to control
insects); spot fumigation  of grain milling machinery  (to control insects);  post-
harvest fumigation of citrus,  tropical fruits,  and vegetables under USDA
Animal, Plant  and Health Inspection  Service (APHIS) quarantine (to control
fruit flies);  fumigation of felled logs  (to control bark beetles);  fumigation
of stored beehive supers and honeycombs  (to control the greater wax moth);
vault-fumigation of stored furniture and clothing  (to control various pests);
treatment of plants,  soil  and  grass  sod  in accordance with the OSDA/APHIS
Japanese Beetle Quarantine Program;  and  fumigation beneath concrete slab
foundations of buildings and porches to  control subterranean termites,  and  spot
fumigation of  their galleries  to  control drywood tenriites.

    The first  regulatory action  the  Agency is proposing is to allow
registration to continue for preplant soil fumigation,  the largest  volume use
of EDB,  However, the Agency will permit this use  only.if certain  restrictions
are implemented and additional data  requirements are  fulfilled.  Specifically,
the Agency will require applicators  to wear respirators and protective
clothing; reentry into a treated  field will also be prohibited  for  24 hours
following fumigation.  The  Agency is also  requiring soil  fumigant  registrants
to conduct studies to determine  if this use results in  groundwater
contamination.  The need for these studies is based on  the concern  that SDB  is
structurally similar to dibromochloropropane  (OBCP),  a  chemical  also  used as  a
preplant soil  fumigant, which has been found  in many  groundwater supplies.   In
addition, the Agency will  require registrants  to provide  data  to demonstrate
that residues of EDB in crops grown  in EDB fumigated  soils  at maximum allowed
application rates do not exceed the  limits of  sensitivity of  analytical
methodology (currently 1 ppb).   When results  from  these  investigations  are
available, the Agency will  reexamine the adequacy  of  its  regulatory measures
for this use.

    Second, the Agency is  proposing  to cancel  the  EDB registrations for  two
other major uses:  fumigation of stored grains  and spot  fumigation of grain
milling machinery.  The bases for these proposed cancellations  are  that  the
public- health risks of cancer,  heritable genetic damage, and reproductive
disorders outweigh the economic benefits of these  uses.  The Agency has
concluded that residues of EDB conraminate  the human diet as a result of these
registered uses.   Applicators,  persons near a  treatment  site and workers who
handle treated commodities are  also exposed to significant levels of EDB via
inhalation and dermal contact.

    Third, the Agency is proposing that on July 1,   1983, EDB registrations for
post-harvest  fumigation of citrus, tropical fruits and vegetables will be
cancelled.  The interim period  of allowed  use  will give time for the
development and' implementation  of efficacious  alternatives.  There  are
currently no  effective means of protecting these commodities from fruit fly

                                      119

-------
                                      120
infestations other than BOB.  However, gamma  irradiation  treatment methods are
currently being considered.  These have been  used successfully  in  several
foreign countries.

    Fourth, the Agency is proposing  to cancel  the EDB  registration for
fumigation of felled logs, a relatively low volume use.   The basis for  this
action is the Agency's determination  that the  potential risks to applicators
outweigh the economic benefits of this use.   Based on  the current  application
methods for this use, the Agency has  concluded that applicators may be  at  risk
of incurring unreasonable adverse effects resulting from dermal contact and
inhalation of EDB vapors.  Because the use of  the registered alternative
endosulfan would result in a national reduction in treatment costs, the Agency
has established that there are no benefits for this use of SOB.

    Fifth, the Agency is proposing to defer the decision on the use of  SOB  for
control of termites, but in the interim will  reinforce and expand  current  label
directions by requiring that it be used only by certified commercial
applicators; that applicators must wear protective clothing and 3  respirator;
that premises be vacated by non-applicators during fumigation and  subsequent
aeration; and for use against subterranean termites, that injection holes  be
plugged temporarily dur\Lng, and sealed permanently after, treatment.  The
Agency defers further regulatory action on uses against termites until  SOB  can
be considered in a "cluster analysis" together with all other pesticides used
for control of subterranean and drywood termites (including aldrin, dieldrin,
chlordane, heptachlor, pentachlorophenol, chlorpyrifos, methyl bromide,
sulfuryl fluoride, hydroger :yanide,  and silica aerogel).

    Finally, for the remaining uses:  fumigation of beehive supers, vault
fumigation of stored clothing and furniture, and the OSDA's Japanese Beetle
Quarantine Program, the Agency is proposing to allow these uses to continue
only if certain use restrictions are  implemented and additional data require-
ments are fulfilled.  Specifically, the Agency will require protective clothing
and other low-cost measures designed  to reduce applicator exposures to SOB.  In
addition, the Agency will require registrants of products with each of  these
uses to conduct studies to determine  the EDB air levels to which applicators
are exposed, the number of applicators involved in these uses,  and the  amounts
of EDB applied annually for these uses.

-------
                             Acknowledgements

Special Pesticide Review Division  Team

Richard Johnson, Team Leader
Joseph Panetta, Project Manager
Lynn Brown, Proj ect Manager
Owen Seeder, Proj ect Manager
Cathy Kessler, Student Assistant
Wendy Weinstein, Student Assistant
Gary Otakie, Project Manager
C. Jeffrey Kempter, Section Head

Technical Support Team

James Holder, EFB, HED
Abrahan Mittelman, EF3, HED
Linda Zygadlo, SAB, BFSD
Kenneth Bailey, TB, HED
Ann Barton, HED
George Ludvik, ASI3, 3FSD
Carl Bayer, OGC
John Worthington, EFB, HED
Willard Cummings, PSB, BFSD
Henry Appleton, EFB, HED
Roger Holtorf, EAB, BFSD
Robert Heath, RES, HED
William Burnam, TB, HED
Christine Chaisson, T3, HED

EPA PESTICIDE CHEMICAL REVIEW COMMITTEE   (PCRC)

Marcia Williams, Director, SPRD
Michael Winer, OGC
Allen Jennings, 0PM
John J. Neylan, OE
Donna Xuroda, ORD
Richard N. Hill, OTS
Raymond Smith, OANR
Charles Gregg, OWWM
                                    121

-------
                               TABLE  OF  CONTENTS

Executive Summary

I.  Introduction                                                    1

    A.  General Background and Organization                         I

    3.  Legal Background                                            1

         1.  The Statute                                            1

         2.  The RPAR Process                                       3

    C.  Chemical Background                                         3

         1.  Nomenclature                                           3

         2.  Chemistry                                              3

         3.  Uses and Registration                                  4

    D.  Regulatory History                                          7

         1.  Tolerances                                             7

         2.  Pre-RPAR Actions                                       7

         3.  Post-RPAR Actions                                      7

II. Risk Analysis                                                   g

    A.  Rebuttal Evaluations and Qualitative Assessment of          9
        Risk Data

         1.  Oncogenicity                                           9

         2.  Mutagenicity                                          ^7

         3.  Reproductive Effects                                  24

    B.  Hunan Exposure to BOB from Pesticidal Uses                 3Q

         1.  Dietary Exposures                                     ^1

         2.  Inhalation and Dermal Exposures                       42

                                  123

-------
                                 124
    C.  Quantitative Assessment of Risfc Data                        53
         1.  Oncogenicity                                           53
         2.  Mutagenicity                                           51
         3.  Reproductive Effects                                   e-|_
    4.  Summary and Conclusions                                     g1
    D.  Risks of the Alternative Pesticides to EDB                  ,,

III.  Benefits Summary                                              gg
    A.  Introduction                                                gg
    B.  Fumigation of Stored Grain                                  gg
    C.  Quarantine (APHIS)                                          gg
         1.  Grapefruit                                             7^
         2.  Papaya                                                 71

         3»  Mango                                                  y^

         4.  Other Citrus                                           72
         5.  Miscellaneous Fruits and Vegetables                    72
         6.  Summary                                                72

    D.  Spot Fumigation of Grain Milling Machinery                  72
    E.  Preplant Soil Funigation                                    7-
         1.  Tobacco
         2.  Pineapple
         3.  Citrus

         4.  Peaches

         5.  Peanuts
         6.  Cotton
         7.  Vegetables

-------
                                 125
    F.  Minor Uses                                                  75

         1.  Fumigation of Felled Logs                              75

         2.  Termite Control                                        7g

         3.  Fumigation of Stored Beehive Supers and                73
             Honeycombs

         4.  Fumigation of Stored Clothing and Furniture            79

         5.  Fumigation of Quarantined Plants, Soil and             go
             Grass Sod

IV.  Identification of Regulatory Options                           32

    A.  Introduction                                                32

    B.  Legal Basis for Options                                     32

    C.  Data Gathering for Identification of Options                33

         1.  Additional data                                        33

         2.  Information-Gathering Hearings                         33

    0.  Legal Options Available Under FXFRA                         33

         1.  Classification for Restricted (7se                      g3

         2.  Amend the Terms and Conditions of Registration         33

         3.  Labeling                                               34

         4.  Establish or Re-evaluate Tolerances in                 g4
             Conj unction with FOA

         5.  Cancel After a Specified Period of Time                g4

    E.  Risk Reduction Measures for EDB                             „,-

         1.  Use Restrictions for All Uses of EDB                   gs

         2.  Additional Requirements for Soil                       -_
             Fumigation Uses

-------
                                126
         3.  Additional Requirements for Stored Grain               37
             Fumigation

         4.  Additional Requirements for Spot Treatment of          gg
             Willing and Cereal-Handling Machinery

         5.  Additional Requirements for Post-Harvest               gg
             Fumigation of Citrus and tropical Fruits.

         6.  Additional Requirements for Fumigation of Beehive      g^
             Supers

         7.  Additional Requirements for Vault Fumigation of        g2
             Stored Clothing

         8.  Additional Requirements for Fumigation of Sub-         n?
             terraean and Drywood Termites

         9.  Additional Restrictions for Fumigation of Bark         g-j
             Beetle

        10.  Additional Restrictions for Fumigation of              g^
             Japanese Beetle Under USDA Quarantine Program

V.  Selection of Regulatory Options                                 gg

    A.  Introduction                                                gg

    B.  Fumigation of Stored Grain                                  gg

         1.  Summary of Risks and Benefits                          gg

         2.  Risk/Benefit Analysis                                  97

    C.  Quarantine Fumigation

         1.  Summary of Risks and Benefits

         2.  Risk/Benefit Analysis

    D.  Spot Fumigation of Grain Milling Machinery                 ,,„

         1.  Summary of Risks and Benefits                         ...

         2.  Risk/Benefit Analysis                                 ,,-

    E.  Pre-Plant Soil Fumigation

         1.  Summary of Risks and Benefits

         2.  Risk/Benefit Analysis                                 ..,

-------
                                  127

    P.  Minor Uses
                                                                   116

         1.  Summary of Risks and Benefits
                                                                   116

         2.  Risk/Benefit Analysis
                                                                   1,18

Bibliography


Appendix 1  - Rebuttals Submitted in Response

             to the RPAR Notice on EDB


Appendix 2  - List of Clearances and General
             Survey of Irradiated Pood Products
             Cleared for Human Consumption in

             Different Countries

-------
TABLE NUMBER
Table
Table
Table
Table
Table
Table
Table
Table
Table
Figure
1
2
3
4
5
6
7
3
9
1
Table  10

Table  11


Table  12


Table  13



Table  14


Table  15


Table  16



Table  17
                    LIST OF TABLES

                    TITLE                              PAGE

 Physical  Properties of  SDB                            3

 EDB  Tolerances  and  Exemptions from Tolerances         5

 Emergency  Exemptions  for EDB  Use  on Soybeans         10

 Summary of Mutagenicity Testing of EDB                19

 Sperm Count Distributions by  Cohort and              28
 Occupation

 Sperm Count Distribution for  Males from              29
 Infertile  and Fertile Marriages

 Lesions in Reproductive Organs  of Male  Rats           30
 Exposed to EDB

 Calculated Dietary  Burden (DB)  of EDB Residues  in     33
 Human Diet Resulting  from Soil  Fumigation  Uses

 ZDB  Residues (ppm)  Reported in  Fumigated Grains      34

 Time Line  of Wheat  and  Wheat  By-Product              35
 Distribution with Expected Times  for EDB
 Fumigation

 U.S. Quarantine Requirements  for  EDB Treatment        33

 Measured EDB Residues in  Fruits Obtained from         40
 California Wholesale Produce  Markets

 Estimated  EDB Food  Residues Resulting from            40
 Required Quarantine Fumigation

 Total Dietary Burden  (DB) of  EDB  Residues in          42
 the Human  Diet Resulting  from All  Commodity
 Fumigation  Uses

Measured Air Concentration of SDB during Specific     45
 Soil Fumigation Practies

Estimated Annual Duration of  Exposure to EDB of       47
 Soil Fumigation Workers

Total Estimated Annual Exposure to EDB of             43
 Private Applicators/Transfer Operators during
 Soil Fumigation

 Inhalation  Exposure off Workers to SDB at Citrus       5^
Fumigation  Stations and at a Citrus Warehouse
                                   129

-------
                                   130
Table  13    Inhalation Exposure of Spot Fumigation
             applicators to SDB

Table  19    Inhalation Exposure of Hill Workers

Table  20    Populations at Risk

Table  21    Summary Table of Inhalation Exposure to EDB

Table  22    Risk and Estimated Cancer Cases Per Year due to
             "Probable EDB Dietary Burden"

Table  23    Risk and Expected Cancer Cases Per Year due to
             "Realistic Worst Case EDB Dietary Burden"

Table  24    Upper Bound Risks and Expected Cancer Deaths
             Per Year due to Ingestion of Crops Grown in soil
             Fumigated with EDB

Table  25    Risk due to Occupational Inhalation of EDB

Table  26    Hunan Hazards of Substitute Chemicals

Table  27    Annual Economic impacts of Cancelling
             Ethylene Dibromide

Table  28    Alternative tfematocides Used to Replace EDB
             on Selected Vegetables, Cost of Materials and
             Changes in Mematode Control Cost, by Crop

Table  29    Impacts of Regulatory Options on Stored Grains

Table  30    Impacts of Regulatory Options on Post-Harvest
             Funigation of Citrus, Tropical Fruits, and
             Vegetables

Table  31    Impacts of Regulatory Options on Spot
             Funigation
 54


 56

 57

 58

 62


 63


 64



 65

 66

 70


 77



 99

105



111
Table  32    Impacts of Regulatory Options on Soil
114

-------
                                       131

I.   Introduction

    A.  General Background and Organization

         The Federal Insecticide, Fungicide  and  Rodenticide  Act,  as amended
(FIFRA), requires the Environmental Protection Agency  (SPA)  to  review the  risks
and benefits of the uses of pesticides  suspected of  causing  adverse effects to
hman health or the environment.  On December  14,  1977, EPA  issued  a Notice of
Rebuttable Presumption Against Registration  and  Continued  Registration (RPAR)
of pesticide products containing ethylene dibromide  (SDB), a  fumigant
insecticide and nematocide.  The rebuttable  presumption was  based on validated
studies showing that EDB is oncogenic and mutagenic, and has  adverse
reproductive effects on several species of animals.  Since then EPA has
reviewed rebuttal comments and conducted thorough  risk and benefit  analyses of
the pesticidal uses of EDB.

    This docvanent determines what regulatory actions EPA should take under
FIFRA to change the current use patterns of  pesticides containing EDB.
Included are discussions about the risks and benefits of the  pesticidal  uses of
EDB, and explanations about the rationale for the  Agency's proposed final
decision.  This document is written in  five  chapters.  Chapter  I  is this
introductory section.  Chapter II summarizes and evaluates the comments
received on the RPAR, evaluates the potential health risks of EDB,  and briefly
describes the laboratory evidence of toxicity, available exposure data,
available epidemiology data and the Agency's revised risk  assessment.  Chapter
III summarizes the use—by-use economic benefits  of EDB and the assumptions  and
limits of these estimates.  Chapter IV describes possible  regulatory options
for the reduction of risks.  Chapter V evaluates the risks and benefits  and
adverse impacts of each possible action for  each of the nine  general
categories of the registered uses of EDB, and selects the regulatory action  the
Agency is proposing for each use.

    B.   Legal Background

         1.   The Statute

    The FIFRA (7 U.S.C. Section 126  at sea.) regulates all  pesticide
products.  Under Section 12(a)(1)(A) of FIFRA all  pesticide  products  must be
registered before they may be sold or distributed.  Before a  pesticide
niay be registered, the Administrator must determine that its  use will not
result in "unreasonable adverse effects on the environment,"  which  is defined
by Section 2(bb) of FIFRA to mean "any unreasonable risk to man or  the
environment, taking into account the economic, social, and environmental costs
and benefits of the use off any pesticide."   In sum, any decision  to  rsgistsr or
continue to register a pesticide must weigh  both the risks and the  benefits
from the use of the pesticide.

Section 6(b) of FTFRA 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/her 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."

                                         -L-
      418-574 O - 83 - 10

-------
                                 132

           Table 1. Physical Properties of EDB j_/
Boiling Point
Melting Point
Vapor Pressure
Specific Gravity

Solubility
Stability
Flammability
Appearance
Heat of Vaporization
Corrosiveness
Flash Point
Fire Point
Weight 4 Volume
Conversions
-  -131.5JC
      9.3°C
—  11.0 mm Hg § 25 C
—  liquid (water @ 4°C =1)
    2.172 § 20°C gas (air =1) 6.487
—  0.43 g/100 g water § 30°C, soluble
    in ethanol, ether and most organic
    solvents.
—  Stable
—  Nonflammable
—  Colorless liquid
—  46.2 cal/g
    vapor or liquid attacks, many
    paints and some metals
    (especially aluminum)
    None
    None
    1 U.S.
    1 U.S.
gallon
gallon
                              —  1 Ib. liquid
                              —  1 oz/1000 ft
                3  :
                                  1 ppm s  7.69 mg/nr
                                  1 mg/nr = 130 ppb
18.11 Ibs.
8.215 kg.
208 ml §
1  og/liter
  1 gm/ or
                               25°C
Density of EDB in
saturated air
Cone, of EDB in air
at saturation
--  1.08 (air s 1)

— 1.3? by vol. e  20°C
I/ Data taken from several sources including Monro (1969),
Thomson (1977), Martin (1971), Spencer (1973), Merck (1976).
                                 -2-

-------
                                      133

    The Administrator may also cancel the  registration  of a  pesticide if its
labeling does not comply with the misbranding provisions  of  FIFRA  which  require
the labeling to contain language "adequate to protect health and the
environment" [FIFRA Section 2(q)].

         2.   The RPAR Process

    The purpose of Pebuttable Presumption Against Registration  (RFAR)  is to
gather risk and benefit information about pesticides which appear  to  pose risks
to health or to the environment.  As described  in 40 CFR  162.11, this  process
is designed to allow an open/ balanced decision with participation.by  all
interested groups.  The RPAR regulations describe various risk  criteria,  and
provide that an RPAR shall arise if the Agency determines that  any of  the risk
criteria have been met.  Cnce a notice of rebuttable presumption is published,
registrants, applicants and interested persons may submit evidence in  rebuttal
or in support of the presumption.  Ml parties may also submit  evidence  on the
economic, social and environmental benefits evidence of any  use of the
pesticide.  If the presumptions of risk are not rebutted, the evidence of the
benefits from use of the pesticide submitted to or gathered  by  the Agency must'
be evaluated and considered with the risk information.  The  Agency analyzes
various risk reduction methods and their costs.  The Agency  then determines
whether or not the risks of pesticide use may be outweighed  by  its benefits.
If a balance between risks and benefits cannot be reached for a specific use,
the registration for that use must be canceled or otherwise  restricted, or,  in
the case of a new application, denied.

         C.   Chenical Background

              1.   Nomenclature

    Sthylene dibromide (SDB) is the common name for 1,2-dibromoeth,ine, CAS
number 0001060934.  It is a soil and commodity ffumigant having both nematocidal
and insecticidal use.  The term HDD as used in this document refers
specifically to the organic molacule, ethylene dibromide, and does  not include
its breakdown products, such as inorganic bromide(s)  or total bromide*s).
These latter two terns are used in the food additive tolerances (21 CFK 123)
and raw agricultural commodities tolerances (40 CFR 180).

              2.   Chemistry

    At room temperature, pure EDB is a colorless, heavy, nonflammable  liquid.
It is prepared commercially by reacting bromine with ethylene gas and  has a
characteristic mildly sweet odor detectable in air at levels ranging  from 10  to
25 pom (77 mg/mJto 192.5 mg/m"1).  Its chemical formula is CH_3rCH.,3r,
its molecular weight is 187.38,  and its  structural formula is:

                                  Br  Br
                                   I   I
                              H - C - C  - H
                                   I   I
                                  I!   H

    Table T lists other chemical and physical properties of SDE.
                                         -3-

-------
                                      134

              3.   Uses and Registration

    SDB has been produced in the United States since the  1920's  and  used
primarily as a lead scavenger in leaded antiknock gasoline  additives.   Because
of Che Agency's regulations requiring a phase-out of leaded gasoline,  this  use
of EDS has steadily dropped over the last few years and is  expected  to
eventually become negligible.

    During the 1950's and 1960's, federal pesticide registrations  for  uses  of
EDB as a fumigant were granted under FIFRA.  Currently, the major  registered
pesticidal uses of EDB are:

    o Pre-plant soil funigation by injection for a wide variety
      of food and non-food crops including vegetables, fruits,
      grains, peanuts, cotton, and tobacco;

    o Post-harvest commodity ftmigation for grains, fruits
      and vegetables, including various State, Federal or
      international quarantine programs on citrus, fruits,
      nuts, and vegetables;

    o Fumigation of grain milling machinery and flour mills
      to control insect infestations in milling remnants and
      processed milled products.

    Other uses of EDB include:

      o Control of wax moths in beehive supers;

      o Control of mountain pine bark beetles in the western
        states by Federal and State forestry agencies;

      o Control of drywood and subterranean termites in
        structural pest control operations;

      o Control of clothes moths, dermestid beetles and similar
        pests of furs, rugs, and other items in fumigation
        vaults; and

      o Control of Japanese beetles in the soil of balled ornamental
        trees and shrubs under the APHIS Quarantine Program.

    EDB is usually formulated as a liquid concentrate or as a gel.  There are
122 federal pesticide registrations, held by 53 registrants, of products
containing EDB as an active ingredient, and 24 state registrations, held by 12
registrants.  In addition, emergency exemptions for SDB use on soybeans
(discussed in detail in the next section) were granted to six states early this
year for use through the month of July, 1980.  There are no Federally
registered products containing EDB as an inert ingredient.  Most of the
registered products are mixtures of EDB and other active ingredients, such as
carbon tetrachloride/ ethylene dichloride, methyl bromide, chloroform,  carbon
disulfide, sulfur dioxide, chloropicrin, and benzene.

    The Agency (Holtorf, 1990) estimates that about 340-360 million pounds of
active ingredient EDB are produced annually in the U.S. by four basic
producers:  Dow Chemical, Ethyl Corp., Great Lakes Chemical and PPG


                                         -4-

-------
                              135
   TABLE 2.  EDB  (Inorganic Bromide)  Tolerances  and  Exemptions
             from Tolerance.
Commodity

Soil Fumigation Uses

  Broccoli
  Carrots
  Melons
  Parsnips
  Potatoes
  Eggplants
  Okra
  Summer Squash
  Sweet Corn
  Sweet Corn Forage
  Sweet Potatoes
  Tomatoes
  Pineapples
  Cucumbers
  Lettuce
  Peppers
  Cottonseed
  Peanuts -7
  Asparagus
  Cauliflower
  Lima Beans
  Strawberries
Tolerance
75 ppm
50
            Citation and Form
            40 CFR 180.126
            Inorganic bromides
            (calculated as Br)
ppm
40 ppm
30 ppm
25 ppm
  •
10 ppm
  «
 5 ppm
Post-Harvest  (Commodity) Fumigation Uses
  Barley -
  Corn
  Oats
  Popcorn
  Rice
  Rye
  Sorghum (milo) 2/
  Wheat
50 ppm
            40 CFR 180.146
            Inorganic bromides
            (calculated as Br)
I/  40 CFR 180.126(a) prohibits use of treated peanut hay
~~   and hulls as feed for meat and dairy animals.

2/  40 CFR 180.1006  exempts these grains from the require-
~~   ment of a tolerance  for organic bromide  residues of
    EDB.

-------
                                136
Table 2 - continued

Commodity                    Tolerance

Quarantine Fumigation Uses

  Beans (String}             10 ppm
  Bitter Melons
  Cantaloupes
  Cavendish Bananas
  Citrus Fruits
  Cucumbers
  Guavas
  Litchi Fruit
  Litchi Nuts
  Longan Fruit
  Mangoes
  Papayas
  Peppers (Bell)
  Pineapples
  Zucchini Squash
  Cherries                   25 ppm
  Plums (Fresh Prunes)         •
               Citation and Form
               40 CFR 180.146
               Inorganic bromides
               (calculated as Br)
               40 CFR 180.146
               Total inorganic and
               organic bromides
Food Additive Tolerances

  Milled Grain Fractions
125 ppm
21 CFR 193.225
1977 Inorganic
bromides (Calculated
as Br)  from all
fumigation sources
including grain-mill
machinery.
  Fermented Malt
    Beverages  from
    corn grits and
    craciced rice
25 ppm
21 CFR 193.230 (d)
(1977) Total in-
organic bromides
(calculated as Br)
from all fumigations
with bromine compounds
                            -6-

-------
Table 2 continued

Commodity

Dried Eggs
                                137
Tolerance
  400 ppm
Parmesan and
 Roquefort Cheese
  325 ppm
Citation and Form

21 CFR 193.250 Total
inorganic bromides
from all organic
bromides used as a
soil, commodity, or
processed food fumi-
gant (calculated
as Br) .
Concentrated Tomato           250  ppra
  Products

Dried Pigs                       "

Processed Foods               125  ppm
  (including processed
 grain and cereal products)
Milled Fractions  in
 Animal  Feeds  from
 barley, corn, grain
 sorghum (milo) ,  oats
 rice, rye, and wheat
  125 ppm
21 CFR 561.260(b)
Inorganic bromides
(calculated as Br)
resulting from
carry over or
concentration of
residues from
fumigation of grains
with MBr or SDB.
                           -7-

-------
                                     138

Industries.  The bulk of this production, about 230 million pounds,  is  used  as
a gasoline additive.  Pesticide usage of EDB  in comparison  to  total  production
is quite snail, comprising about four percent or  13 to  15 million  pounds  of
annual U.S. production.  Estimates of quantities  imported and  exported  for
pesticide usage are considered confidential and are not  publicly available.
The major pesticide uses of EDB are dominated by  agricultural  soil fvmigant
use, comprising about 90% of U.S. consumption.  The remaining  10%  is  used for
commodity and space ftmigation.

    0.   Recrulatory History

    1.   Tolerances

    There are no tolerances for EDB per se in or  on raw  agricultural
commodities because it had been asguned, on the basis of data  originally
submitted by petitioners, that there would be no  EDB residues.  This  conclusion
was based on the hypothesis that the parent EDS compound had either completely
evaporated or been converted to bromide ions which were  subsequently  taken up
by plants as inorganic bromides.  Residue analyses then available  for EDB in
crops grown in treated soil were negative for the organic molecule.   The
analytical method for organic bromide, employed from about 1950 through 1965,
had a sensitivity of only 0.2 to 1.0 ppm and was  not specific  for  EDB, but
rather measured only the organic bromide ion.  Consequently, tolerances were
established in 40 CFR 180 for inorganic bromides  resulting from the use of EDB
in or on raw agricultural commodities, and as a food additive  in processed
foods by 21 CFR 193 and 561.  These tolerances are presented in Table 2.

    2.  Pre~RPAR Actions

    On October 16, 1974, the National Cancer Institute (NCI) issued a
"Memorandum of Alert" describing a preliminary finding of EDB's carcinogenic
activity in rats and mice.  On the basis of this  information,the Agency began a
general review of the production and use status of EDB.

    On July 14, 1975, the Environmental Defense Fund (EOF)  petitioned the
Agency to investigate EDB on a priority basis and to take suspension or
cancellation action against the compound, as justified (Butler, 1975).  In
November, 1975, EDB was accepted by the Off-ice of Special Pesticide Review as
an RPAR candidate, based on the final NCI study which showed SOB to be
carcinogenic in both rats and mice.  After this initial petition,  EOF amended
and renewed its request for action against EDB under FTFRA on January 21,  1976,
September 30,  1976, and August 26, 1977 (Hinkle,  1976,  1977).

    Additional evidence for this referral included studies showing EDB to be
mutagenic and capable of producing adverse reproductive effects in several
species of animals.  On September 7, 1977,  the Agency's Carcinogen Assessment
Group issued a Preliminary Risk Assessment for EDB which stated, "There is
strong evidence that EDB is likely to be carcinogenic to man"   (CAG, 1977).

    Following intensive review and validation of the literature, the Agency
published a Notice of RPAR and the RPAR Position Document 1  in the Dec. 14,
1977 Federal Register.  The basis for the RPAR is the risks  to man of
oncogenicity,  mutagenicity, and adverse reproductive effects as a  result of
pesticidal exposures to EDB.
                                          -8-

-------
                                       139

         3.  Ppst-RPftR Actions

    During the initial rebuttal period,  December  14,  1977,  to January 30,  1978,
the Agency received twelve requests  for  extension of  the  rebuttal  period.   The
Agency granted an extension to April 3,  1978.   During and after  the  extended
rebuttal period, 91 rebuttal submissions were received.   After the extension
deadline, several rebuttals were amended, resulting in  a  total of  105 docunents
in the rebuttal file (Appendix 1).  Thirteen of these submissions  contained
rebuttal information alone or in combination with benefits  information.  Thirty-
four of the submissions provided some  type of benefits  information only, while
an additional thirty-five contained  "testimonial" statements  of  need  which  were
not supported by any data.  There were several other  submissions,  including
support for an immediate "ban" or other  action  against  EDB.

    Following the 1979 suspension of DBCP as a  preplant pesticide  for soybean
nematode control, several states claimed that there are no  other registered
pesticides that are economical or effective against nematodes.   Based on test
results showing EDB to be as efficacious as DBCP, six states  applied  for
temporary permits (e.g., exemptions from FIFRA) to allow  this  use  under section
1B(c) of FTTSA.  These exemptions were granted  for the  period  beginning April
24, 1980 and ending July 31, 1980, with nunerous restrictions  to reduce health
risks to applicators.  Table 3 describes these  exemptions in  detail.

II.  Risk Analysis

    The Agency issued the Notice of RPAR and Position Document 1 (PD-1) based
on three criteria:  oncogenicity, mutagenicity, and reproductive effects.
After issuing the RPAR, the Agency began a detailed risk/benefit analysis
according to applicable guidelines and policies.  This chapter is divided into
three sections:  Section A evaluates the rebuttal comments the Agency received
following the issuance of the PD-1, and discusses the scope and quality of the
available toxicological data, including new data obtained since  the publication
of ?D-1; Section B estimates hunan exposures; and Section C quantifies the
potential health risks of EDB to humans.

    A.   Rebuttal Evaluations and Qualitative Assessment off Risk Data

         1.   Oncoorenicity

    In issuing the RPAR, the Agency relied on evidence of oncogenicity reported
in an NCI-sponsored study (Weisburger,  1977).  This study showed a  statisti-
cally significant increase in the incidence of squamous cell carcinoma of the
forestomach in both rats and mice as compared to the controls.  Based on its
preliminary review of the data from NCI, the Agency stated:

         .. EDB causes a significant increase in the incidence of gastric
    carcinomas in both sexes of rats and mice.  Metastases of these traors
    are reported.  The tuner rates appear to be high and the differences
    are highly significant. (Anderson,  1977).

    The Agency's position that an oncogenic rebuttal should arise is  based on
its evaluation of this study and 40 CFR Section 162. 11(a)(3)(ii)(A).   This
statute states that rebuttable presumption of oncogenicity shall  arise if a
pesticide "... induces oncogenic effects in experimental mammalian  species or
in man as a result of oral, inhalation or dermal exposure..."  As a further
clarification of the provision,  the preamble to the Agency's Interim  Procedures

                                          -9-

-------
                                             Table 3.  Emergency Exceptions Cor ECU Use en Soybeans (4/2V80-7/31/80)
o
                     location of tJbe/Appl leant

                     Alabawi/Depart«Bnt of
                       Agriculture and Industries
Pest

Soybean Cyst Heretoda
Boot-Knot Neaatode
                                                                    Acreage (Haxlnnt)

                                                                         160,000
                 olSoybean Cyst
Agriculture and Consumer Services   Southern Root-Knot Nematnde
                                    Peanut Root-Knot MemaKVte
                                    Renifon Nenatoda
                                    String tteroatoda
                                                                                                130,000
                     GeorgiaTbepartaenfc of
                       Agriculture
                     Tennessee /Department of
                       Agriculture
Soybean Cyst Mematnde
teanut Root-Knot Neroatode
Lance Neaatode
                                                                          98,000
Soybean Cyst Nematoda
                                                                       2,118,818
 Product

Soilbroa 90GC
                                                               "Sollbroai 90BC
Soilbroa 90BC
 Dow Fume W-85
 Ethylene Dibronide
Hissisaippi/Departoent of
Agriculture and Conuerce
South Carolina/Oejnuon
University
Spiral
i Cyst Nenatode
Coluobia lance Nenatoda
5,000
1.000,000
Sollbrou 90EE
Sbilbrcn 90BC
Dow fiws W-85
Ethylene Dlbronide
                                                                                                                                                             *>-
                                                                                                                                                             o
SoTlbran 90EC
                     Restrictions on use for all above oxeai>tioi)a include>
                     1.  Application rate »ay not exceed 32 Ibs/acre/year of EEB active ingredient
                     2.  Ajplication is to be carried out in accordance with the suppleoantal labeling subnltted on April 1, 1980, except
                         uae must be United to pests authorized by this exeuption.
                     3.  Applicators and others handling EDO oust wear protective clothing and a respirator, unless a "closed eysUaa"
                         is used, which prevents applicator contact with EfiB.
                     4.  All applicable precautions and restrictions on tlte registered product label are to be observed.
                     5.  Ma wore titan tire acreages designated above are to be treated.
                     6.  Soybean luty frua treated fields way not be used for feed.
                     7.  Residues of inorganic bromide must not exceed 125 ppo. in or on soybeans or 150 ppu in soybean Heal.  Soybeans
                         with lean than 125 ppa and uoybean uual with leaa than 150 ppn inorganic bcomicle i-eslduea nay be moved in
                         interstate com*roe.  Ilie food and Drug Administration, DICW, has been notified.

-------
                                      141

and Guidelines for Health Risk and Economic  Impact  Assessment of  Suspected
Carcinogens (May 25,  1976; 41 FR 21402) states  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."

    Cn the basis of the evidence summarized  in  Position Cocunent  1, the  Agency
concluded that the cancer risk criteria had  been exceeded by all  registrations
and applications for registration of pesticide  products containing SDB,  and
that a rebuttable presumption against new or continued registration of such
products should arise.

    Six respondents (Rebuttal nos. 3C, 47, 48,  64,  78  and 79) addressed
rebuttal comments to the oncogenic criterion.   The  Agency has divided these
rebuttals into two broad groupings, those which address uncertainties in
extrapolating to man from the laboratory studies conducted  on animals, and
those studies which investigate the health of workers  occupationally exposed to
EDB.
          a.  Response to Rebuttals of Animal. Oncogenicity

         (1)  Direct Response to Rebuttals

    The comments addressing the uncertainty of  extrapolating from animals  to
man can be grouped into broad categories:

    o The route of administration, intubation,  was
      inappropriate because human exposure would be
      primarily by inhalation;

    o Massive doses both obscured detection of  a
      no observable-effect-level (NOEL) and
      overwhelmed the animals' metabolic and
      detoxification processes;

    o The forestomach, where tumors occurred in the
      test animals, is without a counterpart in humans;

    o The oncogenic response at the site of intubation
      was due to an extreme
      irritant effect.

    The Agency has .concluded that the first rebuttal listed above is invalid.
Human exposures to EDB also result from its presence in the human diet.   At
least three uses of EDB result in its presence  in the human diet:  fumigation of
stored grains; "spot" fumigation of grain milling machinery; and postharvest
fumigation of citrus, tropical fruits and vegetables.

    In addition, the latter part of this section describes recent  evidence
which demonstrates that EDB is a potent carcinogen in animals following
inhalation exposure.  Midwest Research Institute (MR!), under a  contract  with
The National Institute of Occupational Safety and Health  (NIOSH),  found a.
significant increase in the incidence of tumors of the  spleen, mammary gland,
adrenals and subcutaneous tissues in a group of rats exposed to  20 ppra of 1,2-
dibromoethane alone (Wong, 1979).  Most recently,  Hazelton Laboratories,  under
a contract with NCI (1980), also found inhalation of cDB  to be causal  to  a
statistically significant increased incidence of cancers  in both rats  and mice.
                                           -11-

-------
                                      142

    The second rebuttal comment, that the massive doses overwhelmed  the
animals' metabolic systems, obscuring a no-observable-effeet-level  (NOEL),
is one of the most common criticisms of NCI cancer bioassay studies  (Mishra,
1973a).  It has been argued by the Agency (Mishra, 1978a) and SCI that the use
of maximum tolerated dose (mtd) and 1/2 maximum tolerated dose  in cancer
bioassay (even in the absence of a NOEL) is now accepted by the Agency, and its
use has been included in the proposed EPA/OPP oncogenicity testing guidelines.
Results from the above-mentioned NCI study (1977) show that mice receiving only
1/2 the mtd experienced extremely high and statistically significant cancer
incidence.  In addition, results from the NIOSH and NCI-sponsored studies
discussed above demonstrate that EDB is carcinogenic in animals at low doses
(20 ppm).  The Agency's Interim Cancer Guidelines (41 FR 21402) state that in
the absence of definitive information to the contrary, any exposure, however
small, will incur some risk of cancer on the exposed population.  In essence,
the Agency position is that there is an assumed cancer risk front chemical
carcinogens for any level above zero.  The Agency concludes that this rebuttal
is invalid.

    The Agency also concludes that the third rebuttal is not valid.  It is true
that humans do not have a counterpart to the forestomach of rats.  "However,
the site of tumorigenic effect, and the tumor cell type produced by a chemical
is known to change depending upon the animal species, dose schedule and routes
of administration used.  Thus, the absence of forestomach in huaans may not
make humans immune to carcinogenic effects of EDB."  {Mishra, 1978a).  This
rebuttal is further invalidated by results of the NIOSH study in which inhaled
EDB was found to induce tumors of the spleen, mammary gland, adrenals, and
subcutaneous tissues.

    The fourth rebuttal, that the oncogenic response was due to an extreme
irritant effect, is based on the chronic "irritation" theory of carcino-
genesis.  This general theory that there is high correlation between chemicals
which are irritants and are also carcinogens has long been discarded and the
evidence against it is sunmarized in the book Carcinogenesis as a Biological
Problem  (Berenbluo, 1974).  Most irritants in mouse dermal tests were
noncarcinogenic and the same was true for irritants introduced  in subcutaneous
and other types of exposure tests.  Among the carcinogens which could be
described as irritants, there was no correlation between carcinogenic potency
of the  agent and the degree of irritant inj ury produced.
                   (2)  Pertinent New Studies

    Since the issuance of PD-1, three new studies were completed which
investigated the carcinogenic potential of EDB in animals.  These are the Dow
Chemical Company Rat Inhalation Study, the Midwest Research Institute Rat
Inhalation Study, and the NCI Rat and Mouse Inhalation Study.

    Dow Chemical Co. Rat Inhalation Study;  A final report entitled  "13
Week Repeated Inhalation Study on Ethylene Dibromide  (EDB) in Male and Female
Rats"  (Nitschke, 1980) was reviewed by the Agency (CAG,  1980b).  The study
con :isted of exposing male and female CDF (F-344) rats to 0,  3,  10 or 2076
mg/m   ethylene dibromide, 6 hours/day, 5 days/week for 13 weeks  for  a total
of 67-68 exposures in 95-96 days.  Scheduled sacrifices  occurred after 1, 6,
and 13 weeks of exposures.
                                           -12-

-------
                                      143

Findings from the study included:

    o rats exposed to 23 mg/ra  EDB showed no consistent  effect  in  any
      parameter;

    o at 77 mg/m  EDB caused slight epithelial hyperplasia of the
      nasal turbinates in animals necropsied after  1, 5  or 13 weeks of
      exposure; however, 88 days after the last exposure to EDB,
      no morphological difference from control animals was observed;

    o rats exposed to 3076 mg/m  EDB showed a definite adverse
      response characterized by a decrease in body weight gain, an.
      increase in liver and kidney weights, and pathologic effects in
      the nasal epithelium.  The nasal turbinates of rats progressed
      from very slight hyperplasia of the epithelium to  hyperplasia and
      nonkeratinising squamous metaplasia•

    The Dow report concludes:

    While this study has shown that repeated subchronic  exposures  of rats to
77 mg/m  or 3076 mg/m  EDB induce pathologic changes in  respiratory
epithelioa of the nasal turbinates,  a subsequent post-exposure phase revealed a
lack of progress of the lesions with almost complete reversions toward normal
histologic appearance of the nasal turbinates.  In view of these findings and
the lack of any lesions subsequent to repeated exposure  to 3 ppm EDB, short-
term repeated exposure to these concentrations of SDB would not be expected to
result in any long term irreversible effects upon the nasal turbinates or other
tissues of the body.

    The Agency (CAG, 4/30/80) has reviewed this study, noting the  short
duration of exposures, and has concluded that this study  involved  exposures to
animals which are clearly too short in exposure duration  to be of  value in
determining chronic risks present from long term exposures to EDB.  (CAG,
1930b).

     Midwest Research Institute Rat Inhalation Study (Wong, 1979):  This
study involved significantly longer  exposures and lower concentrations of EDB.
This 18 month study with Sprague-Dawley rats was designed to assess the
carcinogenicity of inhaled ethylene  diJbromide (EDB)  and the effect of dietary
disulfiram on the carcinogenic response to ethylene dibromide.   For this
purpose, the tumor response of four  groups of males and females, each with a
size ranging from 120-134 animals was evaluated as follows:  (1)  controls;  (2)
154 og/m  of inhaled EDB administered for 7 hrs/day, 5 days/week for 18
months? (3) 0.05% disulfiram in the  diet for 13 months,  and (4)  both  SDB and
disulfiram.

    The preliminary draft report was reviewed by the Agency (CAG,   1979).  The
project officer, Dr. H. B. Plotnick,  stated that although the report  is in
draft form, the pathological data have been thoroughly reviewed and will not be
altered in subsequent versions.  The CAG analysis states  that,  this report
permit us to make a direct estimate  of the carcinogenic effect  of inhaled ED3
and frees us from having to rely on  the unsatisfactory upper  bound estimates
previously derived from intubation studies.
                                           -13-

-------
                                      144

    The authors of the MRI report found that EDB alone (group 2) and EDB with
disulfiram (group 4) produced a statistically significant increase in tumors at
several sites, compared to the group of controls.  The sites are listed below.

                                     SOB	          EPS S Disulfiram
                              Males       resales         Males    Females
    Spleen heoangiosarcoma      *            *               *         *
    Adrenal tumors              *            *               *         *
    Liver hemangiosarcoma       XX               X         *
    Liver hepatocellular        XX               *         *
      carcinoma
    Mesentery heoangiosarcoma   XX               *         *
    Kidney renal tumor          XX               *         *
    Subcutaneous tumors         XX               XX
    Thyroid tumors              XX               *         *
    Mammary tumors                           *               -         *
    Lung tumors                 XX               *         X
    * denotes statistically significant increase, p<0.05
    X denotes not statistically significant
    - not applicable

    Thus, this study demonstrates that BOB is also carcinogenic in rats
following prolonged inhalation exposures.  "A high mortality rate and a
significant increase in the incidence of tumors of the spleen, mammary gland,
and adrenals was noted in the group  (of rats] exposed to 154 rag/m 1,2-
dibromoethane alone."  (Plotnick, 1980).

    National Cancer Institute Rat and Mouse Inhalation Study   (NCI, 1980):
The possible carcinogenicity of 1,2-dibromoethane in rats and mice was tasted
by Hazelton Laboratories of America, Inc., and published recently in a cancer
bioassay under a contract sponsored by the National Cancer Institute (NCI).

    In the rat study, 50 F344 rats of each sex were exposed to 77 or 307S
ag/m  of 1,2-dibromoetthane by inhalation 6 hours/day, 5 days/week for a
maximum of 103 weeks.  Untreated control groups comprised of 50 rats of each
sex were exposed in chambers of ambient air.

    The body weights of the rats were monitored throughout the study.  The mean
body weights of the high exposure group of either sex were lower than those of
the corresponding untreated controls.  Survival periods in the high-dose rats
of both  sexes were shorter than in the controls.

    The histopathologic examination of all the major tissues and organs
revealed a statistically significant increase in the incidences of carcinomas,
adenocarcinomas, adenomatous polyps of the nasal cavity, and hemangiosarqcmas
of the circulatory system in male and female rats exposed to a 3076 mg/m"
dose level as compared to the controls.

    In addition, a statistically significant increase in tumor incidences of
the following organs of rats was observed:  adenomas of the pituitary in low-
dose rats of either sex when compared with the corresponding controls; and
alveolar/bronchiolar adenomas or carcinomas and fibroadenomas of the mammary
gland  in high-dose female rats when  compared with ths corresponding controls.

                                         -14-

-------
                                      145

    In summary,  1,2-dibromoethane was carcinogenic  for F344  rats.   These
results confirm  the carcinogenic effect  observed  in the JIIOSH  study in Sprague-
Dawley rats tested by Midwest Research Institute.

    In the mouse  study, fifty B6C2F1 mice of  each sex were exposed  to  77  or
3075 mg/m  of 1,2-dibromoethane by inhalation 6 hours/day, 5 days/week, for  a
minimum of 78 weeks.  Untreated control  groups consisting of 50 nice of each
sex were exposed  in a chamber to ambient air.  Throughout the  study, mean body
weights of high-dose mice of either sex  were  lower  than the controls.   Survival
of the low and high-dose female mice was significantly shorter than survival
for the corresponding controls.

    The histo pathologic examination of all major organs revealed statistically
significant incidences of alveolar/ bronchiolar adenomas or carincinomas  in
high-dose male mice as compared to the corresponding controls.  In  high-dose
female mice, the  incidences of hemangiosarcomas of  the circulatory  system,
fibrosarcomas of  the subcutaneous system, carcinomas or adenomas of the nasal
cavity, and adenocarcinomas of the mammary gland were significantly higher than
those in the corresponding controls.

     In summary,  1, 2-dibromoethane induced a carcinogenic effect in male and
female B6C3F1 mice.  The Agency (GAG) will extensively review this  srudy  and
evaluate data for passible use in the quantitative  risk assessment  for human
population exposure to 1,2-dibromoethane via  inhalation.

                    (3)  Conclusion - Animal  Oncogenicity of EDB

     On the basis of the discussion presented above, the Agency concludes that
the presumption of oncogenic risk as stated in PD-1  has not been rebutted,
because no respondent to that notice has sustained  the affirmative  burden of
proof set forth  in 40 CFR 162.11(a}(4).  In providing an assessment of the
"weight" or meaning of the available studies on oncogenicity of EDB to animals,
the Agency concludes that SDB is a potent animal oncogen.

                   b.  Response to Rebuttals _gf_Human Oncogenicity
                       Of EPS:

     In response  to the human oncogenic  potential of EDB, one respondent
(Rebuttal No. 3c) submitted three studies which investigated the health of
workers exposed  to EDB.  Each of these studies is summarized below,  together
with the Agency evaluation.

                    (1) The Mortality Experience of  161
                        Employees Exposed to Ethylene Dibromide
                        in Two Production Units (Ott,1977)

    This study was undertaken to investigate the mortality experience of
persons occupationally exposed to EDB in two manufacturing units.   The emphasis
of the study was on cancer mortality and cancer due  to  respiratory  disease.
The total cohort population consisted of 161 employees  from both units.

    The estimated EDB exposure and duration of exposure  ranged  from 1-96 ppm
and 1-16 years respectively.   In both Units I and II, employees were also
exposed to several other potentially toxic  chemicals, such as vinyl  chloride,
carbon tetrachloride, benzene, nickel acetate and silica.   The  study did not

                                         -IS-

-------
                                      146

mention the exposure levels or duration of  exposure of  employees  to  these and
nunerous other chemicals.  In addition, a number of employees  in  Unit  II  were
exposed to arsenic, a known carcinogen.

    The investigators reported that no increased malignancy  was noted  in  Unit
I.  The observed cancer incidence was 1 case against  3.6 expected.   The deaths
from all causes were not elevated over expected deaths.  In  Unit  II, after
excluding the workers exposed to arsenic, a slight increase  in both  cancer
deaths and deaths due to respiratory diseases were observed.  A total  of  5
cases of malignant tumors were observed in contrast to  the expected  incidence
of 2.2 (p—0.072).  Two of these 5 cases were stomach  cancers against an
expected incidence of 0.7 (p » 0.150).  The remaining 3 cancer cases were also
higher relative to 0.9 cases expected (p «  0.063).  The investigators  concluded
that, because of the limited size of the study population and the variety of
toxic agents to which individuals in Unit II may have been exposed,  it would  be
difficult to interpret these findings as being demonstrative of effects of EDB
specifically, or any other specific toxic agent.

    In submitting this study, the rebutters noted that  it was of doubtful
significance (Great Lakes Chemical Co., 1977).  The Agency has concluded the
following:

     "This study shows slight increases in  stomach cancer incidence and deaths
due to all causes; however, this may not be solely attributed to EDB exposure
alone, since workers were also exposed to other chemicals, and that two cancer
deaths occured in the same family (father and son)."  (Mishra, 1978a)

                   (2)  The Associated Ocrtel Company Ltd: A Mortality
                        Survey on Employees at an Sthylene Dibromide Plant
                        at Havle, England (Turner, 1977)

    The study investigated the death rates of 117 persons occupationally
exposed to EDB where employment was known to exceed 4 years.  Rates per 1000
man years were compared between EDB workers and the population of southwest
England.  The number of man years in the four groups in the age range of 25 to
44, 45 to 64, 65 to 74, and 75 and over,  were 1797, 1440, 177,  67 respectively.
The author claims that when these groups were compared with the population in
southeast England, there was no observed increased incidence of cancer deaths
in the workers exposed to EDB.

    No monitoring or exposure levels of EDB in the plant are discussed.
Without knowing the exposure Levels of EDB, the Agency is unable to use this
study to assess the human oncogenic potential of EDB.    (Mishra, I978a).

                   (3)  Dibromoethane;  A Survey of Employee Records at
                        Amlwch Factory (Turner, 1976).

    This study was a retrospective investigation off the mortality experience of
workers occupationally exposed to EDB at a manufacturing plant in southwest
England.  The health records of the personnel employed at the factory during
the period from 1952 to 1975 were examined.  These employees were an unusual
group, in that they were preselected for employment on the basis of above
average health.  Of the 297 employees known to have worked at the site, a
cohort of 273 persons was accounted for and used in the study.
                                        -16-

-------
                                       147

    Ambient air  levels of  EDB  in  the  plant were not given.   Although the
duration of exposure of  the  entire cohort  was not determined, the investigators
reported that the  31 men,  who  died by the  end of 1975,  had  worked on the site
an average of 13 years.

    The average  age of the group  was  40  years.  Mortality rates for the entire
cohort were calculated and were compared to the death rate  for males in an
unexposed population in  the  45-64 age group in the area in  which the Amlwch
Factory was located.  The  average ages of  the test and control groups
differed.  It was  the researcher's opinion that this would  provide the most
significant evidence of  any  association  between EDB exposure and increased
morbidity.

    The investigators determined  that among the workers who died,  the incidence
of deaths due to all causes  as well as heart disease,  cardiovascular disease
and cancer of various sites  was in most  cases lower than the control group (45-
64 years of age), or approximately the same as the control  group.   The author
attributed this  lower death  rate  to the  preselection of healthy workers prior
to employment.   The authors  concluded that, "it is difficult to conclude frotn
the data on this relatively  small group  of persons who  have had regular slight
exposure to dibromoethane  that there  is  any association between such exposure
and any particular disease."

    The Agency agrees with the investigators,  and also  finds that  it is
difficult to conclude that there  is no EDB exposure-related cancer  death in  the
workers. (Mishra,  1978a)

                   (4)  Conclusions-Human  Oncogenicity  of EDS

    Based on the evaluation of the  rebuttals  discussed  above,  the Agency
concludes that,  because no respondent has  sustained  the  affirmative  burden of
proof set forth  in 40 CFR  162.11(a)(4),  the presumption of  oncogenic risk to
humans has not been rebutted.  The  Agency  concluded  that the  latter  two  studies
conducted in England are not useful in evaluating the htanan health risk of
exposure to EDB.  The Agency (CAG,  1973) considers only  the first study  listed
above as technically acceptable.   Though it did  not  show a  statistically
significant cancer incidence in exposed  workers,  the Agency concludes  that the
animal data provide substantial evidence that  EDB  is a likely  carcinogen for
huaans.

   2.  Mutagenieity

    In PD-1 the Agency evaluated  all  of the  available studies on the mutagenic
potency of EDB.   These studies, summarized  in Table 4, tested EDB for three
types of nmtagenic effects: point  (gene)  mutations, chromosomal aberrations and
primary DNA damage.  Each of these categories was further subdivided into
positive or negative mutagenic effects.  These are the conclusions reported in
the PD-1:

    "In a memorandum dated 9-10-77, Dr. R.   Pertel  stated that there is ample
    evidence to  fulfill both the multitest  criteria for EDB as a mutagen as
    well as the  scientific criteria of EPA's Science Advisory Board's (SAB)
    study group  on mutagenicity.   This evidence shows EDB to be positive in
    both prokaryotic (microbial)  and eukaryotic (higher forms including
    mammals)  for point ,(gene) mutational  effects, with and without mammalian
    metabolic activation."  (December  14,  1977; 42 FR 240,  page 63141)."
                                        -17-
    418-574 O - 83 - 11

-------
                                       148

    40 CFR 162.11(a)(3)(ii)(A) states  that a  "...  rebuttable  presumption shall
arise if a pesticide's  ingredient(s),.. induces autagenic affects  as  determined
by multitest evidence."  Further, Section  162.3(4)  defines  the  term tnutagenic
as "... the property of a substance or mixture of  substances  to  induce  changes
in the genetic complement of either somatic or germinal  tissue  in  subsequent
generations."  Based on the induction of point mutations by EDB  in several
tests, the Agency concluded that the mutagenic risk  "... index  has been
exceeded by all registrations and applications for registration of pesticide
products containing EDB, and that a rebuttable presumption  against new  or
continued registration of such products has therefore arisen."   (December  14,
1977, 40 PR 240,  page 63135).

              a.  Mutagenicity Rebuttals

    Four respondents submitted rebuttals specific  to this presumption of the
autagenic risk of EDB (rebuttal nos. 3c,47,48,64).   Rebuttal comments included
general arguments about 'the mutagenicity testing systems used and  specific
statements about the studies reviewed in ?D-1.  These rebuttals, together with
the Agency's scientific review and evaluation, are summarized and  responded to
in this section.   Mutagenic studies on EDB which have become available  since
the EDB PO-1 was issued are also evaluated.

                   (1)   General Comments

    The rebuttals which relate to the mtitager ;,city RPAR express four general
areas of concern:  the significance of the testing systems; the limitations of
the tests on EDB? available-information; and  the validity of specific studies.

                        (a)  Significance of  Mutagenicity Tests

    The Great Lakes Chemical Corporation. (Rebuttal No. 3C)  and the  Hopes
Company (Rebuttal No. 64) questioned whether  in vitro (or "test tube")  tests
indicate a potential for in vivo  nutagenicity or carcinogencity:

    "While EDB has produced some mutations in in vitro testing,  the
significance of these results in assessing the mutagenic or
carcinogenic risks of EDB is not established  and subj ect to serious
scientific debate." (Rebuttal No. 3C).

    Ethyl Corporation (Rebuttal So. 43) also  questioned the predictive  utility
of existing test systems.  They mentioned microbial, insect, and cell culture
studies which require extrapolation of results to apply to humans:   "...The
ability of SDB to induce mutations in a variety of test systems Ls  only
suggestive of its potential to induce mutations in the human population..." and
"...there is no evidence to determine the relative risk..." (Rebuttal No. 43).

    In response,  the Agency supports the mutagenicity studies of non-human test
systems.  Data from non-mammalian test systems are relevant in considering the
potential mutagenicity of a compound.  In the absence of adequate mammalian
systems which are sensitive enough to exhibit the many mechanisms of possible
genetic damage in reproductive cells, other test systems are utilized to expose
the chemical's intrinsic capacities to alter  genetic components.  This evidence
has generated a presumption of hazard as stated in the PD-1.
                                        -18-

-------
Table 4  Summary at Mutagenlclty Testing  of  EDB
Genetic Event
and
Test System
Gene (Point) Mutation

Bacteria), In vitro
Bacterial, llost-
Mudiated Assay
(mouse)
Eukaryotic Mlcroblal,
ifi v*tro
Insect, In vitro
Somatic Cell
Culture In vitro
Higher Plants
Higher Plants
Bacterial, in vitro
Bacterial, in vitro
Bacterial, llost-
HudiaLed Assay
(mouse)
Species/Strain/Type

Salmonella typhlmurhro
• his G ^6
TA T530, 1535, 100
S.typhlmurlira
G 46/hls
Netirospora crassa
' ai'-3" '
Drosophlla melanogaster
males
Mouse Lymphoma cells
L517By TK lloterozygotes
TradescantiaL
Mutable Clones
Barley Seeds
S. typhimurlum
his G 46
TA 1538
8. typhlnmrlun)
LT 2
3al-clil A
S. marcescens
A 21/lcu
Concentration
or
Dose

10-10,000 ug
(10 ul)/ plate
500 «*jAg
i.m. to mouse
1.2-1.63 ul/nl
0.3 niM fed
ad. lib. 3 days
1.0-3.0 Brt
for 2, liours
3.6 - 222.9
ppra (gaa)
0.1-1.1 mH
11.5 UH
19 ul/plate
5 ul/plate
500 nq/kg
i.m. to moitsa
Genetic Indicator
or
Effect

Reversions by
Base-Pair
Substltlon
Reversions
Recessive Lethala
Sex-linked
Recessive Lethala
Induction of
TK lloraozygotea
Stamen llalr Color
(blue to pink)
Oilorophyl 1
Mutations
Reversions by
Frame-Shift
Deletions
Reversions
Result References

Positive (Bran, et. al, 1974)
., (Buselwaler, et al, 1972)
m/o H^(McCann, et al. 1975)
Positive (Buselmaier, et.al, 1972)
Positive (De Serres, at al, 1970)
K/o HA (Mailing, 1969)
Positive (Vogel, et al, 1974)
(early gem
cells affected)
Positive (Clivo, 1973)
(dose response)
Positive (Nauman, et al
(dose response) 1974} (Sparrow
et al.)
Positive at (Ehrengerg,
lir0 et al., 1974}
Negative (Breo, et al, 1974)
w/o Mh
Negative (Alper, et al, 1975
w/o MA
Negative (Buselmaier, et al.
1972)

-------
                                      150

    Experimental animals are utilized throughout  the various  disciplines  of
toxicology to predict adverse effects and  the probability of  inducing  certain
effects in humans for a given range of doses of a compound.   Certainly there
exist numerous examples of cases where humans did not display the  anticipated
effect predicted from animal studies.  Hunans have also shown effects  not
demonstrated in the experimental studies,  or have been more or less  susceptible
than the non-human species.  However, the  theorum that "the basic  biological
mechanism explored and documented in experimental studies with non-human
subjects is valid when extrapolated to humans to  predict risk,"  still  holds
firmly, as evidenced by the universal acceptance of non-human experimentation
in science.  (Chaisson, 1978)

    Predicting potential of a compound to  induce mutagenic effects in  humans  is
more difficult than predicting other types of toxicities, but the  logic can
still be founded on establishing the mechanism in appropriate non-human
species.  This concept is discussed at length in the DHEW document by  the
Subcommittee on Environmental Mutagenesis, 1977.  Valid studies  demonstrating
mutagenic effects in experimental animals, and the mechanisms of those effects,
coupled with serious consideration of pharmacologic principles and biodistri-
bution in humans do establish the framework for presuming a rautagenic  hazard  to
humans.  The evidence of mutagenic potential in non-human species  can  be  used
to predict potential for humans.

    The application of test results was addressed by the Great Lakes Chemical
Company: "...None of the test systems used were capable of detecting anything
other than short-term effects of a mutagen.  The  same applies to the
evaluation'of the genetic risk."  (Rebuttal No. 2C).

    The philosophy summarized above has been considered by EPA and is  addressed
in the document by the DHEW Committee to Coordinate Toxicology and Related
Programs, 1977.  The problems in definitively identifying human rnutagens  are
recognized."  However, interpretation of the available test results with the
pharmacokinetic and biochemical considerations, and matching  of results of
reproductive studies can give presumptive  evidence on which hazard to hunans
can be predicted and regulatory decisions  contemplated (Chaisson,  1978).

    Great Lakes Chemical Co. expressed a related point: "Even an established
human mutagen is not likely co have any mutagenic effect unless the exposure
reaches the testes or ovaries." (Rebuttal  No. 3C).

    The Agency agrees completely with this statement.  It is  obvious that in
order to have a heritable effect,  the mutagen must reach the  reproductive
tissue.  As discussed in the next section, the available reproductive  studies
provide evidence that the reproductive tissues in exposed animals can be
adversely affected by EDB.

                        (b)  Limitations of Available Tests

    Hopes Company states that there is a necessity for a  multitest approach to
mutagenicity testing, with strong emphasis on those tests which relate  directly
to man, and that the only real way to test for heritable  mutation in  man is to
utilize a test system with guidelines from the DKEW document  "Approaches to
Determining the Mutagenic Properties of Chemicals:  Risk  to Future
Generations."  The Agency agrees with the  principle set forth in this docizient
that a multiple-system approach is necessary since no single adequate test
exists.  Hopes Company has outlined the optimal criteria,  but has  failed to
                                        -20-

-------
                                      151

cite what the species and  test  protocol would be.   The Agency recognizes the
shortcomings of each test  system but  also  the utility of the tests when they
are considered as a group.

                         (c)  Validity of Specific  Studies
    The Douglas Chemical Company criticized  the  3uselmaier,  1972 study:
"Response data indicated the data  presented  are  insufficient for evaluating the
effect of EDB in the host-mediated assay  with  S^ marcescens, since  results of
only a single dose were reported." (Rebuttal No.  47).

    The Agency agrees that an ideal protocol in  the  host-mediated test would
include at least three dosage levels, with at  least  ten  animals  per dose
level.  Utilization of a different protocol  tempers  the  resulting strength of
conclusions, but does not nullify  the fact that  the  500  mg/kg dose
(subcutaneous) increased the mutation frequency  over the control group by a
factor greater than eight.

    The rebuttor also rejected  the use of the  inconclusive  results  of Epstein
et al., 1972, to support the RPAR.  The rebutter's reasons  for this position
were unclear.  The Agency did not  use this study to  support  the  presumption of
hazard; the  study was mentioned to establish that it was considered,  but that
the results  were inconclusive.

    In addition, the rebuttor disagreed with EPA's rejection of  the results
ofBrem, (1974) and implied that the single dose  used in  the  study was evidence
of no nmtagenic potential.  The rebutter's position  is not adequately
developed.   EPA has considered  this study to be  invalid  because  no  information
on the effective dose or range  is  given.  Also,  the  strain  used,  S.typhimurium
TA 1538, would not be expected  to  predict activity since this strain  is
designed to  detect frameshift mutagens and EDB is more likely to  cause base-
pair substitution mutations.  The  rebutter's position does not rebut  the  RPAR
presumption.

              b.   Agency Position o_n Mutagenicity Rebuttals

    The Agency has reached the  following  general  conclusion  on all  the
rebuttals submitted in argument against the  Agency's position in  PD-1:

    "The methodology currently  used in mutagenesis assay studies  has been
generally accepted on theoretical  grounds and  offers the only means to predict
adverse genetic effects in hunans." [and]  "The comments  in the Rebuttal Nos.
2C, 47, 48 and 64 underscore the limitations of  these test systems  but do  not
rebut the Agency's presumption of  hazard  as  outlined in  the  RPAR  document."
(Chaisson. 1973).

    Thus, the Agency concludes  that the mutagenic risk as stated  in the RPAR
Notice has not been rebutted, since no respondent to that Notice  has sustained
the affirmative burden of proof set forth in 40 CFR  162.11 (a)(4).
                                         -21-

-------
                                     152

              c.   Re-evaluation £f Mutaaenicity  Data  and  Review
                   of New Studies

    In the light of recent research, the Agency has  re-evaluated the
rautagenicity data presented in the RPAR Notice and reviewed  additional  studies
as to the mutagenic potential of EDB.

    In the PD-1 the Agency concluded that  "no positive chromosomal  damage
studies have been found  for EDB."  After a review of the study  by Ehrenberg,  e_t
ai ., (1979), the Agency (Mauer, 1979) has concluded that, in addition  to  the
point (gene) mutations cited in the PD-1,  this study also  demonstrates  that EDB
causes chromosome-damaging effects as discussed below.

    EDB was shown to cause chromosome exchange in barley seeds  which were
treated with 1.4 mM of EDB for 24 hrs.  Cells were analyzed  for chromosomal
aberrations in the first mitosis following treatment.  The observed frequencies
of induced chromosomal aberrations were determined to  be significantly  higher
than the control material.

    Additional studies,  not available at the time of the RPAR IJotice, have
since been reviewed by the Agency (Mauer,  1979).  These studies  lend additional
support to the Agency's  position that there is multicast evidence indicating
that SDB Induces mutagenic effects in laboratory  tesr  systems.   The studies
cited below all show positive effects.  They are  discussed according to the
specific genetic indicator involved, i.e.: (1) point (gene) mutation; (2)
chromosomal aberration;  (3) primary DNA damage.

                   (1)  Point (gene) mutation

    SDB at concentrations as low as 0.2 parts per million has been shown to
induce sex-linked recessive lethal mutations in Drosophila melanogaster (Kale
and Baum, 1973).  Early germinal stages were affected at a rate  3 times as
great as the spontaneous mutation rate, indicating that even lower
concentrations could cause mutagenic effects.

    EDB has also been shown to cause reverse mutations in Aspergillus nidulans
and Sscherichia coli as well as induction of ochre suppressors  in E.coli
(Scott, B.R.,  £t ^1 , 1978).  The potency of EDB was  shown to  increase with
metabolic activation by a plant extract of Tradescantia clone 4430.


                  (2)  Chromosomal damage

    EDB produced micronuclei in tetrads of microsporogenesis of Tradescantia
clone 4430 by a 6-hour treatment with 20-160 ppa of SDB gas (Ma, et al.,
1978).  A range of 0.001-0.002 micronuclei/tetrad/ppm-hr. was established by
do se-e ffect exper iments.

                   (3)  Primary DNA damage

    SDB was shown to indirectly interact with nucleophilic  sites and strongly
inhibit RNA transfectivity via the  production of  "half-suistards" (Xondorosi,
et al.,  1973).   Q -RNA was inactivated by EDB in the presence of thioglycol
after treatment for 1 hr. in a phosphate buffer.
                                         -22-

-------
                                      153
    These recent studies, not available at  the  time  PD-1 was  issued,  reaffirm
the results of earlier citations and  indicate that in addition  to  causing point
(gene) mutations and primary DNA damage, EDB possesses  chromosome-damaging
affects as evidenced by several in vitro assays.

    An additional study investigating the potential  mutagenicity of EDB  in the
rat and mouse has been reviewed by the Agency.  The  study, an investigation of
dominant lethal mutations, concluded that EDB at a dosage of  10 or 30 mg/kg in
mice and 100 or 150 mg/kg in rats did not induce dominant lethal mutations in
either of the species (Teramoto,  et al., 1980).
    It is the Agency's position (Mauer, 1979) that the-  negative results of this
test are no indication of other genetic events  which can occur  as  a result of
the transport of EDB or its metabolites to mammalian germ calls.   The genetic
end-point of this test is gross chromosomal damage,  a relatively insensitive
and limited indicator of total genetic damage.  This test does  not detect  the
variety of genetic damage demonstrated in the other  test systems reviewed  by
the Agency.  Thus,  the Agency concludes that this study is of very limited
significance as compared to positive mutagenicity assays en SDB in multiple
test systems.

              d. Conclusions

    3as«d on the evaluation of the rebuttals and new studies discussed above,
the Agency concludes that no respondent has sustained the affirmative burden of
proof set forth in 40 CFR 162. 11 (a) (4), and that the presumption of oiutagenic
risk to humans from EDB has not been rebutted.  The  Agency (Mauer, 1979) states:

    "Sufficient evidence exists to indicate SDB presents a potential tnutagenic
    risk for humans, especially for individuals exposed to the relatively  high
    levels of the chemical measured in occupational  settings, such 33 during
    soil application and commodity fumigation.  In summary,  the evidence is as
    follows:

         o EDB is a potent mutagen, producing a broad spectrum
         of mutational events in submammalian in vicro  systems,
         higher plant, and_mammalian cell cultures.

         o EDB is mutagenic to germ cells in vivo, as shown by
         the sex-Linked recessive lethal test in Drosophila,
         including significant increases in lethals  in a dose-
         dependent fashion.

         o Notwithstanding the removal of a major portion of EDB
         administered experimentally to mammals, EDB is
         transported to mammalian germ tissues in amounts
         sufficient to produce significant effects as indicated
         by sperraatogenic effects in bulls and rats.   In
         addition,  tissue distribution studies show measureable
         [sic]  amounts can be transported to testicular tissue
         which may result in an accumulation of genetic lesions
         induced by this bifunctional alkylating agent. "(Mauer,
         1979).
                                         -23-

-------
                                       154

    In regard to the suitability of the data to support a quantitative estimate
of mutagenic risk, the Agency asserts:

    "Although there is a strong indication of risk at current  levels of
    occupational exposures, the present data base does not permit
    quantification of mutagenic risks from the levels of EDB exposure
    likely to be encountered by the general population.  As discussed
    by an EPA consultant...no coherent body of relevant data is
    available which could either be used to relate hunan EDB exposure
    directly to mutagenic events in any mammalian test system, or from
    which dose-response relationships can be extrapolated to hunans.
    Further, previous indirect attempts to quantify risk from mutagenic
    studies in Djrosopfrila did not take into account differences in mutational
    responses between ionizing radiation and many chemical rautagens...
    Also, there are basic differences in transport mechanisms between
    mammals and Drosphila which can affect the pharmacokinetic disposition
    of chemicals, especially via the inhalation route of exposure.

    Experimental work is currently in progress for estimating the mutagenic
    risk of EDB... patterned after studies already published on EMS...
    This approach will attempt to overcome the previous shortcomings of
    quantitative risk estimation mentioned above by utilizing a common
    unit of 'dose1 to the germ lines of both Drosophila and male mica
    exposed to radioactive-labeled EDB.  This 'dose1  is defined as the
    amount of chemical bound to germinal DNA and is measured as the nunber
    of alkylationa per DNA-nucleotide.  The 'dose'-mutation curve in
    Droaophila will be compared to.the exposure-'dose1 curve in the mouse.
    In this way it may be possible to derive a quantitative estimate of the
    genetic hazard to hunans from a given concentration or exposure level
    of EDB, since one could estimate the dose (as alkylations) to mammalian
    germ cells from given body exposures, and the genetic consequence of
    that dose in the Drosophila "test." (Mauer, 1979).

         3. Reproductive Effects

    In Position Docunent 1 (PD-1), the Agency cited several studies in bulls
and rodents which established that 2DB could adversely affect mammalian
reproductive development by interfering with the production of male gametes and
the development of embryos.  Adverse effects in hen egg production were also
noted.  In summary, the animal studies reported the following effects:

          Bulls - reduced sperm production; reduced sperm motility;  and
         abnormal sperm structure.  (Amir, 1973; Amir and Lavon, 1976; Amir and
         Ben-David, 1973)

          Rats and mice - selectively damaged spermatogenic cells resulting in
         "transient" sterility; decreased fetal implants and fetal weight.
         (Edwards et al., 1970; Short et al., 1976)

          Chicken Egg Production - reduced egg weight; reduced egg
         production; reduced fertility; generalized reduction in the
         permeability of ovarian membranes.  (See Table 9,  of ?0-1)

    An Agency Working Group concluded that the reproductive effects  studies met
the RPAR risk criterion stated in 40 CFK 162.11 (a)(3)(ii)(B),  that  a
rebuttable presumption shall iriae if a pesticide "produces any other chronic
                                         -24-

-------
                                      155

or delayed toxic effect in test animals  at  any  dosage...which is substantially
higher than that which hmans can reasonably be anticipated to be exposed."
The Working Group also concluded that, because  sufficient data did not exist
for determining a no observable effect level for EDB  via oral, inhalation or
dermal routes of exposures, acceptable levels of exposure could not be
calculated for persons exposed by any of these  routes following the pesticidal
uses of EDB.

    Based on the above evidence, the Agency adopted the  following position in
the PD-1:

     "On the basis of scientific studies and information
     summarized in the Position Docunent [(1)],  the Agency has
     concluded that the risk index  for reproductive effects has
     been exceeded by all registrations  and applications for
     registration of pesticide products  containing  EDB and that
     a rebuttable presumption against new or continued
     registration of such products  has therefore arisen."

    Four rebuttals (Rebuttal nos. 2C, 48, 64, and 79) were submitted in
response to the reproductive RPAR.  The  Agency  has  divided these into two broad
groups: those which addressed uncertainties in  extrapolating to man from  the
positive animal laboratory studies; and  those studies which investigated  the
health of workers occupationally exposed to SDB.

                  a.  Response to Rebuttals of  Extrapolating
                      from Animal Reproductive  Disorder  to Humans

    So rebuttals seriously questioned the validity  of the  animal reproductive
studies evaluated in the PD-1.  However,  all respondents noted the
reversibility of the effects at all but  the highest doses  in  hens and  the
general reversibility of effects in bulls following cessation of dosing.
Respondents argued the effects produced  in  hens  are probably  specific  to  the
hens and that the bull is specifically sensitive to EDB.

    Generally, the rebuttals disputed the significance which  the  Agency placed
on the animal studies in the PD-1.  More specifically, the rebuttals  argued
that the reproductive disorders observed in bulls orally administered  EOS
cannot be used to determine a safe  level of hunan exposure to inhaled  EDB.   The
basis for this argument was that the bull is  unusually sensitive to EDB.   In
addition, the rebuttals contended that it is inappropriate to  determine a
margin of safety by comparing levels to  which hunans  are exposed via  inhalation
with the amount of EDB to which bulls were  exposed  orally.

    The Agency concludes these rebuttals are  not valid and do  not change the
Agency's position.  In the absence of adequate  hunan  data,  the  Agency's normal
policy is to use data from the most sensitive species which in  this case is the
bull.  This procedure is dictated by regulatory  prudence;  in  the absence of
adequate h\snan data it is reasonable to  assune  that man  is at  least as
sensitive as the most sensitive species  for which data are  available.  The
question of assorting that all EDB inhaled is  retained is basically  similar to
the question of the most sensitive  species:   in  the absence of  adequate data it
is prudent to assuae that the worst case applies  (e.g.,  all inhaled EDB is
retained).  Thus, the Agency concludes that it  is in  the  interest of public
health to use the results from the  reproductive  studies on bulls to evaluate
the reproductive risks to hunans.   In further support of this assumption,  the
                                          -25-

-------
                                      156
Agency has recently completed an inhalation  study  on  rats (discussed later in
this section) which demonstrates that  this exposure route also causes
deleterious effects upon the reproductive organs.

               b. Response _to Rebuttals  Addressing Human
                  Studies

    In response  to the reproductive RPAR in  the  PD-1,  two epidemiology studies
were submitted (Rebuttal no. 48).  At  the same time the Agency conducted its
own study (Griffith, et al, 1978).

    The first study submitted in rebuttal was a  retrospective  study  of inferred
fertility of married men occupationally  exposed  to EDB (Wong,  1977).   Basic
marital reproductive histories were collected from five groups of  male workers -
one group from each of four SDB manufacturing plants  in the  U-S. and  a fifth
group from a plant in Great Britain.   For each of  these groups,  expected and
observed births were compared for the  families of  workers exposed  to  EDB.
Basically it was the conclusion from each of these "mini" studies  that expected
and observed human fertility were essentially the  same and thus  there was  no
evidence that EDB had an effect upon human fertility.

    The Agency concludes that this study is  of limited value in  evaluating the
human reproductive hazards of EDB.  The  Agency bases  its  rejection on numerous
inherent weaknesses in the study and the general fact  that negative fertility
studies such as these are a very insensitive measure of the  possible
reproductive hazards of EDB.  Two of the deficiencies  in  these studies, which
were also noted by the author, included: 1)  none of the U.S. groups had
matching controls; and, 2) the data for  the  U.S. groups were compared  to
fertility data published by the National Center  for Health Statistics  which
included non-married women, while the  plant  data did not,  thus tending to
artificially inflate the plant rates.

    The second rebuttal study (Ter Haar, 1973) presented  summary results of  a
sperm count survey of 59 employees of  an EDB manufacturing plant at Magnolia,
Arkansas.  (This study was also referred to  in rebuttal number 3C.).   Workers
were divided into two groups:  those exposed  to air  levels  of less  than 0.5 ppm
EDB and those exposed to levels between 0.5  - 5.0  pom.

    The Agency concludes that this study is of limited value in evaluating the
possible reproductive hazards of EDB.  In general,   the Agency considers sperm
count studies such as this one,  which are negative, to be  insensitive
measuresof the possible human reproductive hazards  of a substance.
Reproduction is comprised of numerous facets, only one of  which is  the sperm
count in males.  Other parameters in males,  including  the  number of abnormal
sperm cells, motility of cells,  and ultimately male fertility,  can  be
affected.  In addition, the potential effects on female reproduction must also
be examined, even if a substance has been demonstrated to have  no apparent
effects on male reproduction.   However, human sperm count  studies which are
postive and were conducted according to scientifically accepted protocols, do
provide adequate evidence to demonstrate that a substance  possesses risks of
disorders to human reproduction.   In addition to these general  considerations,
this particular study has specific technical  flaws, including:  1) the failure
to document or substantiate the  finding of correlation between  sperm count and
exposure index;  and 2) the failure to provide documentation on  how  workers were
chosen for the study,  the percentage of all exposed individuals included in the
study, and the individual exposure estimates.
                                         -26-

-------
                                     157

               c.  Recent Agency Sponsored  Study

    The Agency study (Griffith, et  al.,  1973)  was  undertaken to provide
additional information on potential health  effects of agricultural  uses of EDB,
with particular focus on reproductive  effects.   Semen samples were  obtained
from a total of forty-four individuals  in four  cohorts:   (1)  citrus fumigators
at two locations in Florida;  (2) rice  funigators in Texas;  (3)  citrus
fumigators in Texas; and (4)  a mixed group  of  carrot farmers and researchers in
New Jersey.  Semen samples were immediately analysed for  volume, sperm density,
and motility.  In addition, blood samples were  drawn from all participants and
tested for blood chemistries, testosterone  levels,  certain  hormones and
organochlorine residues (only for the  Florida  and  Texas cohorts).

    Air monitoring for SDB was conducted on a  very limited  basis.   The .limber
of data points were inadequate for  use  in standard statistical  analyses to
determine if SDB had a discernable  effect on observed sperm measures.
Initially, lifetime days of use and use-index  (average pounds  per day  of use)
were multiplied together and  used as a  surrogate measure  of potential  exposure
to determine if an inverse relationship existed with the  measured sperm
counts.  This surrogate was used because _the Agency had very little data on the
exposure to SDB of users.   The Agency has  since concluded  however,  that this
surrogate exposure is an invalid measure of potential exposures  to  EDS.   Decent
monitoring studies demonstrate that citrus  funigators have  significantly higher
EDB exposures (on a yearly basis) than any  other group of EDB  users,  and yet,
this is one of the smallest volune  uses of  EDB.  In contrast,  soil  fumigators
use considerably more SDB, but are  exposed  to  considerably  lower levels of EDB
than the citrus f unigators, because they soil-injected EDB  outdoors  using
tractor-drawn chisels.  Thus, this  surrogate use index (days  x  Ibs.  applied per
day) was a very inaccurate measure of the actual exposure to  EDB for each
cohort.

    The Agency has sunmarized the sperm counts  of  the cohorts  from  this  study
in Table 5.  In addition, in  Table  6, the Agency has  summarized  results
fromstudies conducted by McCloud which  investigated sperm counts for males
from"fertile marriages" and from "infertile  marriages".

    The Agency is reexamining the results from  its  study, but recognizes that
the study will have very limited value to assessing the reproductive hazards of
exposures to EDB.  The study  consisted of a  very small number of individuals (a
total of forty-four) and a much smaller numbers  of  individuals within each of
the 4 cohorts.  Standard statistical test results  which take  into account
age and sex consist of such small sample sizes,  would be of  limited
reliability.

               d.  Recent Agency Animal Study

    The Agency has sponsored  a study in which certain reproductive parameters
were measured in rats exposed to EDB via inhalation  (Short  et al.,  1978).
While a variety of reproductive effects were noted  in this  study, the most
noticable effect was atrophy  of the prostate as  illustrated  in Table 7 below.
                                          -27-

-------
                                                                            TABLE 5

                                                       SPERM COUNT DISTRIBUTIONS  Mf COHORT AKD OCCUPATION
                                                    Hadian                                                 fi
                         Cohort                     Count         Frequency Distribution  ($) at Counta  (10 at) by Cohorts    jfcounta

                     Stata      Occupation          ( x 106/ml)    <\0      10.1-20    20.1-40    40.1-60   60.1-100    >100    _<20 (106/n>1)
00
1                  Florida    Cltrua Punlgatora     53-5           14-3       7.1         14.3      21.4        7-1       35-7         21.4
                              (N=14)
                   Texas      Cltrua Fumlgatora     59-5          5.9        0          23.5      23.5         17.6     29-4         5-9            £
                              (N=17)                                                                                                               0°
                   Texaa      Rice Fumlgntora       76.4          000          12.5         50.0     37.5
                   New Jersey  Carrot  Farmera &    123.0         0          0          0         0           20.0     60.0        0
                               Reaearchera

-------
                                                                           TABUJ   6

                                              SPENM oouwr DiswiuuncNS toa MAI.ES FRCH INFERI-IIS AND FEKTIU HAMUAT,ES
                                                   Median                                                6
                                                   Count         tVaijuency Distribution  (1) of Counts  (10 nil) by Cohorts     iCounta
                                                   (10%H     <10     10.1-20    20.1-40   40.1-60    60.1-100     >100      _<20  (I
10                 Macl*od    (H=9,000>             74            U.9        6.1         11.4       11.8        25.3      36.5        15.0
                  1977*
                             (N-1,000)             90                                  12         12         27        44
                  'Males from infertile inarriayea
                  "Hales frun fertile niarrlcKjes
                  + Cateyorica <10 and 10.1-20 confined

-------
                                     160
                             Table 7

   Lesions in Reproductive Organs of Male Rats Sxposed  to £DB

 For 10 Weeks and Sacrificad at the End of Exposure

Lesion
Test is
Atrophy (Severe)
Calcification
Epididyrais
Atrophy (Moderate to
Severe)
Prostate
Atrophy (Moderate to
Severe )
Prostatitis
Seminal Vesicles
Atrophy (Moderate to
Severe)
EDB
£
a/
0/10^'
0/10

0/10


0/10

0/10

0/10

(ppm)*/
19

0/10
1/10

0/9


0/10

0/10

0/10


39

0/10
0/10

0/10


10/10

0/10

0/10

                                                       as
                                                         6/6
                                                         2/6

                                                         5/6
                                                         6/6

                                                         1/6

                                                         6/6
* ppm in inspired air.

     Number of rats affected/number of rats inspected.

    The results of this study suggest that the prostate is the most sensitive
tissue.  Thus, while the testis, epididymis, prostate and seminal vesicles are
markedly atrophied in the presence of 88 ppm EDB, only the prostate is affected
at 39 ppm.  Based on this study, the NOEL for atrophy of the prostate is  19 cpra
(146.11 mg/m  ) in the rat.

               e. Conclusions on the Reproductive Effects STAR

    The Agency has examined the rebuttals submitted and other pertinent data
and has concluded that the presumption of reproductive disorders as stated in
the RPAR notice has not been rebutted because no respondent to that notice has
sustained the affirmative burden of proof set forth in 40 CFH 162. 11 (a)(4).
While there are no reliable human data available to help assess the human
reproductive effects of EDB, the available animal data clearly indicate that
SDB causes adverse reproductive effects in the bull, rat and hen.  The
inhalation NOEL for EDB in the rat is 19 ppm (146.11 ng/mg"'). based on atrophy
of the prostate.  Based on this NOEL and the exposure levels discussed in the
next section, the Agency concludes that (as was originally stated in PD-1 with
respect to the bull) there is not an ample margin of safety from adverse
reproductive effects for humans exposed as a result of the current pesticidal
use practices of EDB.

         S.   Human Exposure To EDS From Pesticidal Uses

    Human exposures to EDB from pesticidal uses occur through three exposure
routes:  dietary, inhalation by evaporation of pesticides containing ECS, and
dermal by contact with these pesticides.
                                         -30-

-------
                                      161
         1.   Dietary Exposures

              Residues of EDB may enter the  human  diet  as  a  result of she
following raaistsred uses:

    o preplant soil funigation of agricultural  commodities and  the
      possibility of groundwater, and, ultimately, drinking  water
      contamination resulting from this use;

    o post-harvest fumigation of stored grains  and grain products;

    o fumigation of grain mills, mill machinery, and  handling equipment
      (including transportation equipment)?

    o post-harvest fumigation of fruits, nuts or vegetables  for quarantine
      purposes;

    o possibly through control of wax moths  in  beehives and  associated
      equipment.

    In the next five sections, dietary axposure levels resulting front each of
these registered uses are examined.  These data are taken  from  two major
investigations on SDB residues occurring in  the hunan diet:  "Dietary Burden of
Ethylene Dibromide Used as Soil, Grain, Grain Machinery, and Fruit Funigant"
(Holder, 1979), and ""Final Dietary Exposure  Analysis  for Sthylene  Dibromide
Resulting from Residues in Fumigated Crops and  Food Commodities" (Worthington,
1978).

              a.   Residues From Preplant Soil  Fumigation

    The Agency has examined the four known studies which investigated EDB
residues in crops grown in EDB-fumigated soil.  These include the  following:

    o An unpublished series of reports to the Great Lakes  Chemical Corp.
      from Litton Bionetics, Inc. (Rebuttal  No. 3C);

    o An unpublished report from the Ministry of Agriculture and Forestry
      of Japan (Kenichi, 7978);

    o Data compiled by the Food and Drug Administration, DHEW,  from their
      Compliance Program for OBCP and EDB;

    o A report by the California Department  of  Food and Agriculture (CDFA)
      analyzed 29 samples of nine crops grown in EDB-funigated  soils
      in California, with a stated level of  detection of 1 part per billion
      (Maddy, 1979a).

Taken collectively, these studies indicate that residues of EDS in crops  grown
in EDB-fumigated soils are, if they occur, below the level of instrunent
detection.  However, the Agency cannot be certain that residues will not
occur.  Each of these studies has been carefully reviewed by Agency scientists
and each was found to have some deficiencies which precluded the Agency from
categorically accepting their findings.  The studies from both Litton Sionetics
and Japan's Ministry of Agriculture and Forestry used analytical procedures
which wera only sensitive to 0.01 ppm and involved application rates of EDB
which were generally less than standard rates.  The FDA study involved analysis


                                         -31-

-------
                                      162

of foods obtained from retail groceries with no knowledge  that  EDB  was  even
applied to the soil  in which these crops were  grown.   In addition the
analytical method was only sensitive to 0.01 pan.  The CDFA  study (Maddy,
1979a) is probably the most convincing of all  the  studies  to date.   The level
of detection is stated to be 0.001 ppm.  However,  because  the analytical method
used was not validated by CDFA through mass spectrophotometry,  these data  must
be considered as preliminary.

    The fact, that residues of the chemically  related  pesticide DBCP (in the
.001-.01 ppm range)  have been found in crops treated by similar use  patterns,
augments the concern that residues of EDB may  occur.  Therefore,  in  view of
these findings, the  Agency assumes that treated crops can  bear  EDB residues up
to the limit of detection, which the Agency considers to be  0.001 ppm.

    Using the 0.001  ppm residue limit, the Agency  has calculated  the  potential
body burden which may result from soil fumigation  uses (Table 3).  The  body
burden is calculated in mg EDB/kg body weight/day, assuming  a 1,500  g diet  and
a body weight of 60  kg.  The 60 kg body weight value is considered to be the
average for an individual from a family of four.   The food factors (the  average
percentage of the human diet) listed are the values reported  in the  update of
the Food Factor Tables (Schmitt, 1973).  The percentages of  crops currently
treated are docuner.rad by L. Zygadlo (1978) and the OSDA/State/EPA Assessment
Report (1978).  Th^  values for the estimated percentage of'crops  treated with
EDB, if all DBCP uses are cancelled, are based on  the portion of  D3C? uses  for
which EDB is an alternative pesticide.  The calculated dietary  burden (DB)  per
day per person is derived from the calculation:

    DB(ng/kg/d?y) »  ppm Residue x % of the Human Diet x
                         100

          kg Diet   x  % Crop Treated
         kg Body Wt          100

    As indicated in Table 8, the estimated average daily intake of EDS from
residues in crops at current use levels is 2.59 x  10   mg/kg  per day.  If it
is assumed that 100% of the crops bear EDB residues,  the estimated dietary
burden (DB) is 4.1 x 10   mg/kg per day.  This latter figure  is appropriate
if an individual eats 100% treated crops.

    Site preparation for fruit tree planting,  an additional registered preplans
use of EDB, is not listed in Table 3.   This is because no detectable residues
have been found in the fruits,  and also because the Agency believes that it is
extremely unlikely that this fruit would contain residues,  due to the high
volatility of EDB coupled with the many years between EDB application and first
fr^uit harvest.

              b.  Postharvest Fumigation of Stored Grains

    Grains are usually treated  post-harvest to prevent insect infestation.
Liquid mixtures of organic solvents such as carbon tetrachloride (CT), ethyLene
dichloride (EDO, and/or carbon disulfide (CS_) are commonly used for this
purpose.   SDB generally constitutes inthe range of 1% to 9% of the weight of
the lioTiid mixture.   A mixture  of chemicals is used because ED3  alone would net
penetrate sufficiently to the bottom of the storage bin.  CT is  an effective
                                         -32-

-------
                                                                              Table a
                           Calculated dietary burden (CD) of tXB residues in the human diet resulting from U>e noil funtgatlon uses.-
LJ
 I






Crop
Asparagus
Utocooll
Carrot a
Cauliflower
Com, sweet
Cotton
Cucuiitoers
Egyplant
latluca
Lima Beana
Melons
Okra
Pa ran i pa
teanuta
tempera
Pineapple
Kjtatoea
$JUaSh
Strawberries
Sweet Potatoes
'Juiitdtoea
Crop
Ibleranoes
of
Inorganic
Bromide
Icon Soil
Incorporation
(Ppm)
10
75
75
10
50
25
30
50
30
5
75
50
75
25
30
40
75
50
5
50
50


Haxiuun
Expected
Residue
of EDO
Jppw)
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001


t of diet
matte up of
Currently
Treated
Cropa
HfcUL
9.5
21.0
3.2
Ntcr.
0.2
28.0
1.4
0.2
45.0
11.0
71.0
NBGI.
0.1
NBGL.
27.0
1.1
MflGL
4.4
5.4
6.8



Hood
Factor
of diet

0.14
0.10
0.48
0.07
1.43
0.15
0.71
0.03
1.31
0.19
2.00
0.07
0.03
0.36
0.12
0.30
5.43
0.03
0.18
0.40
2. 07


CO If all of
Crop Consumed
Dears Residue
(ing/kg/day)

3.50 » 10~n
2.50 x 10 ~*
1.20 x 10~'
1.75 x 10~"
3.58 x 10 '
3.75 x 10 *
1.78 x 10"'
7.50 x 10 ,n
3.28 x 10 ""
4.75 x 10 "
5.00 x 10"'
1.75 x 10 a
7.5 x 10 *
9.00 x 10 °
3.00 10 „
7.5 10"°
1.36 10"*
7.50 10~*
4.50 10""
1.00 x 10" '
7.18 x 10-7


Estimated ECB
at Current
Due tevel
{wq/kg/fay}

0 -9
2.38 * 10 a
2.52 x 10_d
5.60 x 10 1U
0 -a
7.50 x 10 "
4.97 x 10 "
1.05 x 10 |!J
6.55 x 10 *u
2.14 x 10~"
5.50 x 10 ~"
1.24 x 10 °
0 ,,
9.00 X 10~"
o a
2.03 x 10 °
1.49 x 10 U
0 „
1.98 x 10~"
5.40 x 10 B
4.38 x 10-8
                                                                                Total
                                                                                               4.1
10
                                                                                                        .-6
                                                                                                                  2.59 x 10
                                                                                                                           r7
                                                                                                                                        05
                                                                                                                                        CO

-------
                                      164
"carrier" and EDC is added to the mixture to create  the  proper vapor  pressure.
A typical example of a commodity funigant is Dowfune E3-5, which contains CT,
EDC, and EDB in 63:30:7 proportions.

    The data available to the Agency demonstrating residues  in EDB-treated
grains are taken from a variety of published papers which report a wide range
of SOB residue levels.  This broad range results from numerous factors,
including the type and condition of grain, construction variation in  storage
bins selected, EDB application methods and rates, aeration and sampling
techniques, and other related factors.  Table 9 summarizes the ranges reported
in the available studies*

                                        Table 9

                 EDB residues (ppm) Reported in Fumigated Grains *

                     Aeration Period (Weeks) after.Fumigation
    Grain       0             1            4             13


    Wheat     12-1500       1-240         5-33       0.6-5.3
    Corn      37-1565       21-61         2.5-531    0.3-3.5

    Sorghum   <1-5.7        14-45         2.1-15     0.9-3.5

    Barley                  2.2-6.3       1.-2.2     0.9-1.9

    Rye       20-32

    Oats                 10-15 "several weeks after fumigation"
    * (Worthington, 1978)

    In spite of the wide variation in values reported, a consistent pattern of
decline, which will be discussed later in the text, is apparent from the daca
presented in Table 9.

    Cf the grains for which EDB is registered as a fumigant, only wheat and
rice make up appreciable portions of the human diet.  Wheat forms by far the
larger portion of the human diet than does rice:   10.36% for wheat and 0.55%
for rice.  Therefore, persistence of SOB in wheat  is expected to have a much
greater impact on the human- diet (Holder, 1980).   Accordingly, the Agency's
analysis will focus on EDB residues in wheat, flour and bread.

    It is important to note, however, that human dietary exposure can result
from the use of SOB on other grains and rice.  Corn oil, corn aeal, corn
cereals, othar grains, and rice can contain residues of EDB as a result of the
fumigation of stored corn, although most of the stored corn fumigated with EDB
is used as livestock feed (DPRA, 1973).  The Agency recognizes that these other
uses could result in secondary contamination of the human diet through residues
of EDB in neat, milk, rice and other products.  However, the Agency has no data
to support or dismiss this hypothesis, and therefore considers it to be a data
                                         -34-

-------
                                                 FIGURE I
                                   TIME-LINE OF WHEAT AND WHEAT
                             BY-PRODUCT DISTRIBUTION WITH EXPECTED
                                      TIMES  FOR EDB FUMIGATION
                    -ON FARM STORAGE
    1 Ynnr
                       8 rr.y.
                        I
                      (average)
i
CO
en
I
                      r,ob*.;x
                          *
                  -OFF FARM-
                   STORAGE
                            PROCESSING-
^. COMMODITY
  \ DISTRIBUTION
                                              1 mo.
                                                    1 wk.
JT""1
i.nu

GRAIN
SALE

                                                                                   HUMAN FOOD
   'Broad
   Cereal*
   PreUnls
   Noodles
   Cookies
   .Cakes
                                                                                                          05
                                                                                            Constimptloit
                                                                                             As Tort of
                                                                                            Dally Dletnry
                                                                                              Intnkn
                                                                                          (Along with nosldiies?)
                                                                               ANIMAL FEED
                                                               GRAIN EXPORTED

-------
                                      166

gap.  The presence of SDB residues, resulting  from  fumigation of these other
commodities, would increase  the estimated  SDB  dietary burden for its general
use as a fumigant for stored grains•

    Because residues of EDB  entering  the human diet from its use as a funigant
of stored wheat appears to dwarf its  use as  fumigant  for all other  grains,  the
Agency has focused its efforts on estimating the  quantity of EDB contained  in
fumigated wheat and wheat products.   The quantity of  EDB in  wheat,  flour  and
bread varies considerably as a result of variations in the amount of EDB
applied to stored wheat, duration between  application and final  use of product,
type of storage facility used, and  numerous  other considerations.   In order to
determine the amount of EDB  entering  the human diet as a result  of  these
registered uses, the Agency  has considered all the  available data on residues
of EDB in wheat, flour and bread, obtained additional residue levels on these
commodities through in-house studies, and  developed several  mathematical  models
which estimate expected EDB residues  in these  commodities  at various points in
movement through commerce.

    Figure I (Holder, 1980)  illustrates the  typical movement of  a shipment  of
grain through commerce.  The top line indicates the general  areas through which
grain moves.  Below this line are the Agency's estimated average periods  for
which the grain is held at each locus in commerce;  the occurrence of SDB
fumigations which the Agency has determined  will  be applied  during  the grain's
movement through commerce; and a chronological list of the commerce  areas
through which the grain moves.  As  indicated in Figure I,  the  principal methods
of grain distribution are export, animal feed,  U.S. mills  where  grain  is made
into flour, and U.S. bakeries where flour  is further  processed to various baked
goods•

    The fumigation schedules shown  in Figure I were used by  the  Agency (Holder,
1980) to estimate the concentrations of EDB  in wheat  and wheat products at  each
major point in its movement through commerce.

    In addition to these time estimates, the Agency (Holder,   1980) developed
three mathematical models (using results from  laboratory studies) of the loss
of EDB residues from wheat, wheat flour, and baked goods following post-harvest
application of EDB.  Using the schedule shown  in  Figure  I, the Agency  (Holder,
1980) calculated the "probable" concentration of EDB.in bread  to be  0.07 nob,
from which the resulting dietary burden of 1.8 x  10   mg/kg/day was  estimated.

    This same procedure was used to develop  a  "realistic worst case" estimate
of the concentration of EDB at each point  in commerce, and the dietary burden
which would result from these concentrations.  Using the same residue  loss
curves (RLC), but assuming a maximum application rate of SDB  of 7.2  lbs/1000
boishels (instead of an average of 5.1  lbs./1000 bushels) and  minimum treatment
to market intervals,  the Agency estimated the residue levels  of EDB   in bread
and baked goods to be 31 ppb and an associated dietary intake of 3.3 x 10~3
mg/kg/day.

                   c.   Fumigation of Grain Milling Machinery

    In addition to residues of EDB that occur in wheat and wheat  products  as a
result of grain storage fumigation,  residues in many wheat products  also result
from the use of EDB in spot application to  flour mill machinery and  as a
general space fumigant in these mills.  The Agency has evaluated  the three
known studies (Wit,  1969,  1977,  1973)  which have investigated the contribution

                                        -36-

-------
                                       167

of flour mill and machinery  fumigation  to SDB  residue levels in wheat flour and
baked breads (Holder,  1980).  These  studies  generally involved sampling for SDB
concentrations in incoming wheat,  processed  flour,  and bread baked from that
flour*  Each of these  sets of samples was taken,  at  various tiaes after spot
fumigation.

    Because the EDB concentration  in the  incoming wheat varied greatly from one
study to the next, it  is difficult to draw parallel conclusions from these
three studies.  The first two studies demonstrated  that incoming washed wheat
contained EDB, and that the  resulting flour  absorbed EDB from the spot-
fumigated equipment (Holder, 1980).

    In the Agency's opinion  (Holder, 1980) the third and most recent study had
such high levels of SDB in the incoming wheat that  it probably "overshadowed
any pick-up of SDB residues  from the milling machinery due to spot fumigation."
However, this study did show an increase  in  EDB  residues in flour and bread up
to 3-4 hours after spot funigation.  Using a time-weighted average of the
differences between input wheat and the flour derived from the wheat,  the
Agency estimates that  an average of 3 ppb of EDB persists in bread as a result
of spot fumigation (Worthington, 1978).   In  the  absence of the quantity of
wheat flour processed  annually through mills where  EDB-treatnient was used, the
Agency has assumed that all  flour mills fumigate with EDB.   Using the residue
level of 3 ppb, the Agency (Holder, 1980)  estimates the dietary burden from
spot fumigation for all wheat by-products to be  0.76 x 10   mg/kg/day,
assuming all wheat by-products are equally contaminated with EDB.

    Combining the dietary burden (OB) obtained for  spot fumigation with the DB
estimates for EDB used on stored grain, the  Agency  proj ects that the  total
contribution to dietary burden from residues in  wheat and flour ranges from
0.24 x 10   to 4.1 x 10   mg/kg/day for all  wheat fmigations (Holder,
I960).

    In addition to these derived dietary  burdens for all SDB fumigations,  the
Agency (Holder, 1980)  has summarized data on residues of SDB measured  in breads
made with wheat treated with EDB.  A total of 57 samples from 5  Independent
studies yielded an average observed residue  level of EDB in bread  of  28  ppb
with a corresponding dietary burden of 4.1 x 10   mg/kg/day.

    The Agency has efforts underway to obtain additional data on residues  as
they actually occur in commerce.  The Benefits and  Field Studies Division  of
OPP, in cooperation with the USDA's Federal  Grain Inspection  Service,  has
analyzed flour samples obtained from several of  the  USDA's  programs  for
distributing flour to domestic charity programs,   including  the  School  Lunch
Program (Marshall, 1980).  Preliminary results of these  analyses for SDB
(Bbntoyan, 1980) confirmed EDB residues in all of the  samples  analyzed  in  the
first phase of this study.

    Because this DB is derived from actual measurements  of  EDB  residues
observed in bread, it  includes SDB residues  resulting from  both  spot  fumigation
of grain milling machinery and grain storage ftmigation.

     It is-worth noting the markedly good  agreement  between  the DB  of
4.1 x 10   determined  from actual  bread samples  and  the  projected  maximum  DB
of 4.1 x 10  . derived  from the assumptions and calculations  discussed  above
                                         -37-

-------
                                      168

(Holder, 1980).  Thus, the Agency concludes that the total dietary burden
resulting from EDB fuaigation of wheat in storage and of  flour machinery is
best estimated from the data collected from actual bread  samples, which is 4.1
x 10   mg/kg/day.

    For an "average" 60 kg individual in the U.S., this dietary burden
translates into 2.5 micrograms of EDB consxmed per day from the fxanigation uses
of EDB on stored grains.  As discussed above, only funigated wheat and wheat by-
products were considered.  The estimated dietary burden may slightly
underestimate the actual DB resulting from the commercial use of EDS to
fvanigate stored grains.  Considering the evidence currently available to the
Agency, 4.1 x 10   mg/kg/day is the Agency's best estimate of the dietary
burden resulting from all EDB fumigation.

              d.   Quarantine Fumigations

    Several agricultural fruit and vegetable commodities  are regulated by
federal and interstate quarantines.  These regulations were enacted to protect
the importing state from the introduction of new insect or disease pests.
U.S. quarantine treatments are legally mandated by authority of the Plant
Quarantine Act of 1912 as amended (7 D.S.C. 151-167) and  the Federal Plant Pest
Act (7 U.S.C. iSOaa,  iSOjj) through quarantine administrative instructions, and
regulations.  The U.S. Department of Agriculture's Animal and Plant Health
Inspection Service (APHIS) is responsible for carrying out these mandates.

    DSDA regulations require treatment of imported commodities and of certain
commodities transported out of quarantined regions (Table 10).

                                     Table 10

             U.S. Quarantine Requirements for. EDB Treatment

   Regulation                Commodities              Exporting Countries

7 cm 318.13-45         bitter melons, bananas,       Hawaii
                        litchi nuts, cucumbers,
                        papayas, pineapples,
                        cayenne, string beans,
                        zucchini squash

7 C?R 318.58-2d         mangoes                       Puerto Rico

7 CTR 319.56-2c         oranges, tangerines,          Mexico and
                        grapefruits                   certain Central
                                                      American countries
7 CFH 319.56-2J         mangoes, plums                Mexico,
                                                      Guatemala

    As a consequence of the U.S. quarantine requirements, EDB is. registered =is
a commodity fumigant for post-harvest control of insects infesting citrus,
tropical fruits (including mangoes, papayas, litchi fruit and nuts, and
guavas), pineapples, bananas, cherries, plums, peppers, cucunbers, string
beans, and zucchini squash.  For many of these commodities, SDB is the only
fumigant authorised by the USDA regulations.  In addition, certain states also
                                         -38-

-------
                                      169

require fumigation of interstate  transported  commodities.   In California, the
State Administrative Code  (section  3252)  restricts importation of several
citrus fruits from Puerto  Rico and  Florida.   These commodities may only be
received in the Stats after being fumigated at  the point of origin in
accordance with APHIS reaulations.
    EDB treatments are carried out  in  fumigation  chambers as specified in
Quarantine 13  (7 CFR 318.13), Quarantine 58  (7 CTR  318.5S),  and Quarantine 56
(7 CFK 319.56).  Quarantine specifications call for the  use  of technical grade
EDB of at least 97% purity, which is typically introduced into the  fumigation
chamber in the liquid state onto a  home-variety electric frying pan 'for
volatilization.  This is the method used in  Florida.   An alternate  method,  used
in Texas, involves the dripping of  liquid EDB  in  front of a  high speed fan.
Because of the high density of EDB gas  (more than six  times  heavier than air),
forced circulation within the chamber  is also  required for volatilized EDB.
The exposure period for both methods is two  hours from the time all the
required EDB has entered the chamber.

    Over 99% of the 83,500 pounds of EDB used  in  1977  for quarantine  purposes
was applied on citrus and tropical  fruits (EPA, 1979).   The  remaining amount  of
EDB, less than 850 pounds, was used on  an estimated 750,000  pounds  of
miscellaneous  fruits and vegetables which constitute only niniscule percentages
of the total U.S. production of each of these  commodities (Zygadlo, 1978).  The
Agency recognizes that some SDB food residue exposure  can occur from  these
minor uses.  However, the national  dietary burden which  results from  these
minor uses is  negligible and insignificant in comparison with  that  expected
from EDB fumigation of citrus and tropical fruits (Table 13).   Thus,  these
minor,uses are not discussed further, although a maxiinvsn dietary burden of  IS.8
x 10   mg/kg/day (Holder, 1980) could be expected if  100% of these  minor
crocs were fumigated with EDB or an individual ate  100%  of SDB-treated food.

    There are  some monitoring studies available on  the residues of  EDB in
various fruits and vegetables resulting from quarantine  fumigation  with EDB.
The studies show that, while EDB penetrates  these commodities  more  rapidly and
completely than it does grains, it also dissipates  more  rapidly from  them
(Worthington,  1973).

    These limited data were compiled by the  Agency  (Worthington,  1978)  to
estimate the average EDB residues occurring  in citrus  and  tropical  fruits one
day after fumigation.  The Agency (Worthington, 1978)  estimated the average SDB
residues occurring in citrus and tropical fruits one day after  fumigation to be
8.0 ppm and 5.0 ppm, respectively.  Using these data and a residue  loss
equation, derived from the longest half-life time of two days,  residues it any
tinra (t) can be calculated.  See Table  12 for calculated results.

    Subsequent to these estimates the Agency obtained  data from the State of
California (Maddy, 1979b) on residues of EDB found  in  citrus and tropical
fruits collected in 1979 from wholesale produce markets  in that state.   These
fruits had beer, fumigated "to comply with OSDA quarantine  directives  prior to
entry into California."  The Agency (Worthington, 1979)  has  evaluated  these
data, and summarized them in Table  11.
                                         -39-

-------
                                       170

                                     Table  1 1

Measured EDB Residues in Fruits Obtained from California Wholesale
Produce Markets

Fruit     Period after     Number of         Range of  EDB       Average  ED3
          Fumigation       Samples Taken     Residues          Residue Level
           ( days )                            ( ppm }               ( ppm )

Papayas  3 to 16              24             0.13 to  1.24         0.69

Oranges  3 or 16               9             0.06 to 6.35         r. 37

    The Agency (Worthington, 1979) has analyzed these data to  determine  if  they
support its estimate that one day after fumigation the EDB levels in citrus and
tropical fruits are 8.0 ppm and 6.0 ppm, respectively, and has concluded that
they do support these estimates.  3y taking  the obser~ 3d residue levels  and the
duration of these levels after fumigation, the Agency  (Worthington, 1979) has
extrapolated to one day after fumigation using dissipation equations.  These
results generally cluster around the SDB concentrations one day after
fumigation as previously estimated (Worthington, 1973).

   In order to calculate the most likely residue levels occurring in these
commodities at the retail market, the Agency (Panetta, 1979) has obtained
specific estimates of both the minimum and the average treatment-to-market
intervals.

    These time intervals and the initial residue levels were put into  the
residue loss equation discussed above to obtain estimates of residue levels  in
citrus and tropical fruit at the retail market (Table  12).

                                      Table  12

Estimated EDB Food Residues Resulting From Required Quarantine Fumigation

Commodity          Estimated Initial                  Estimated Market
                     Residue Level                     Residue Levels
                        (ppm)                      a/      (ppm)   b/
                                            Maximum"       Average"
Citrus
Trooical Fruits
8
6
1.42
0.76
0.425
0.32
a/       Maximum residue levels assume a minimum treatment- to-market interval
of 6 days for citrus and 7 days for tropical fruits, b/average residue levels
assume 7 and 9.5 days, respectively (Holder, 1980).

    The last t-wo steps in calculating the DB for the U.S. population resulting
from EDB fumigation of citrus and tropical fruits are to multiply the estimated
residue levels from Table 12 by the percent crop treated, and multiply this
product by the food factor (i.e., % of human diet which the commodity
represents).  The percent crop treated and consumed in the U.S. was first
estimatad to be 1.7% for citrus and 72% for tropical fruits (Sygadlo, 1978).
                                        -40-

-------
                                       171

The estimate of the percent of crop treated  for citrus  was amended (Luttner,
1980) to 2.6% because the original estimate  did not  include citrus which is
fumigated and imported to the U.S.  The  estimates  of the  food factors are 3.3%
and 0.03% respectively for citrus and  tropical  fruits.

         Accordingly, che following DB estimates have been calculated:

                                  Maximum DB          Average DB
                                  ( mg/ kg/ da y)         (mg/ kg/ day)

Citrus                            3.5  x  10~           1.05 x 10~5

Tropical Fruits                   0.41 x  1 0~5         0.17 x 10~5

              e.   Control of Wax Moths  in Beehives  and
                   Associated Equipment

    There are two Federally registered products  containing EDB  for controlling
the wax moth (Galleria mellonella) in  stored honeycombs.   [The  label  of  one
product states that it is to be used on  "...empty  combs and ampty  hives" (Reg.
No. 10092-1), while the other label gives directions  for  use,  "When bee  combs
and/or hives are stored for the winter..."   (Reg.  No. 5785-46)].   No  tolerances
have been established for either organic  EDB or  the  inorganic bromide  ion
residues for the use of EDB on honeycombs, because both registrations were
considered nonfood uses.   Although these labels do  not specifically  prohibit
the use of EDB on honey or comb wax, it  is not normal practice  among beekeepers
to do so.  There are no known data on  the potential  for indirect EDB
contamination through translocation to edible honey or  beeswax  as  a result  of
fumigating stored comb-bearing frames  or  comb foundations.   The Agency believes
that no residues in honey are likely to result.  Consequently,  this registered
use is not expected to contribute to the  dietary burden of EDB.

              f.   Other Registered Uses  of SDB

    The remaining registered uses of EDB  involve non-food  uses  and  are not
included in the dietary burden estimates.  These uses include vault fumigation
of stored clothing, structural termite control,  control  of bark beetles  in
coniferous trees, quarantine treatment of regulated articles  for Japanese
Beetle (including soil, plants with roots, and grass sod),  and  parasitic
nematode control (including nursery plots, greenhouses,  and  potting soil).

              g.   Summary of SDB Dietary Exposures
                   Resulting from Registered Use Practice

    Table 13 is a summary of the daily dietary burden resulting from each of
the food uses of EDB. _This table shows the dietary burden  at current use
levels to be 1.21 x 10   rag/kg/day and 6.31 x 10   mg/kg/day if 100% of  the
crops on which EDB is used were treated.   Included in this table are the
relevant input data used to develop the dietary  burden at current use levels.
These include the average treatment-to-market interval,  estimated EDB residue
level at the time of entry into the human diet,  the percent of  total human diet
made up by the given commodity, and the current  percent  of the commodity which
is treated with ZDB.  As shown in this table, fumigation of  stored  grains and
post-harvest fumigation constitute over 90% of the dietary burden which  results
from registered EDB use practices.  The  table does not  include  soil fumigation,
where no residues have been found.

                                         -41-

-------
                                          table 13
Total Dietary burden (DB) of EDB residues in the human diet resulting fro* all commodity fuaigation uses*
Average
Treatment to
Market Time
(days)
Commodity
1.

2.

3.
4.
5.
6.
7.
a.
9.
10.
11.

All
Grain Products
(Bulk Fumigation)
Grain products
(Spot fumigation)
Tropical Fruits
Citrus
Cherries
Pluma
Peppers
Cucurbits
Bananas
String Beans
Pineapple

218

1.0

9.5
9.5
5
5
5
5
5
5
5

Eat. Residues
at Time of Entry
into the
Ikman Dietary

-------
                                       173

          2.    Inhalation  and Dermal  Exposures

    All registered pesticidal uses of SDB  may result in inhalation exposure.
Applicators may be exposed  during  transfer,  mixing,  loading and actual
application operations, calibration and  repair of equipment,  and reentry into
fumigated structures or fields.  Other workers noc involved in the actual
application of  SDB may be  exposed, including those working at or near grain
storage sites and mills which have been  fumigated, citrus quarantine fumigation
stations, apiaries where fumigation has  been used to control  wax moths in
beehives, and fields fumigated with SDB.   The general public  is not likely to
experience significant inhalation exposure,  although seme exposure may occur in
areas near quarantine fumigation stations, bark beetle control oper.ations,  and
structures which have been  fumigated  for termite control.  The major use
patterns (for which quantitative exposure  data are available)  involving SDB
inhalation exposure and risk estimates for specific  populations at risk are
presented in the following  discussion.  They are:   a. preplant soil fumigation,
b. citrus fumigation, and c. grain mill  fumigation.

                   a.   Exposure from Preplant Soil
                        Fumigation

    Ethylene dibromide is applied as  a liquid pre-plant soil  fumigant from
tractor-mounted tanks or cylinders, using chisel injection  or  plow-sole nethorfs.

    The injection chisels or plow soles  normally remain in  the earth as the
tractor moves across the field, and are  removed  only when the  applicator turns
the tractor around at the end of each pass.   A small quantity  of EDB may be
released above  ground when  the chisels or plow soles are  removed.

    When EDB is applied with chisels,  a  float,  cultipacker  or  roller is dragged
behind the chisels to seal  the soil.   When plow-sole applications  are  used,  the
chemical is metered in front of the plow on  the  soil surface and immediately
covered with soil by the plow behind  and to  the  side of the treated  area.   Less
than 5% of EDB  is applied using the plow-sole method.   However,  in  the
preparation of  pineapple fields, SDB  is  soil  injected under a.  plastic mulch
which prevents  rapid volatilization.

    Tractor operators and other applicator personnel may  be exposed  through
volatilization  of EDB before and/or after the sealing of  the soil.   In
addition, exposure may also result from  the  calibration,  cleaning or  repair of
equipment, from equipment failure, and from  transfer  and  loading operations.

    SDB is generally applied by open  system operations, in which the pesticide
aiust be transferred from 30 or 55 gallon drums by  hand, powered pumps or
gravity flow devices into tanks mounted above the  injection chisels.  There is
potential for applicator exposure by  inhalation of vapors and skin contact with
the liquid during these open system transfer  operations.

    There is also a closed  system method of  transfer,  in  which SDB Is fad to
the chisels or  plow soles from cylinders prepackaged  with valves and diptubes,
or containing a convenient orifice for attachment of  suction tubes.  The Agency
is currently studying (but  has not yet determined) the  effectiveness of closed
system technology in reducing worker  exposure  for many chemicals, including SDE
(Hittelman,  1980).
                                         -43-

-------
                                      174

    The time duration for transfer operations and application varies, but  it  is
roughly 10 minutes (0.17 hrs) for each transfer operation and 1.75 hrs  for each
application operation (Mittelman, 1980).

    The Agency has estimated the annual exposure duration for total soil
funigation operations by site.  These estimates are presented in Table  14.  The
estimates assume that transfer operations occupy 10% of the total application
time.

                        1) Inhalation Exposure

    There are relatively few measurement data regarding applicator exposure to
ECB.  The only sources of information are reports by the California Department
of Food and Agriculture (Maddy, 1979c), Dow Chemical's Industrial Hygiene
Laboratory (Vaccaro, 1976), and the Great Lakes Chemical Corporation (White and
McAllister, 1977).  The Dow and Great Lakes studies were submitted as parts of
the rebuttal responses of these two manufacturers (Rebuttal IIos. 78 and 3C).
These studies have been analyzed by the Agency (Mittelman, 1980).

    Column 1 of Table 14 identifies the soil fmigation operation or location
for which EDB air concentration was measured.  The only measurements Listed arp
those in which individual activities were monitored, in order to properly
assess worker exposure.  Air samples were taken during application alone (with
open and closed cab), during transfer operations, while respiratory equipment
was in use, and at sites on the treated field during application and up to 48
hours afterward.

    The average EDB concentration measured in air (milligrams of EDB per cubic
meter of air) at each of the sampling sites is listed in colunn 2.  The use of
closed cab tractors and closed transfer systems apparently did not reduce
exposure to applicators involved in these studies; however, closed system
research is currently being carried out, and insufficient data exist at this
time to evaluate the 'future impact of closed cabs and transfer systems on
applicator exposure.

    The range of concentrations measured in each location and the nunber of
samples taken are listed in columns 3 and 4.  The greatest range of
concentrations was measured during application activity in an open cab tractor
where the exposures ranged from 0.008 to 4.75 mg/m .  No range is listed for
closed system transfer operations because a single sample was taken.

                   2)   Dermal Exposure

    The CDFA study (Maddy, 1979c) included measurements of SDB levels in hand
washings taken from tractor drivers.  The five samples taken ranged from 0.52
to. 30.90 ug, with an average of 3.38 ug.  Actual dermal contact incurred during
transfer of EDB was not measured; however, the Agency assunes that the actual
dermal contact incurred during normal operations will be at the same or higher
levels than those measured from the hand washing samples because some EDB  is
expected to have evaporated or penetrated the skin (Mittelman, 1980).

                   3)   Combined__Inhalation and Dermal
                        Exposure

    The Agency (Mittalman, 1980) has estimated tor.al inhalation and dermal
exposures  for private (non-professional) persons who both apply and transfer
                                         -44-

-------
                                                  Table 14
A
in
Exposure Situations Average
Application/Closed cab
Application/Closed cab
charcoal filtered intake
Application/open cab
Transfer operation/open
system
Transfer operation/closed
system
Area samples at field level:
(A) During application
(B) 24 hrs post-application
(C) 48 hrs post- application
concentration
in air
(mg/m3)
1.66
1.27
0.81
3.03
5.60
0.17
0.02
0.07
Range Number of
(mg/
-------
                                      176

SDB during soil fumigation.  These,estimates, contained  in Table  16, ass one  a
constant respiratory rate of 1.2 m /hr for light physical activity  (No  heavy
activity breathing rates were used), a constant ambient  air concentration
(average of monitored values), and a range of exposure duration  (see Table
15).  Exposure estimates have been made  for  each use site (colunns  1 and 2).

    The total respiratory intake of EDB  (milligrams per  year)  is  presented in
colunn 3 as the sum of exposures resulting from transfer and application
operations.  It is assuned that the applicator performs  the transfer operation
himself; 100% retention of inhaled EDB is also assuned.  Total respiratory
intake has been calculated in the following  way, using data given for peaches
as an example:

Inhalation Rate x (Transfer Exposure +• Applicator Exposure) - Total Exposure
For peaches:  1.2 m /hr [(0.06 hrs._x 3.03 mg/m ) •*•
             (0.54 hrs. x 0.81 mg/ra )] » 0.7 mg.

    Total estimated dermal exposures are presented in column 4.  An average
dermal exposure value of 4.53 micrograms per hour is assumed (Maddy, 1979a).
This value is based on EDB levels in water from hand washings, and assumes chat
the hand contact is the largest source of dermal exposure.  Therefore,  the
dermal exposure estimates given here probably underestimate actual total dermal
exposures.

    One hundred percent of the EDB remaining on the skin is expected to be
absorbed, based on data for DBCP, a chemically similar compound (Mittelnan,
1980).  Total estimated dermal exposure  has been calculated by multiplying
average dermal exposure by the duration of exposure.  The following example,
for peaches, uses figures from Table t5.

 Dermal  x (Transfer  -*•  Application) =* Total Exposure

4.53 ug/hr x (0.06 hrs. + 0.54 hrs.) » 0.003 mg

    The total estimated annual exposure  (milligrams per year)  is listed
according to use site in colunn 4.  It is expressed as the sun of the
inhalation and dermal exposures for each use site.  Assuming a 70 kilogram
applicator, the maximum estimated exposure to EDB is experienced by pineapple
applicators and transfer-operators.  The total estimated annual exposure for
this use site ranges from 303.3 to 532.4 mg/yr.

    Commercial (professional) applicators'  exposures are unknown, but the
Agency (Mittelman, 1980) estimates that the duration of exposure of the
approximately 25 professional applicators nationwide is roughly four times as
long as that of the estimated 13,200 potentially exposed private applicators,
based on data received from Great takes Chemical Co. (1977).

    The CDFA (Maddy, 1979c) and Dow (Rebuttal Ho.  78)  studies  reported airborne
concentrations of EDB in the breathing zones of applicators  while respirators
were in use.  Based on the data from these studies,  the Agency (MitteLman,
19SO) concluded that the use of respirators can effectively  reduce inhalation
exposure by a factor of 14 for half-face and 24 for  full-face  respirators,  but
the use of closed tractor cabs is not necessarily effective, based on current
data that demonstrate higher exposures in closed cabs  than  ir.  open cabs.
                                         -46-

-------
Tobta IS Eutl«k»ttd  mutual dutat Ion of expouure to trtl to Boil fumigation workera (I)
             Annual  iteration of exposure
                                eutloiated nuotxir of exposed
Hke Transfer Application
teaches
Ibbacoo
fleanuts
(HA, AIA,
Nne^lea
Wstja tali lea
Sweet
potatoes
Irish
potatoea
Cotton'TCA)
Cotton (All,
GA, H!i, NCI
Cltnia (pre-
plont)
.06 0.54
1 - 2.5 9 - 22.5
1.2 10.4
24.5 - 43 220.5 - 387
1-5.2 8 - 46.8
0.2 - 0.3 1.8 - 2.7
0.4 - 0.7 3.6 - 6.3
JO 90
2-6 IU - 54
2.5 - 3.2 22.5 - 2U.8
applicatora (private only)
670
10,800 naxiiim
35
45
414
650
150
25
410
1
Citrus (bur-
rowim nuuia-
                 41.6
uunt
             374.4
        duration
of 52 man day a)
                                                         unknown
(T)  MffteTman, T9BO
12)  tor putpuaea  of eatiwating expouuro,  tlie Ajency lias aiiuiuiud a
transfer tturation  of 101 of total e*i
-------
Table 16  Total estimated annual exposure to M)B for prlvato applicators/transfer operators during soil fumigation.  (I/)
Uaa
Bite
Peaches
Tobacco
Peanuts
Pineapples
^ Vegetables
DO
1 Sweet
Potatoes
Irish
Potatoes
Cotton
(CA)
Cotton
(AR, OA,
M3, NC)
Citrus
(preplant)
Citrus
(burrowing
nematode
containment
program)
Total respiratory Intake of KDD (mg/yr)
Transfer Application Total
0.2
3-6-9
4.3
88.2-154.8
3.6-18.7
0.7-1.1
1.4-2.5
36
7.2-21.6
9-11.5
149.8
0.5
8.7-21 .8
10.1
213-9-375.4
8.7-45.4
1.7-2.6
3.5-6.1
87.3
17.5-52.4
21.8-27.9
363.2
0.7
12.3-30.8
14.4
302.1-536.3
12.5-64~.l
2.4-3.7
4.9-Q.6
123.3
24.7-74
30.8-39.4
513
Total estimated dermal Total exposure
exposure (mg/yr)(2/) isB/Vt
0.003
0.05-0.125
0.058
1.2-2.2
0.05-0.26
0.01-0.015
0.02-O.035
0.5
0.1-0.3
0.125-O.16
2.06
0.7
12.4-20.9
" " " ' " ' 14.5
303.3-532,4
2.4-3.7
4.9^.6
123.8
24.8-74.3
30.9-39.6
515.1
(aggregate exposure)
TI
(2
 IT KfA. Hittalmon, 19Gb.  Auaumptlona are discussed In tho text.
 2)  Itaaed on average dermal expoaiire of 4.53 ug/hr (Hoddy, 1979 ).
                                                                                                                                     00

-------
                                       179

                    4)    Summary  of Preplant: Soil
                         Fumigation Data

    The data ir.dicata  that  exposure to ZDB results from all soil f-jnigation
operations.  The highest applicator exposures occur during transfer or loading
operations.  Exposures  also  result from broken hoses or other delivery lines
and practices such  as  lifting  chisels  out  of the ground while turning at the
end of a row or passage  and  knocking off caked soil or organic debris from the
chisels (Mittelman,  1980).

    Data are not sufficient  to estimate annual exposure resulting from dermal
contact with EDB during  soil funigation.   However,  the Agency has collected
data and estimated  that  EDB  levels on  the  hands of tractor drivers range from
0.003 mg/yr for peaches  to  2.15  mg/yr  for  pineapple (Mittslman,  19SO).

    More data are needed about the use of  closed tractor cabs and closed
transfer systems before  their  effectiveness can be evaluated.   The Agency has
been receiving comments on  current efforts to develop a closed transfer
system.  SPA encourages  such efforts.

              b.    Exposure  from Citrus Fumigation

    Ethylene dibromide  is used in  quarantine fumigation programs administered
by the Animal and Plant  Health Inspection  Service (APHIS)  of the U.S.
Department of Agriculture (USDA).   Target  pests include several  species of
fruit  flies which may  infest various fruits and vegetables exported from the
U.S. or shipped from state  to  state.

    Quarantine fumigation of citrus is carried out at commercial facilities in
all major ports of  entry in  the  U.S..   In  Florida,  the Stats Department of
Agriculture conducts quarantine  fumigation at its own facilities.

    The Agency has  obtained  data on inhalation exposure to ZDB associated with
quarantine fumigation  of grapefruit in Florida (Mittelman, 1980).   Two large
fumigation centers  in  Wahneta  and  Fort Pierce, which operate under  a
cooperative program between  the  USDA and the State of Florida, and  a Tampa
warehouse were selected  for  study  by EPA to investigate worker exposure  to
EDB.

    The fumigation  centers  consist of  stations where whole truckloads  of citrus
are fumigated in large  drive-in  chambers.   The Wahneta station has  twelve
chambers and i-.he Ft. Pierce  facility has sixteen.   The individual  fumigation
chambers are designed  to accomodate truck  trailers  loaded  with approximately
1000 boxes of pre-packed citrus.  After the chamber is loaded  and  clc:2d, a pre-
determined amount of liquid  EDB  is dispensed into  an electric  evaporator to
start  the  two-hour  fumigation.   After  the  fumigation has been  completed, the
EDB is exhausted from  the chamber  into the outside  atmosphere, and  the  trailer
is hooked  up to the tractor  rig  for transportation to dockside warehouses;  most
frequently at the Port of Tanpa.

    Several groups  of  people may be exposed to SDB  during  fumigation
operations:  fumigation center employees;  truck drivers, when  they  enter the
chamber to hook up  the tractor-trailer rig, and later when they  open the
trailer doors to unload fruit  at the warehouses; warehouse and dock workers,
who unload fruit from  the trucks,  sort and stack the fumigated boxes in the
warehouse, and move the boxes  to ships for loading.   There is  also  some

                                         -49-

      418-574 0-83-13

-------
                                       180

potential for exposure of people who  live,  work  or  drive in the vicinity of
fumigation operations, from EDB exhaust  in  the atmosphere and the venting of
SD3 from departing trucks.

                   1)   Data or. Applicator  Exposure

    The Agency has assessed inhalation exposure  data  from a nunber of studies
on fumigation station workers at Ft.  Pierce and  Wahneta,  Florida and warehouse
workers at Tampa, Florida (Mittelman,  1980).  Estimates  of the  annual
inhalation exposure  to EDB for these  occupationally exposed groups are
presented in Table 17.

    The Agency (Mittalman, 1980) calculated the  average  annual  Inhalation
exposure to EDB  in milligrams per kilogram  per year,  assuning a 70 kg worker,
in the following manner:

                                  Exposures

                          Wahneta-Indoor Operations

Avg. Concentration     Breathing     Hrs./   Working
   of EDB               rats        day     Days/yr.

            3           3
  0.207 mg/n   x   1. 8 m hr  x  8.0   x   112 - 70 kg =4.8 mg/kg/day

    These calculations assune a concentration of SDB  that is  based  on the
average of all survey samples taken at each site, a moderate  breathing rate
(assigning 50% of the work shift is spent in  slack activity, and 50% in heavy
activity) a 112-day  season at the funigation station, and a 155-day season at
the citrus warehouse.

    The figurss  in the table indicate  that  the workers inside the  warehouse
apparently experience the highest exposure  to EDB.  These  exposure  levels are
estimated to be  up to one hundred times the average yearly  dose  experienced  by
fumigation station workers.

    Major fumigation treatment operations are also  performed  in  Texas  and
Hawaii.  However, these operations are slightly different  from  the  method
employed in facilities in Florida.  In Texas, EDB is  applied  by  dripping  the
compound into a high-speed blower rather than by evaporation.   Of the  36  total
facilities in the O.S., there are 25 commercial stations  and  a  nunber  of
private operations in Texas.   There are no  known air  sampling data  available
from these stations.  In Hawaii, the  funigation of  papayas, bananas,  citrus,
Litchi nuts and cucumbers is performed by evaporating EDB  from a heating  pan.
Treatment operations occur throughout the year.   The  fruit  is hand-carried in
field boxes into and out of the fumigation chambers.  High  potential  for
exposures therefore exist.  The Agency estimates that a total of 12 ArHIS and 5
State of Hawaii employees monitor and supervise  the funigation activities.
Though measurements using detector tubes have failed  to detect SDB  levels in
worker space,  the limits of sensitivity of these methods are relatively high  in
the range of 1-100 ppm, as compared to .008 ppm in Florida measurements
(Mittleman,  1930).

    In addition to the personnel employed at funigation stations and
warehouses,  other persons are potentially exposed to SDB through quarantine
fumigation uses.  At the Wahneta and Ft.  Fierce stations and at the warehouses,

                                         -50-

-------
 I
in
                         Table 17.    Inhalation exposure of workers, to H>B at citrus fumigation stations and at a oltrua warehouse

                    Location      fcrkera            Concentration of                                                         Avg exposure to
                                                     H)B (mg/B3)       Number of      Breathing    Hra/day  \J  Daya/yr  2/  H)B /•»»/i'»/«-t *>
tbhneta
- Indoor operators
Ft. Pierce - Indoor operators
Wahneta
- Outdoor operators
Ft. Pierce - Outd9or operators
Uahneta
- Truckers/Station
Personnel
Ft. Pierce - Truckers/Station
Personnel
Uahneta
Tanpa -
Tampa -
Tanpa •
- Truckers
- Forkllft operators,
Laborers
(inside)
- Forkllft oparators,
Laborers
(outside)
- •Stlcknen*
0.207
0.607
0.406
0.280
0.048 §/
0.092
1.93
8.99
2.36
4.76
14
1
19
2
3
1
2
1
3
2
1 .8m3/hr
1 .6a»3/hr
!.8m3/hr
t.8«3/hr
1 .8a3/hr
1.8a3/hr
t.8m3.hr
1 .8m3/hr
1.8n3/hr
1 .8m3/hr
8 112
8 112
8 112
8 112
8 112
8 112
O.Oa H2r/
a 155^
a 155^
8 155^
4.8
14
9-4
6.5
1.1
2.1
0.4
287
75.3
151 .1
                    1.  lira/day are very variable depending on work load during year. Average 8 hr. work day la assumed hare'.
                    2.  Data from APHIS (1977)
                    3.  Assuming 22 working days a vonth over the period of October 11 through Hay 14
                    4.  Aauualng a 70 kg worker
                    5.  Assuming heavy activity (Spector, 1956).
                    6.  Assuming trucker returns once each day
                    7.  Avg. of 48 monitoring study samples                                 _
                    8.  Including "non-detectable" samples at level of detection (0.003 «g.»}).
                                                                                                                                                       oo

-------
                                      182

there are  private  residences  located a  short, distance  from the station.   This
snail non-occupational population may be exposed.   Truck  drivers  transporting
citrus to  the Florida chambers are also potentially exposed for several  hours
while waiting at the fumigation station, and when  entering the fumigation
chambers for short periods of time (approximately  5 minutes)  to couple and
uncouple their vehicles.  Very little data  are  available  on the levels to which
these populations  are exposed.

                    2)   Summary of Exposure  Data for Citrus

    The data indicate that significant exposures to EDB are encountered  by
workers in all phases of quarantine f-umigation  operations.   The highest
exposures apparently result from warehouse work.   However,  the  limited data
available from Florida fumigation centers indicate  that high  levels of EDB
occur at these facilities.  There are no data available on  fumigation
facilities in other states.   It appears that there  is  high  potential for
exposure in Texas  and Hawaii  (Mittelman, 1980).

    A number of other people  are likely to be exposed  to  EDB  from  fumigation
operations.  These  include truck Drivers and the general  population living or
driving near fumigation facilities.  Although those living  in  homes adracent
tofumigatign stations are exposed to very low levels of EDB (.025  ng/m  to
0.952 mg/m ) there  may be significant numbers of people exposed (Mittleman,
1980).

              c.   Exposure from Grain Mill  Fumigation

    EDB is used as  a "spot" funigant in grain mills.  This  use  is  a-consequence
of regulations of  the Food and Drug Administration  and O.S. Dept. of
Agriculture which specify maximum acceptable levels of insects or  insect  parts
in processed foods.  EDB is particularly effective  in  flour milling machinery
and equipment.   There are many ledges and obstructions where the product
accumulates, and infestation  tends to occur.  Proper "spot" treatment  of  these
critical areas prevents insect growth and reproduction.   Typically, a
formulation containing 70% EDB and 30% methyl bromide  is  used,  although  a
mixture of 59% EDB, 32% carbon tetrachloride and 9%  ethylene dichloride  is also
common.

    Spot fumigations are performed on flour  milling  and cereal  handling
equipment after running the machinery until  it is as empty as possible,  leaving
only small quantities of the product trapped in the equipment.  The funigant is
typically applied by means of applicator gun injection into holes in the
milling equipment designed specifically for  this purpose.    Spot f-umigants are
usually applied over a weekend when the mill is shut down  and personnel are
away.  The fumigation period typically lasts about  24 hours, after which  the
mjj.1 is aerated prior to the return of the mill workers.  Mill fumigations are
repeated on a 3-4 week cycle in order to interrupt the 4-6 week life cycle of
invading target pests.

    In addition to  spot fvaigators who apply EDB, personnel involved in opening
the mill and mill workers themselves are also potentially exposed to
substantial quantities of EDB during start-up procedures.    In some cases, the
same mill personnel are involved in fumigation, aeration and start-up
procedures.
                                         -52-

-------
                                       183

                    1)   Data on Grain  Mill  Fumigator
                        Exposure

    A report commissioned by the Millers' National  Federation and prepared by
Industrial Health Engineering Associates  (IHEA,  1978)  provides the only data
available to the Agency on EDB exposures  resulting  from  grain mill fumigation.
IHEA measured exposure levels of EDB both inside  and outside  respirator face
masks worn by selected applicators during fumigation operations at five
different mills.

    Based on the IHEA data, the Agency (Mittleman,  1980)  has  calculated a  worst-
case exposure for applicators of 50.8  to 59.3 mg/kg/yr.   This exposure  estimate
assumes that the respirator is not worn tightly against  the face or that it is
removed during application, and that there  is no  rotation of  applicators.
Alternatively, with proper use of the  respirator  and no  rotation of
applicators, an inhalation exposure of  4.3-12.5 mg/kg/yr  is estimated.   The
bases for calculating these figures are presented in Table  18.

                   2)   Summary of Spot Fumigation  Data

    The IHEA study indicates that exposure  to EDB can  occur during and  after
spot fumigation.  The highest levels of EDB were  measured during the opening
and start-up operations in grain mills.  The data indicate that  spot funigators
are potentially exposed and encounter  significantly lower levels of EDB  when
respirators are properly worn.

    Estimated exposures to SDB for workers  associated  with spot  funigation  are
presented in Tables 18 and 19.  Workers involved  in spot  fumigation are  usually
also mill employees therefore, application, opening and  start-up are included
in the inhalation exposure estimate in Table 18.  Table  19 presents the
inhalation exposure estimate for those millworkers  not involved  in  funigation.

                   3)   Conclusions

    The Agency has reviewed and evaluated all of  the available data on
inhalation exposures to EDB.   Uses of EDB for which data  were available  include
soil fumigation, quarantine funigation of citrus, and  grain mill  ("spot")
fumigation.  The populations  at risk from these uses of EDB are  listed in Table
20.  The remaining registered uses of EDB are listed in Table 20 and include
vault fumigation of stored clothing,  control of bark beetles in coniferous
trees, -termite treatment,  fumigation of beehive supers, and control of Japanese
Beetles and parasitic nematodes.   Exposure data were not available for these
uses.  Table 21 is a sunmary of the estimated average  dose of EDB experienced
by soil fumigation applicators,  quarantine funigators  and spot fumigators.
These estimates are derived from exposure data which was analyzed by the Agency
(Mittelman, 1980) .


    C.  Quantitative Assessment of Risk Data

          1.   Oncoaen ici try

    The Agency's Interim Cancer  Assessment Guidelines (41 FR 21402) state that
when a chemical is judged  to be  a  potential  human  carcinogen,  the Agency will
estimate its possible impact on the  public health  at current and anticipated
levels of exposure.   The Cancer  Guidelines also  recognize that the available
                                         -53-

-------
                                            IB.    INIIMATIOH EXPOSURE OF SPOT FUMIGATION APPLICATORS TO ETB
 I
ui
Job Category tetlvityj/ Duration Average Number of wg/yi Inhalation nj/kg/yt/ inhalation
(Spot fumigation of activity concentration applications exposure 5/ exposure 6/
applicator) (hra) 2/ of EDfl in rlu» , per year per
breathing zone (og/m ) applicator 4/
Range in parentheses V
I a) Proper use
of respirator
b) Rotation of
applicators
II a) Improper use
of respirator
b) rotation
of
applicator
III a) Improper
use of
respl rator
b) No
Rotation
of
appl Icators
Spot
fumigation
Fogging
Opening alone
(aeration)
Opening f
start up
Normal
operation
Spot
firolqatlon
Fogging
Opening alone
(aeration)
Opening t
start up
Normal
operation
Spot
fumigation
Fogging
Opening alone
(aeration)
Opening -f
start up
Normal
Gyration
1.6S
1.2
0.5
i.e
6.2
1.65
1.2
0.5
1.8
6.2
1.6S
1.2
0.5
1.8
6.2
1.6 (level of detection
0.8 - 2.3 n=3)
4.6 (n=l) 1.6 - 4.7 299.1-878.7 4.3 - 12.5
12.8 (1.9-26.9 n=6)
39.9 (30.6-54.2 n=5)
3.8 (level of detection)
17.1 (n-1)
34.2 (n=l)
12.8 (1.9-26.9 n=6) 1.6 - 4.7 474.1-1392.8 6.6 - 19.9
39.9 (30.6-54.2 n=5)
3.8 (level of detection)
17.1 (rvl)
34.2 (n=O)
12.8 (1.9-26.9 n=6) 12-14 3556-4148.7 50.8-59.3
39.9 (30.6-54.2 n=5)
3.8 (level of detection)
                                                                                                                                                            00

-------
                               185

    Footnotes for Table 18.

I/  Space fogging with organophosphates or bin  fogging with
    methyl bromide may occur following spot fumigation with
    EDB.  Should this be the case, additional exposure to
    EDB will occur due to volatilization of EDB from the
    machinery.  The mill may be opened (aerated) several
    hours to a day prior to start up or may be  opened and
    started up simultaneously (IHE, 1978).  Normal operation
    involves the typical milling cycle on the first day of
    renewed operation.

2/  IHE, 1978. (average of five mills).  Values for each
    fumigation cycle.

3/  IHE, 1978.  Concentrations of EDB are assumed present at
    level of detection.

4/  Estimate based on 6-15 applicators/mills (Soyersmith,
    1977) and 12-14 applications annually (DPRA, 1978).
    If two applicators work together and applicators are
    rotated, then each applicator performs 1.6  to 4.7
    fumigation cycles annually:

     14 applications                12 applications
     3 application teams = 4.7;     7.5 application teams  = 1.6

5/  Assumption of 100% inhalation retention and 1.8 m /hr
    •respiratory rate  (Spector,1956) .  The same  employees may
    or may not be involved in all five activities listed.
    It is assumed here that the spot furaigator  is involved
    in spot fumigation, fogging, opening and start up,  and
    normal operation.  This is a conservative assumption
    made in the absence of any data categorizing actual use
    practices.  Fogging is not always carried out in conjunction
    with spot fumigation, and aeration (opening) is sometimes
    performed several hours to a day prior to start up.

6/  Assuming a 70 kg worker.
                             -55-

-------
                         Table 19.    Inhalation exposure of mill workpra



                                       Ayarage concentration
1
01
en
Duration of EOB in breathing zona
Job Category activity {firs)!/ (wj/p3j2/
Hill worker 3.8 (based on level
of detection in air)
Hunter of
applications
per year='
12-14
I/ During first working day after fumigation.
2/ IDEA, 1978. Estimate based on level of detection in the IHE, 1978 study.
3/ DPRA, 1978.
4/ Assuming 1001 retention of ED6 and a respiratory rate for moderate activity of
•q/yr
inhalation
exposure
656.6-766.
1.8 »3/hr
•g/kg/yr
inhalation
exposure
1 9.4-10.9
(Spector, 1956).
5/  Assuming a 70 kg worker.

-------
                                   187

             Table  20.   Peculations at Risk
     Site
Soil  injection -
Florida citrus  fumigation
Texas citrus  fumigation
Hawaii  fruit  fumigation
Grain storage
                3/
Spot  treatment —'



Baric  treatment

Termite  treatment

Japanese Beetle control

Parasitic nematode  control
               4/
Bee-hive supers-!-'

Vault fumigation
a)  about 25 commercial applicators
b)  up to 13,200 farmer applicators

a)  22 State employees
b)  9 Federal employees
cj  12-30 laborers per warehouse
d)  unknown number of truckers
e)  3 fumigation stations

a)  30 APHIS personnel
by  75 commercial personnel
c)  unknown number of truckers
d}  25 fumigation stations

a).  13 APHIS employees
b)  10 State personnel
c;  unknown number of commercial
    personnel
d)  IS fumigation stations

a)  200 commercial applicators
b)  potentially 50,000 - 60,000
    farmers


-------
                              188

Table 21     Summary table of inhalation  exposure  to  EPS

1.   Soil injection                         Average  dose in mg/kg/yr

    Private Farmer/Applicator          1.   0.01  - 1.8-1-'

2.   Citrus Fumigation (Florida)

    a.  Fumigation station:            2.   a.  2.3-7.5
         Corridor operators

    b.  Fumigation station:                b.  5«1  - 6.5
         Outdoor operators

    c.  Truckers/Station                   c.  0.4-2.1
            personnel
                                                 ?/
    d.  Warehouse:                         d.  287—'
            Indoor laborers

    e.  Warehouse:                         e.  75-3
            Outdoor laborers

    f.  Warehouse:                         f.  151.8
           "Stickmen"
3.


4.
5«
6.
7.
8.
9.
10.
Spot Treatment
a. applicator
b. mill worker
Grain Storage Treatment
Bark Treatment
Bee-Hive Supers
Termite Treatment
Vault Fumigation
Japanese Beetle Control
Parasitic Nematode Control

3-

4.
5-
6.
7.
a.
9.
10.

a.
b.
No
No
No
No
No
No
No

4.3-59
9-4-10
Data
Data
Data
Data
Data
Data
Data

.3
.9







1_/  This excludes pineapple use, for which  the nrivate
applicator/loader dose may be as high  as 7.6 mg/kg/yr.
Specialized cultural practices may  lower this dose  significantly.

2_/  Based on a single personal sample  of 8.99 mg/nr .   Area
samples, however, support the presence of  such high  concentrations
(Mittelman, 1980).
                             -58-

-------
                                       189

techniques for assessing the magnitude of  cancer  risk to  humans based on animal
data are at best very crude.  This is due  to  several  causes,  e.g.,  un-
certainties in the extrapolation of dose-response  data to very low  dose levels
and differences in levels of susceptibility in  animals and hunans.
Accordingly, these risk estimates are neither scientific  certainties nor
absolute upper limits on the risk of cancer from  exposure to  EOB.   The
estimates should therefore be viewed as an index  of potential health hazard
that incorporates both the degree of carcinogenicity  of the compound in animals
and the measured, and/or estimated hunan exposures.

    The CAG has estimated an individual's  lifetime probability of SDB-induced
cancers resulting from:

         .  ingestion of EDB in treated food commodities (CAG,  1980a)
         .  occupational inhalation exposure to  SDB (CAG,  1980c)

Risk estimates for EDB dietary and occupational exposure  were calculated using
a standard one-hit no-threshold model based on  the following  general  equations;
                   —3x
         Risk * 1-e   ; and
            B = (1/y) In [1/(1-P)I ;

     where: 3 » slope coefficient of the one-hit model
            P =« (Pt-PcJ/d-PO
            Pc= incidence of tunors in control  animals
            Pt*3 incidence of tunors in test animals
            y • test animal exposure (ppm)
            x » potential human exposure (ppm)

    As will be discussed below, the Agency adjusted the slope  coefficient  (3)
of this equation to account for differences between exposure  routes of  the
laboratory study used, which was gavage, and  the estimates  of  amounts of EOB
inhaled and consumed via dietary contamination.

              a. Dietary Exposure

    The CAG estimated cancer jisks associated with EDB  residues in  the  diet
using the Agency's dietary exposure analysis  (Holder,  1930), and the results of
a long-term bioassay study of Osborne Mendel  rats and B6C2T1  mice,  completed by
the National Cancer Institute (Hazelton Laboratories,  1975).   EDB was
administered by intubation for two years,  causing a significant increase in the
incidence of gastric tumors.  The Agency used this study  to derive a risk
equation for lifetime exposure, measured in rag/kg/day.  This equation
incorporated both time and dosage in estimating tunor  response rate.

    The Agency recognised that use of the  NCI study could be criticized
because the dose delivered by gavage might not  be comparable to dosing by
ingestion or inhalation.  As a result, the Agency adj  usted  the NCI tunor
incidence results to account for the potential  differences  in tumor  incidence
which may occur because of differences in  exposure routes.  Through a
comparison of the adj usted NCI figures with the results of  a dietary DHCP study
administered to CH albino rats (Hazelton Laboratories,  1977),  the CAG derived
an approximate estimate of the relative carcinogenic effectiveness of the two
methods of administration and adjusted for EDB  potency via  the dietary route.
To obtain an estimate of EDB potency via dietary ingestion, the intubation
value for EDB (3 * 31.73) was multiplied by the ratio of dietary to  intubation


                                         -59-

-------
                                     190

parameters found for DBCP, which gave: 3 » 31.73  x  0.5897 »  18.71.  This
adj ustaient is based on the assumption that the ratio of oral  to  intubation
tumor incidence is the same for EDB as for DBCP because the t*^o  pesticides  are
similar in both their chemical structure and their  carcinogenic  potential  (CAG,
1979).

    Estimates of cancer risks associated with SDB residues in  the diet are
illustrated in Table 22 which is based on "probable SDB dietary  burden."  The
estimated lifetime probability_gf EDB-induced cancer is 2.6 x  10    for grain
fumigation of wheat, 5.60 x 10   for soil fumigation, 2.7 x 10   for
quarantine fumigation of tropical fruit, 5.1 x 10   for spot  fumigation of
wheat, and 1.8 x 10   for quarantine fumigation of  citrus fruit.  The total
lifetime probability of EDB-induced cancer is 3.2 x 10

    Table 23, based on "realistic worst case EDB dietary burden," shows the
lifetime probability of EDB-induced cancer_from general use categories.  The
probabilities range from a low of 6.6 x 10   for  soil fumigation and
quarantine fumigation of tropical fruit to a high of 5.7 x 10    for
quarantine fumigation of citrus fruit.  Other probability values are
1.2 x 10   for spot fumigation of wheat and 5.3 x 10   for grain fumigation
of wheat.  In summary, Table 23 shows that the average lifetime_probability of
EDB-induced cancer from residues in an average diet is 1.5 x  10

    Table 24 lists the risks of EDB-induced cancer  from ingestion of crops
grown_in soil fumigated with EDB.  The range extends from 0 probability co 8.3
x 10   for melons.  The total risk is 4.2 x 10  .  This risk assumes that
crops bear residues up to 0.001 ppm.

              b. Non-Dietary Exposure

    Risk estimates for those groups occupationally exposed to SDB are based on
the Agency's "Non-Dietary Exposure Analysis for EDB Uses" (Mittelman, 1980),
and the results of a recent study designed to assess the carcinogenicity of
inhaled EDB to Sprague-Oawley rats.  The Agency (CAG, 1980c)  derived a direct
estimate of the carcinogenic potency of inhaled EDB using a statistical model,
adjusted for the lack of full lifetime exposures, and applied it to the results
of this study.  The Agecny used the significant tumor types found in the female
rat to estimate the lifetime probability of a canper case in humans.  This
lifetime probability was estimated to be:


                  .   -1.7X
              P » 1-e
where:
        X » estimated human lifetime average exposure

    The CAG applied the above statistical model to the Agency's estimates  of
occupational inhalation exposure and numbers of people exposed.  Table 25
presents the risks resulting from occupational exposures to EDB vapors.   The
exposure level presented in the second column of the table was obtained by the
calculation d » (40X)/(71.3 x 365), where X is the exposure level in terms of
ng/kg/year.  It Is assumed that of a theoretical 71.3 year lifespan, 40 years
are spent in the occupation.  Although the estimates of the lifetime
probability of cancer presented in the table are in the high range,  of
particular concern are the risks to citrus warehouse workers  whose  estimated
lifetime risks of EDB induced cancer can be as high as 4 chances in  10 of (CAG,
19SOc).
                                          -60-

-------
                                     191

    The Agency (Mittelman, 1980) has estimated  that  from 50,000  to 50,000
workers in grain storage facilities are exposed to EDB.   The  risk to  these
workers cannot be estimated because the level of exposure  is  not  known.   Risk
estimates for the lower volune uses of  SDB were not  calculated, because  of  a
lack of exposure estimates.

          2.  Mutagenicity

    As discussed earlier in this chapter, the presumption  of  mutagenic risk to
humans from exposure to EDB has not been rebutted.   The  Agency concludes  the
human exposure to EDB also result In risk of mutagenic damage.
 The Agency is also developing Mutagenicity Risk Assessment
Guidelines that will be utilized by the Agency  for various
chemicals as needed.

         3.  Reproductive Effects

    As discussed earlier in this chapter, the presumption of  r sk of
reproductive effects to humans from exposure to SDB has  not been rebutted.  No
quantitative reproductive risk estimates will be developed, because the
available health studies were determined to be  inadequate to support a
quantitative assessment.

           4.  Summary and Conclusions

    The Agency has applied results from laboratory studies in estimating the
individual lifetime probability of EDB-induced cancers from both dietary and
occupational exposures.  The dietary risk estimates from all registered uses
range from 3.2 x 10   for the "probable dietary burden", to 1.5 x 10   for
the "realistic worst case dietary burden", with quarantine fumigation
contributing 2.1 x 10   and 8.4 x 10   to the respective ranges.   The
occupational risks of SDB-induced cancers resulting from inhalation of SDB
vapor range from 1.8 x 10   to as high as 4 chances in 10 for citrus
warehouse laborers.

    Although the Agency concludes that  there is a risk that mutagenic damage
and reproductive disorders could result from exposures to SDB, the Agency has
not quantified these risks.

    D.  Risks of the Alternative Pesticides to EDB

    Table 26 summarizes all of the known risks for major alternative chemicals
to EDB.  These seventeen pesticides could be substituted for EDB  depending upon
ths use that was cancelled.  The regulatory status for each is also given.
Four of the substitute chemicals are either on the RPAR list or candidates for
RPAR, and most are restricted to use by certified applicators.  In Chapter V,
the risks of each substitute are compared to EDB on a use-by-use  basis.
                                        -61-

-------
                             192

                          TABLE 22

                 CANCER RISK DDE TO  "PROBABLE EDB
                   DIETARY
Source      EDB dietary
            burden
            (lag/kg/day)
                              lifetime  probability
                              of EDB-induced
                              cancer
Soil        2.6 x 10  '
Fumigation

Grain                7
Fumigation  1.8 x 10~A
  (wheat)  (4.1 x 10"5)

Spot                 /.
Fumigation  3.2 x 10
  (wheat)

Quarantine
Fumigation:          -
   Citrus   1.1 x 10~5

   Tropical          ,-
   fruit    1.7 x 10~°
                              5.6 x 10  ^
                              2.6 x  10",
                              (6.28 x 10"4)
                              5.1 x 10"




                              1.3 x 10"


                              2.7 x 10"
Total
            1.6 x 10-|
           (5.7 x 10~5)
 3.2 x 10"
(9.0 x 10"4)
     The parenthesized value  (4.1 x  10""' mg/kg/day)
is calculated from 57 samples of residue.  Non-parenthetical
values were derived from models using application rates.
(Litton Bionetics, Inc., 1978).
                           -62-

-------
                         193

                      TABLE 23

ESTIMATES OP CAHCER RISK DDE TO  "REALISTIC  WDRST CASE
                EDB DrSTART BURDEN™
Source
EDB dietary
burden
(mg/kg/day)
 Lifetime probability
 of EDB-induced
 cancer
Soil
Fumigation

Grain
Fumigation
  (wheat)

Spot
Fumigation
  (wheat)

Quarantine
Fumigation:
   Citrus

   Tropical
   fruit
Total
4.1 x 10"°



3.3 x 10-5



7.6 x 10"6




3.5 x 1
-------
                                                   TABLE 24
       UPPER BOUND CANCER RISKS DUE TO INGESTION OP CROPS GROWN IN SOIL FUMIGATED WITH EDB
Crop
Maximum expected
residue ppb
Estimated dietary burden
at current use level in
  mg/kg/day
                                             2.6 x 10
                                                     ~7
                                                                               Lifetime probability
                                                                               of EDB-induced cancer
Asparagus
Broccoli
Carrots
Cauliflower
Corn, sweet
9
Cotton
Cucumbers
Eggplant
Lettuce
Lima Beans
Melons
Okra
Parsnips
Peanuts
Peppers
Pineapples
Potatoes
Squash
Strawberries
Sweet Potatoes
Tomatoes
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0 q
2.4 x 10j?
2.5 x 10~7_
5.6 x 10"10
0 .
7.5 x 10"
5.0 x 107
1.1 xicrg
6.6 x 10_1U
2.1 x 10J:
5.5 x 10~?
1.2 x 10~b
0 -11
9.0 x 10 n
0 _R
2.0 x 10J?
1 .5 x 10~°
0 q
2.0 x 10 ^
5.4 x 10j(
4-9 x 10"°
0 *
3.8 x 10"S
4.0 x 10 '
9.0 x 10 y
0 q
1.2 x 10"^
8.0 x 10"'
1.7 x 10_|
1.1 x 10 „
3.4 x 10"'
8.8 x 10"'
2.0 x 10"'
0 q
1.4 x 10 y
° -7
3.3 x 10 '
2.4 x 10"'
0 -3
3.2 x 10"?
8.7 x 10"°
7.8 x 10"'
                                                                                                                   <£>
                                                               4.2 x

-------
in
                                                        Table 25
                                 CANCER RISK DUB TO OCCUPATIONAL INHALATION OP EDB
Exposed group
Soil injection
farmer applicator
Fruit fumigation^ '
station corridor
operator
station outdoor
operator
trucker/station
personnel
Warehouse^'
Indoor laborer
Outdoor laborer
Stickman
Spot fumigation
Applicator
Millworker
Range of exposure
(mg/kg/day)
1.5 x
3.5 x
7.8 x
6.2 x

6.6 x
1.5 x
10 5 - 2.8 x
10~3 - 1 .2 x
10~3 - 10.0 x
10"4 - 3.2 x
0.4
0.1
0.2
10~3 - 9.1 x
10 * - 1.7 x
10 3
io-2
io-3
10~3

,o-|
10-^
Lifetime
estimates
1 .8 x 10~5
4.1 x 10~3
9.1 x 10~3
7.2 x 10~4

7.7 x 10~3
1.7 x 10
probability
-3.
- 1.
- 1
- 3.
0.
0.
0.
- 0.
- 2.
2 x
3 x
.2 x
8 x
4
1
2
10
0 x
io-3
io-2
io-2
10~3

10"2
No. of workers
exposed
13,200
n(l)
10
10
4-10<2>
4-10
4-10
2400 -
16,000





- 6000
             hi  Assumed that 31  federal and state employees are evenly distributed.
             (2)  Per warehouse.   Seasonal variation.   Assumed that 12-30 laborers per warehouse are evenly
                 distributed.(Mittelman, 1980)
             (3)  Figures are for  Florida only.   No data are available as to exposures in Texas or Hawaii.
                                                                                                                  en

-------
                                      196
              Table  26  Human Hazards  of  Substitutes  for E2B

           (* - most liVcely  alternative  it  all  EDB  uses  vere cancelled)
Chemical

Aldiearb
*Aluminiua
Phosphide

Carbofuran
*Carboa Disulfide
*Carbon Tetrachloride
Chloropicrin
Dasanit
Potential Hazards to Man
Acutely Toxic
Acutely Toxic
Explosive

Anticdolinestarase
Reproductive Effects
Acutely Toxic
lamuno-tuppressant

Explosive
  (To* Profile
  was not done-
  literature being
  assembled)

Oncogenicity
Mutagenieity
Hephrotoxicity
Hepatotoxicity

Lung Irritant
Skin Irritant
Acute Toxicity

Ancicholinescerase
Regulatory Status

 Some uses restricted
 Others under evaluation

 All uses are
    restricted

 Most uses restricted
 Others under
    evaluation
 Most uses
    restricted
 Notice of 3.PAH. issued
    10/15/80.
 Most uses
    restricted
 No action
D-D
(contains
  1,3  dichloropropene)
 *Ethylene  Dichloride
 Hydrogen  cyanide
Oncogenicity Lang, liver
   & kidney damage
Mutageniciry
Eye & skin
   irritant

Oncogenicity
Mutagenicity

Acutely Toxic
 See Telone
 Accepted
     for pre-a?Aa

 All uses restricted
                                           -66-

-------
Ch**" •*'"'*;

*Methyi Broaide



Mocap


Hemacur
                                      197

                     Hazards of ZDB Substitute

                          Potential Hazards to
Paradichioro benzene
Silica, aercgel
Verier (contains
        1,3 dicfaloro-
        propene
                          CSS lesions
                          fl»rfn irritant
                          Jiitagenicity

                          Anticnoiinesterase
                          Acuta
 Chanicals - (cont'd)

Man     Reguiatory Status
                              4 sJcLn irritant
                          Onccgenicity
                              Irritant
Telone                    Onccgenicitj J!
 (1,3 dichloropropene)    (in vitro)
                                3kL" irritant
references: Federal Register, 1978
            Federal Register, 1979a
            Federal Register, 1979b
            Mishra, L; I978b
        Most uses restricted
        Accepted for Pre-5PAH
        review

        Seme uses.restricted
        Others under evaluation

        Liquid formulations are
          restricted
        Granular formulations
          are proposed for
          restriction

        Accepted for
        pre-SFAR

        Ho Action

        Accented for STAR
        See Telone
                            -67-

-------
                                       198

III. BENEFITS SUMMARY FOR ETHYLESE DIBROMIDE

    A. Introduction

    EDS is used for a variety of agriculturally related  activities-   It  is
produced in the U.S. by the Great Lakes Chemical Corporation and  Dow  Chemical
Company, and its production and marketing data are confidential.  However,  it
has been estimated that in recent years approximately  13 to 15 million pounds
of EDB have been used annually as a pesticide active ingredient.  Sites  treated
include: preplant soil fumigation of agricultural crops  (tobacco, pineapple,
citrus, peaches, peanuts, cotton and various vegetables); fumigation  of  stored
beehive supers; termite control; fumigation of stored  grain; spot ftxnigation of
flour milling machinery; for Federal (APHIS) quarantine programs; and on felled
logs for control of bark beetles.

    Estimates of annual usage were made in  1973 by the USDA/State and EPA
Cooperative Assessment Team.  In 1979,  DHCP, one of the major alternatives  to
EDB, was largely banned from the market.  This most likely resulted in some
increase in EDB usage for certain sites.  Because it. was not possible to
determine the magnitude of these increases, the original 1978 estimates  of
annual usage were used as a base for most of the sites analyzed.

    Benefits analyses for all sites except  stored grain and flour mills  were
prepared jointly by economists of EPA and USDA (Abbreviated Benefits  Analysis
of Pesticidal Uses of Ethylene Dibromide, January, 1979).  Analyses of flour
mills and stored grain were performed under EPA contract by Development
Planning and Research Associates, Inc., of Manhattan,  Kansas.

    The alternatives to EDB were chosen on the basis of cost, efficacy/ and
availability.  Economic impacts of alternatives were estimated using  partial
budgeting.  This methodology allowed measurement of the change in control costs
and yield, if any, while holding all -other  inputs and  their costs constant.

    Cancellation of all current uses of SDB are expected to result in annual
losses to users of approximately $60.0 million (Table  27).  This represents
losses due to both increased cost of treatment and decreased value of
production for certain crops.  The largest single impact would be noticed
through the APHIS citrus quarantine program, where shippers and producers could
suffer a net loss of S24 million, and where balance of payments deficits could
increase by approximately $18 million,  if the export market to Japan were lost.

    In calculating these impacts, it was assumed that only currently registersd
pest control methods would be available at the time EDB cancellation becomes
effective.  Over 30% of the estimated impact of $60 million would occur in  the
AP^HIS quarantine program for which there are no known chemical alternatives to
EDB.  Extensive research conducted in recent years demonstrates that the  use of
gamma radiation can be an effective pest control method (Interdepartmental
Committee on Radiation Preservation of Food, December 1978).   A nunber of
foreign countries and the U.S.A. have approved irradiation of certain  food
products for human consumption (Appendix II).  Though irradiation of wheat,
wheat flour and white potatoes has been approved in the U.S.,  the Food and Drug
Administration, the Federal body responsible for these approvals,  has  .not yet
resolved a 1973 petition (FAP *2045)  from USDA requesting approval of  the use
of gamma irradiation on papayas and citrus.   Should this pest  control  method be
approved and accepted by foreign export markets,  the  impacts of cancelling EDB
may be reduced significantly.

                                         -68-

-------
    EDB is used  in combination  with  ethylene dichlorida and carbon
tetrachloride for insect control  in  stored grain.   A total estimated 630,000
pounds of EDB active  ingredient are  used annually  in grain storage facilities:
409,000 pounds are applied on farms,  while 221,000 pounds are used in
commercially stored grain.  Approximately  225 million bushels of stored grain,
or about 2% of the grain produced in  the country,  are treated annually with
funigants containing  EDB.

    The most likely alternative to SDB-containing  mixtures is the 80%/20%
mixture of carbon tetrachloride and carbon disulfide at on-farm storage
facilities, and  aluminum phosphide at off-farm storage centers.  Because on-
farm and off-farm facilities vary greatly  in air-tighcness the substitutes also
vary; the tighter the  facility, the more likely it is that the substitute would
be aluminum phosphide.  When used as  directed,  both carbon tetrachloride/carbon
disulfide and aluminum phosphide  will be less costly than SD3.  As a result,
fumigation costs will actually  decline if  SDB becomes unavailable and is
replaced with these alternatives.  The Agency estimates a $500,000 to
$-2,260,000 annual savings from  use of these EDB alternatives.

    Another alternative treatment method which is  effective against a broad
spectrum of pests of  stored grain  is  the use of a  controlled  atmosphere of
carbon dioxide,  nitrogen and combustion products of natural gas.   These gases
can be circulated through a sealed bin,  reducing the oxygen in the bin
atmosphere to a  very  low level  and killing any pests.  In all  cases,  bins must
be sealed tightly, grain temperatures must exceed  70  F.  and  the  exposure
time must be at  least  four days.   The commodities  are not contaminated  with any
chemical residues.  The performance of this treatment method  is optimum in
silos and other  readily sealed  sites,  but  use in flat bed storage is  unlikely.
In response to. a petition for exemption from tolerance submitted  by the USDA IR-
4 minor use program,  the Agency has  proposed to grant this exemption  (Federal
Register, Vol. 45, No. 136, July  14,  1980,  p. 47168).  Commerical acceptance
and use of this  controlled atmosphere method could occur  once  the exemption
from tolerance is made permanent;  thus,  another alternative for this  use  of EDB
could become available.  The Agency has no  information regarding  the  costs  of
such a treatment system and would be  interested in obtaining  such information
from any knowledgeable party.

   C. Quarantine Fumigation of  Citrus,  Tropical Fruits and Misc.  Fruits and
      Vegetables.

    USDA's Animal and  Plant Health Inspection Service (APHIS)  requires  the  use
of SDB in its quarantine program  to destroy certain  fruit  flies before  they
become established in  parts of  the U.S.  For the same reason,  Japan also
requires that all fresh grapefruit imported from the U.S.  be treated with SDB.
The target flies include the Mediterranean,  Caribbean,  West Indian, Oriental,
Melon, Queensland, and the Cherry  fruit fly.   In 1977,  approximately 33,500
pounds active ingredient of EDB were  used  to meet  the requirements of APHIS  and
Japan.

    There are currently no registered chemical  alternatives to  SDB  for
quarantine purposes.  Vapor heat  (112 F for 4-6 hours,  followed by chilling)
and cold storage (36  F for 10-22  days)  can control these  flies  as affectively
as EDB, however, its  use often  results in  varying  and considerable degrees of
fr ui t damage.

                                         -69-

-------
                                   Table 21. Annual Economic Impacta of Cancel ling Ethylene Olbromlde
Ifeiu sites
Extent of use
Active ingredient {Units Treated)
Change In Cost
of Treatment
Applied Thousand pounds
Tobacco
Pineapples
Citrus
A. Preplant
B. Burrowing nouatode
containment
Peaclies 2/
Forestry 3/
1 Termites
Q Grain Storage
1 Flour Hilling

Quarantine (AlllIS)





Peanuts
Cotton
Vegetables
Money cowlis
Total
2,10(1.0
572.4

20.0

252.0
170.0
20.0
20.0
630.0
465.0

83.5





4,197.4
721.8
5,600.0
20.0
14,872.1
77,600 acres
11,500 acres

100 acre

500 acres
2,000 acres
68,000 acres
12,400 structures
125.7 to 392. 0 mil. bu.
391 mill

494.5 mil. Ibs.





257,390 acres
55,600 acres
180,000 acres
5,679,000 supers
~
1,200.0
281.5

3.5

44.0
20.3 to 118.2
7.4 to -15.8
mknoun
-2,258.0 to -500.0
4,000.0 to 7,700.0

538.9





2,500.0
435.7
3,800.0
3,686.2
14,895.0 to 19,776.2
Decrease in
Value of
Production
Balance
Market/ of
User Consumer Payment
Impacts Impacts Im[>acla
thousand dollars. I/
0
1,218.5

0

0
0
unquantlfiable
-
0
0

24,1. 4.0





11,000.0
0
0
6,500.0
42,882.5
1,200.0
1,500.0

3.5

44.0
20.3 to 118.2
7.4 to 15.8
-
2,258.8 to -500.0
4,600.0 to 7,700.0
. ,
24,702.9 '





13,500.0
435.7
3,800.0
10,186.2
57,775.5 to 62,658.7

none
minor

none

none
none
none
none
none
$0.016 to $.028
per capita

0
0

0

0
0
0
0
0
0

price 18,384.0
reduction
for grape-
fruit) price
Increase for
papayas fc waivgoes
none
none
none
minor/none
18,384.





0
0
0
0
0
2V  Assumes registration of DOCP.  if DBCP Is permanently cancelled, economic Impacts would range from approximately

~   $379,000 to $610,00 annually.

3/  Including residential use.

    Primarily borne by the growers rather than the actual users.  (ATIUS program of ECU).
                                                                                                                                                      to
                                                                                                                                                      o
                                                                                                                                                      o

-------
                                      201

    An additional method, gamma irradiation,  shows  promise  as  a  substitute  for
EDB post-harvest fumigation.  However, as  discussed above,  the Food  and  Drug
Administration has not reached a decision  on  approval of  the use of  irradiation
on fruits and vegetables.

    Loss of EDB may result in economic impacts  to fruit and vegetable  shippers,
producers, and consuaers.  Shipper impacts would result from an  approximate 700
percent increase in treatment costs,  from  30.87 per 1,000 pounds of  commodity
using EDB to S6.2S per 1,000 pounds of commodity using either  vapor  heat or
cold storage treatments.  Changes in  imports  and exports  of grapefruit,
tropical fruits and miscellaneous vegetables  would  result in anestinated
balance of trade deficit of $13 million  annually, assuming  Japan would not
accept non-SDB treated grapefruit.  However,  if Japan agrees to  accept
grapefruit treated with alternative methods,  this balance of trade deficit
would be eliminated.  These impacts are  summarized  by commodity  as follows:

         1. Grapefruit

    Under current trade agreements, a cancellation  of EDB could  result in a
loss of exports to Japan due to Japanese regulations which  require EDB
treatment.  A loss of exports would result in decreased treatment costs  of
$263,300, while shipments of grapefruit  by other shippers (imports and
interstate) would result in increased treatment costs of  $265,000 (vapor heat
or cold storage), resulting in a net increase in treatment  costs of $1,700.
Under current trade agreements, the domestic market would have to absorb the
grapefruit previously shipped to Japan (about 20 percent of domestic fresh
consumption) which would result in price reductions and an  estimated decrease
in the annual value of production of over $29 million and a balance of payments
loss of over $27 million.  IS Japan agrees to import grapefruit treated by
alternative methods (vapor heat, cold storage, or gamma irradiation), the
economic impacts of cancellation would be significantly reduced.   Specifically,
treatment costs of fruit exported to Japan would increase $1.5 million annually
with the use of vapor heat or cold storage  (changes in treatment costs with
the use of irradiation have not been evaluated).  Also,  there would be no loss
in the U.S. value of production (due to excess  supplies)  and the  $27 million
balance of payments loss would not occur.

         2. Papaya

    Cancellation of EDB could result in a loss of exports to Japan due to
Japanese regulations requiring EDB treatment for papayas.   A loss of exports
would result in reduced treatment costs of $2,700 and a loss in revenue of $1.2
million.  Interstate shipment would result in increased treatment costs of
$73,500 for a net increase in treatment cost of $70,800.   The alternative
treatment methods of heat or cold would result in a  high  level  of fruit
damage.  Although papayas would no longer be exported to  Japan, the  amount  of
fruit damaged would exceed the  previous level of exports.   Therefore,  domestic
supplies would decrease and prices of papaya would increase.
    Cancellation of EDB would result in a loss of imports of mangoes.   This
loss would result in decreased annual treatment costs of $10,900.  The  domestic
producers would not have competition from imports,  resulting in  short  tern
price increases and an increase in the value of domestic (untreated) product
estimated at $2,104,000 per year.
                                         -71-

-------
                                      202

         4. Other Citrus

    Cancellation of EDB would result  in increased treatment costs  of  $473,400
per year due to use of alternative treatments.  The alternative  treatments
would result in product damage with a decrease of $2,003,000 in  product value.

         5. Miscellaneous Fruits and Vegetables

    A cancellation of EDB would result in increased treatment costs of $7,600
due to use of alternative treatments.  The alternative treatments  would result
in decreased value of product of $31,000 per year because of product damage
from use of alternatives.

         6. Summary of Quarantine Fumigation Impacts

    The overall impact of EDB cancellation on the APHIS Quarantine Program
would amount to an estimated increased annual treatment costs of 3538,900 and
an estimated decrease in the value of production of $24,800,800 per year.  In
the short term, consumers would have  price decreases for grapefruit and price
increases for papayas and mangoes.  The magnitude of these increases was not
estimated due to a lack of data.  In  the longer terra, production adjustments
would result in price changes that would partly offset the short term impacts.

    Under current trade agreements, the cancellation of EDB usage  in the APHIS
quarantine programs would have an adverse impact on the U.S. balance of
payments.  This total impact, roughly estimated at $18.4 million, would
eventually be borne by the consvmer.  The major factor would be the $27.4
million loss in the export market of  grapefruit to Japan.  In addition, export
market of papayas to Japan, valued at $1.2 million, would be lost.   Mangoes,
valued at $8.2 million, would no longer be imported.  Other imported
commodities previously treated with EDB would be damaged, resulting in a $2.0
million positive impact on the balance of payment position.

    D. Spot Fumigation of Grain Milling Machinery

    EDB has been used for ever 20 years as a spot fmigant for control of
insect populations which invade flour milling equipment.  Currently, an
estimated 465,000 pounds of EDB active ingredient are used annually for this
use.  A 70 percent EDB and 30 percent methyl bromide combination is common,
although other EDB formulations are also used.  The only alternative spot
fumigant that does not contain EDB is a mixture of 75 percent ethylene
dichloride and 25 percent carbon tetrachloride.  However, the dosage required
for this pesticide is so large that the milling industry does not consider this
mixture to b« a viable substitute.  It must be used more frequently (two  to
four times that of EDB fumigants) to  exert any reasonable degree of control.
The^larga amount of liquid material applied nay cause the system to clog,
requiring extra preparation time before milling can be resumed.  The overall
results of using this alternative mixture include the actual application  of  a
much greater quantity of chemicals, a higher cost for fumigant, the use of more
Labor, and   more down time of the mill.  Carbon tetrachloride is also an RPAR
candidate.  If that compound were cancelled, then no spot fumigant  would
remain.
                                         -72-

-------
                                        203

    The only viable alternative  to  spot  fumigation is general funigation of the
entire mill spacs.  Methyl bromide  and  aluminum phosphide are currently the two
most commonly used general space fumigants,  accounting for approximately 75 and
25 percent, respectively, of  the general space  fumigant market.

    The principal economic impact resulting  from the cancellation of EDB as a
spot fumigant in flour milling equipment would  result from the increased cost
of substituting general fumigation  of the entire mill space (with methyl
bromide or aluminum phosphide) for  the  current  practice of spot  fumigation of
individual items of milling equipment.

    It is proj ected that an average of  four  general fumigations  per year would
be required to achieve satisfactory control.  The current practice is 12-14
spot fumigations and one general fumigation  annually.  Thus,  the  change would
require that mills undertake  three  additional general fumigations per year but
save 12-14 spot fumigations.

    Based on the relative cost of the present versus the  projected systems,  the
most likely increase in annual costs would be a  total of  33,800,000  in  the
wheat, dur-un, and rye milling industries and $800,000 in  the  dry  corn milling
industry, respectively.  These impacts would result  in direct costs  of 514,000
per year in wheat and rye mills and 58,600 in dry corn mills.  Based on total
human consumption of wheat, durum and rye  flour,  and dry  corn mill products,
the consumer impacts would be small, approximately 1.6 cents  per  person.

    If five rather than four  general fumigations  are necessary per year, wheat
 nd rye mill costs would increase $6.4 million  annually,  and  com mill  costs
would increase S1.3 million annually, resulting  in  increased  costs of
approximately $21,500 and $14,000 per mill, respectively.   Based  on  total  human
consumption of wheat, durum,  and rye flour and dry cornmill products, the
consumer impact under the more conservative estimates would continue  to be
small—approximately 2.3 cents per  person  per year.

    2. Preplant Soil Fumigation

         1.   Tobacco

    SDB is used to control root-knot, root lesion, and  stunt  nematodes on  flue-
cured tobacco in Florida, Georgia,   North Carolina, South Carolina  and
Virginia.  Approximately 2.1  million pounds of EDB as  an active ingredient were
used in 1977, two-thirds of this in North  Carolina.

    Alternative registered chemicals to  EDB include  other fumigants  (DD and
Telone II), multi-purpose fumigants (Telone C-17, DD/Pic, Vorlex), and non-
fum\igants (ethoprop, fensulfothion,  carbofuran and oxamyl).  Even  without use
of SDB, nematodes could be eliminated,  through an integrated pest management
program involving rotation, setting healthy transplants, and early root and
stalk destruction.   Growers switching from EDB to other fumigants  or from multi-
purpose formulations containing EDB to other multi-purposs fumigants vrould not
experience any significant change in yield regardless of nematode  population
levels, but the alternatives to EDB are more expensive.

    If EDS were to  become unavailable,  tobacco growers would spend
approximately SI.2 million more for control than they currently spend using
EDB.  These increased control costs represent an increase in total production
                                          -73-

-------
                                      204

costs of approximately 1.2 percent.  Long run  economic  impacts  to  tobacco
growers would be slight.  Likewise, little if  any  impact  -would  be  noted  at  the
market or consumer level if this registration  of EDB were cancelled.

          2.   Pineapple

    All commercial U.S. pineapple production is located in Hawaii,  and all
Hawaiian pineapple acreage must be treated with nematocides.  Damage  is  caused
either by the root-knot nematode or the reniform nematode.  Approximately
572,400 pounds of EDB are used annually for nematode control on Hawaiian
pineapple fields.  The alternative to EDB is OBO?  applied either in combination
with 00 or Telone.  Depending on soil conditions and species of nematode
present, each of these nematocides has its unique  place in  nematode control
programs.  EDB is used on approximately 11,500 acres subject to high  soil
moisture.

    Since pineapple is a perennial crop (with  a normal  rotation of  4  years),
only one-fourth (2,875) of the above mentioned 11,500 acres needs treatment
annually.  Substitution of a combination of DBC? and DO for EDB on  this  acreage
would result in yield losses of approximately  4.8  percent, as well  as increased
control costs of about $98 per acre-  1979 price increases  in DBCP  relative  r.o
other nematocides could further increase these costs by $75 per acre  since DBCP
prices have increased from $9.35 to $20.00 per gallon.  The present net  value
of the combination of these economic  impacts,  (estimated  at 1977 dollars and
prices) over the entire production cycle following removal of EDB (a  5 1/2 year
period), is estimated at $5.2 million (using a 7 per cent discount  rate).  In
the long run, approximately $1.5 million additional cost  per year would  be paid
by Hawaiian pineapple growers until a satisfactory substitute for EDB is
registered for use.  Minor market or  consumer  impacts are expected.

          3.   Citrus

    SDB is registered for preplant use in fruit tree planting sites,  including
citrus groves.  Target pests are the citrus nematode and  the burrowing
nematode.  SDB use in citrus (preplant) has been estimated  at 20,000  pounds
active ingredient annually.  This quantity would treat  approximately  110 acres,
or less than one percent of new citrus acreage planted  annually.  The most
viable alternatives to EDB for preplant citrus use are  DD and Telone.

    A second use of SDB in citrus is  barrier establishment and  maintenance to
control the burrowing nematode in Florida conducted under a cooperative  Federal-
State Program.  In this program, about 250,000 pounds active ingredient
annually are chisel-injected in barrier zones ranging in width  from 16 to 32
feet.

  ^ Economic impacts associated with  cancellation  of EDB  for preplant use on
citrus are expected to be minimal.  In these preplant situations, use of
alternatives will increase direct treatment costs  by an amount  ranging from
$17.50 to $45.50 per acre depending on the alternative  neraatocide selected.
However, because of the low volume use of EDB  for  this  use  site, total cost
increases to affected growers will be only $3,500  per year.  In  the burrowing
nematode control program, use of alternatives  will increase program costs from
$25,000 to $53,000 per year/ depending on the  alternative chemical  chosen.
                                          -74-

-------
                                     205

    No short-tern market or consumer effects  are  anticipated as a result of
this EDS restriction.  The long-run  situation is  Lass certain,  especially if
alternatives prove less effective for  government-sponsored nematode containment
programs.  The extent of long-term impacts  cannot be  precisely  determined,  but
should not be severe.

           4.   Peaches

    EDB is used as a preplant nematocide on peaches.   Primary target nematodes
are ring, root-knot, lesion, and dagger nematodes.  Prior  to  the  suspension of
DBCP, approximately 170,000 pounds of  ECS active  ingredient were  used annually,
enough material to treat about 2,000 acres  of peaches.  Almost  all  of this
usage has been in the East, where over  13,000 acres of peaches  are  planted
annually.  Prior to the 1979 suspension, DBCP was  used to  treat an  additional
8,325 acres annually.  IS the suspension of DBCP  becomes a permanent
cancellation, EDB would be the preferred chemical on  these acres.

    Registered alternatives to EDB for  use  in peach orchards  are  nunerous,  and
include DBC? (tamperarily suspended), DO, Telone, Vorlex,  methyl  bromide,
zinophos, and chloropicrin.  The major  alternatives (DBCP,  DO,  and  Telone)  are
equal in efficacy to SDB.

    If DBCP use is allowed to continue on peaches, increased  grower  costs on
the 2,000 acres currently treated with EDB  *ould range between  $20,300 and
3118,180 annually ($10-559 per acre), depending on the alternative chemical of
choice (CBCP, DD, or Telone).  However, if  DBCP is permanently  cancelled and
EDB is assumed to be the preferred chemical on 10,325  acres,  grower  costs of an
EDB cancellation would range from $373,700  to 3610,100 annually (approximately
537 to 559 per acre) depending on the  alternative chemical of choice  (DD or
Telone).  At the industry level, the increased costs are not  significant under
either scenerio (with and without DBC?).  No  significant market or consumer
impacts are expected if EDB becomes unavailable to peach growers.

         5.   Peanuts

    About 17 percent of peanut acreage  in the major U.S. peanut-producing
regions is currently treated with SDB, resulting in annual EDB  usage of
approximately 4.2 million pounds of the active ingredient.

    There are several registered alternatives to EDB for use on peanuts,
although none are equally effective.   The major liquid funigant alternatives
are Telone II and DD.  Viable granular alternatives are Furadan, Oasanit,
Mocap, Nemacur, and Teraik, all of which are applied at planting.  Each of
thesealtematives is more expensive than EDB.   It is estimated  that
substituting one of the several alternatives (choice of the particular
alternative being dependent on many biological factors) for 2D3  would result in
increased control costs to peanut growers of approximately $2.5  million per
year.

    More important than increased control costs is decreased value of
production.   If granular alternatives are used,  yield  losses of  6  percent are
estimated due to reduced efficacy.  If liquid formulations are used, yield
losses will vary geographically from 7.5 to 8.0 percent.   In the latter case,
yield losses because of a required 10 to 14 day planting delay required to
prevent phytotoxicity to the peanut plant.   These  yield losses would reduce
                                         -75-

-------
                                     206

producers' revenues by S11 million per year in addition to the $2.5 million  per
year increase in control costs.  However, government price support programs  are
expected to prevent these losses from influencing consumer prices.

         6.   Cotton

    SDB is registered for use throughout the cotton belt.  An estimated 721,300
pounds were used in 1977, primarily to control the root-knot neraatode.
However/ since that time,.the suspension of DBCP may have increased usage  to
1.3 million pounds annually.  Telone II and 00 remain as viable alternatives,
and are equally effective under good soil conditions.

    The only economic impact expected from cancellation of EOB would be an
increase in cost of control, from $13.75 per acre with EDB to about $21.00 per
acre with any of the alternatives.  At an application rate of 2.5 gallons per
acre (12.1 pounds a.i per gallon), an estimated 55,560 acres are treated
annually with EDB.  Use of substitutes for SDB on these acres would increase
treatment costs by $435,700 per year,  donpared to the approximate $3 billion
value of the U.S. annual cotton crop, these increases are negligible.
Accordingly, no measureable market or consumer impacts are projected.

         7. Vegetables

    Approximately 180,000 acres of various vegetables (broccoli, carrots,
cauliflower, cucwbers, eggplant, lettuce, lima beans/ melons, okra, potatoes,
squash, strawberries, and tomatoes) are treated by preplant soil fumigation
with EOB to control nematodes.  Depending on the crop in question, EDB is
applied at rates of-1.5 to 6*0 gallons per acre.  An estimated 5.6 oillion
pounds a.i. EDB are used annually.

    Alternatives to EDB (Vorlex, DO, and Telone II) are considered equally
effective.  The loss of EDB on vegetables would result in.increased pest
control costs to vegetable producers of $3.8 million per year.   There would be
no loss in vegetable production due to the cancellation of EOB.  Given a
producer value of these vegetables of about $3 billion, the increased costs  of
the alternative fumigants would not have a significant impact on prices of
these vegetables.

    Producers of melon, lima beans,  tomatoes and cucumbers would incur the
largest impacts.  This is due to the number of acres of these crops treated
with EDB.  Cn a per acre basis, the  increased costs range from 35.38 for
carrots to $28.90 for cucumbers.

     Table 28 shows costs of alternatives to EDB on vegetable crops.

     No - significant market or consumer impacts are expected if EDB were not
available  for use on vegetables.

    ?. Minor Uses

          1.   Fumigation o_f Felled Logs

    Several bark beetle  species in the West attack coni-farous trees and cause
damage which can be reduced through  the use of EDB.  The alternatives are
lir.dane and endosulfan.   However, lindane is not considered a viable
alternative because it is in RPAR review and the lindane ?D 2/3 recoantends its
                                         -76-

-------
                                                                     Table 28
 I
-vj
                                         Alternative Neuatocldea Used to Replace ETUI on Selected Vegetabt
                                            of Materials and Changes In Nomatoda Control Cost, by Crop ->
                                                                                                         ea. Cout
                                                           With alternative ncnatocldes
Increase In neoatodu
Crop


Broccol 1
Carrots
Cauliflower
Cucumbers
Egg Plant
Lettuce
Lima Deans
Melons
(*ra . .
potatoes (white)-"
A. Mdsh.Idalio
U. Oll»r States
Potatoes (sweet)
S(|uat>li
Htrawtxtrrles
Tomatoes
Totals 2,

-' Application rates,
To am based on product
.^ » t .1
With
toe

51
381
10
201
5
5
248
330
59

990
91
214
11
15
312
923

material
labels,

D-D

as
285
6
269
11
9
462
724
129

510
152
291
6
16
524
3,479


Vorlex
$1,000
—
—
—
420
4
—
—
—
67

—
—
—
10
—
—
501

cost par gallon, and the alternative
State recommendations, efficiency of

Telone II

91
202
30
143
1
1
492
406
27

648
122
—
26
35
558
2,788


Total

176
487
36
832
16
16
954
1,130
223

1,158
274
291
42
51
1,082
6,768


Total

125
106
26
631
11
11
706
800
164

168
183
77
31
36
770
3.845

nematocldea used wora specified by the BOB
registered alternatives, and material costs
control costs
Per Acre

22.64
6.88
23.01
28.85
22.00
22.00
23.57
23.11
25.57

16.80
18.48
11.85
25.83
22.50
22.65
—

Assessment
par
                acre.  The a|>pllcatlon rates and material cost per gallon are aa followai

                Current control prou,ra«»
                  Q»i  1.5 gallons par acre for Una boansj 4.5 gallons per acre for catrotaj 6 gallons par acre for swuet
                                potatoes and 1.67 gallons par acre for the runalnlnj vegetables (12 pounds a.1. at $5.50 per gallon).
                 Terr-o-Cidei   B» 6 gallons par acre (8.43 pounds EOU ar»l 7.03 pounds chloroplcrln) at $16.50 per gallon.

                Alternative control prograat
                  r>-D7  16 gallons per acre for sweet potatoes, and 11 gallons/A for remaining crops  (10 pounds a.l.)  at $2.80 pat gallon.
                  Vorlexi  6 gallons pur acre (11 poundu a.l.) at $8.00 pur gallon.
                                                                                                                                                        to

-------
                                      208

cancellation for this use.  Limited information  indicates  that  endosulfan  is
less effective than EDB.  tonchemical controls,  such  as  good  silviculture
practices, careful site selection, avoidance of  monoculture,  and  thinning  and
removal of diseased or injured trees are also  effective.   However,  during  pest
epidemics, chemical control may also be required.

    Records show that in FY 1977, approximately  20,000 pounds of  the  active
ingredient of EDB were used in Colorado and North  Dakota only.  Data  for other
areas are not available.  At label rates, this is  sufficient  material to treat
about 68,000 trees-  SDB is used to treat trees  that  have  already been
attackedby bark beetles after the damaged tree has been cut.  It  is applied  as
an "open" log spray or under a plastic covering.

   The extent to which EDB has been used for forestry purposes  is not
sufficient to appreciably affect commercial timber harvests.  Rather, SDB  is
generally used where bark beetle damage may have aesthetic impacts  to
residential areas, wildlife habitats, or recreational areas.  Such  impacts ara
difficult to quantify, but typical residential lots suffering tree  mortality
from mountain pine beetle infestation could lose front $ 1,500-$3,000 in value.

    Cost: of control could be reduced by as much  as $36,000 per  year if
endosulfan were to replace EDB; however, endosulfan may not be  as effective  as
EDB.

         2.   Termite Control

    SDB is the only pesticide registered by SPA  as a  fumigant for subterranean
termite control.  EDB is also state-registered in California  for  the  control of
drywood termites.  Alternatives registered for control of  subterranean species
include aldrin, dieldrin, chlordane, heptachlor, pentachlorophenol  (under  RPAR
review), and chlorpyrifos.  For drywood species, registered alternatives
include methyl bromide, sulfuryl fluoride, hydrogen cyanide,  pentachlorophenol,
and silica aerogel.

    Estimates of EDB use vary from 5,000 to 12,000 gallon's of formulation, or
approximately 11,500 pounds to 27,600 pounds of  the active  ingredient per
year.  It is estimated that approximately 12,400 homes were treated with ED3
for subterranean termite control in 1977.  No data were available indicating
units treated for drywood termite control.

    Economic impact data regarding the use of EDB and alternative insecticides
are limited; therefore, the magnitude of user, market, and  consumer impacts
cannot be precisely identified.  Because Che current  use of EDB is  snail
relative to other insecticides, overall impacts  should be minor.  However,
certain homeowners, particularly in the southeastern  and the  southwestern  U.S.,
who may experience difficulty in controlling subterranean  termita infestations,
could incur substantial losses.  Because of the  need  to apply a. fumigant into
holes drilled in concrete foundations, there are no known  chemical  alternatives
for homeowners in the Southeast and Southwest.   The only known  solution in this
case is removal of the infested wood and rebuilding.

         2.   Fumigation of Stored Beehive Supers and Honeycombs

    Honeycombs being stored between honey producing seasons an fumigated with
EDB to control the larvae of the greater wax moth (GWM),  which damage the
waxcomb and bases held mounted on frames by hive supers.
                                        -78-

-------
                                      209

    Alternatives to EDB  for  this  use  are paradichlorofaenzene (PD3) and carbon
dioxide.  In the N'orth,  cold weather  is  a natural control.  Paradichlorobenzene
is more practical to use than carbon  dioxide,  as the latter requires elaborate
monitoring equipment to  ens-ure efficacy  and safety for the user.
Paradichlorobenzene is somewhat less  effective than EDB because it does not
provide control of the egg stage  of the  CWM.

    The uss of PDB to  replace EDB in  the treatment of honeycombs will Increase
beekeeping costs by about $3.7 million annually.   The cost per super treated
will increase about 5  cents — 4  cents with SDB versus 9 cents with PDB.
Further, with ?OB and  CO there would be 7 additional treatments needed per
super.

    Using the next best  alternative,  PDB,  it  is estimated that the volume of
marketable honey would be reduced 20  percent  in a super previously treated with
EDB.  This loss is due to decreased control of the GWM larvae which damage the
honeycomb; in turn, the  damaged honeycomb must be replaced with a  fresh comb
base prior to use in the hive during  honey flow.   Approximately one-third of
all supers in the nation would be  treated for  the GWM using PDB in place  of
EDB.  The total quantity of domestic  honey would  thus be reduced by 6.67
percent.

    Annual honey production  in the U.S.  during 1975-77 averaged 191 million
pounds.  Based on a 6.67 percent  reduction in  honey production  without  EDB,  the
annual  loss to beekeepers would be 12.7  million pounds,  valued  at  $6.5  million.

    To  some extent, a  production  loss of this  magnitude  will result in  higher
consumer prices, thus  shifting some of the impact of EDB cancellation from the
producer to the consumer.  The extent of this  consumer price impact cannot be
estimated.

    In  addition to these economic impacts,  there  will probably  be  some  losses
attributable to the reduced pollination  service which bees  provide.   It is
estimated that pollination services of honey bees are valued at $6 billion
annually.  Because the alternatives are  not as effective  as  EDB, there will  be
some losses in pollination capacity resulting  from the reduced  nunber of
colonies.  It is not possible to  estimate  the  impact  on  pollination  which  could
result  from the cancellation of SDB.

          4. Fumigation  of Stored Clothing and Furniture

    Two products containing 5% EDB are registered for use in vault fumigation
to control clothes moths attacking furs,  woolen garments,  rugs,  carpets,  linen,
siohaiz-, or upholstered furniture.  The directions for one  product  state,   "Use
T-H Vault futaig<-it in  a  standard  fumigation system, designed for evaporation  of
a'.^iquid fumigant in a fur storage vault or de-mothing cabinet."   This entails
pouring the liquid fumigant  into  a small flat  container  within  the vault  and,
depending upon the fumigant being used,  placing the container near the ceiling
or on the floor.  The  fumigant then evaporates and the gaseous  fumes  fill  the
vault to penetrate items being treated.   Both  product labels carry statements
of caution regarding handling and  operator protection and  safety.   Also, both
products bear one or the other of  the statements:   "Not  for  household use" or
"For pest control operators only."

    Four alternatives  to SDB were found  for this  use. The  first is  a mixture
of carbon tetrachloride  and ethylene  dichloride in a  ratio of 2  to 7.  since


                                       -79-

-------
                                      210

carbon tetrachloride is undergoing RPAR review,  this  alternative is not
considered fully viable.

    The second registered  alternative  is chloropicrin.   It  is a disagreeable
natarial with which to work, but this  property  tends  to  discourage mishandling
and exposure.

    Methyl bromide is the  third alternative.  The  three  registered methyl
bromide products also contain  0.25 to  2% chloropicrin, as an odor  warning
agent.  For methyl bromide, the term  "vault  fumigation"  would have to  be
interpreted liberally, because the labels cite  barns, storage buildings,  and
other structures infested  with these pests.  Extreme  care must be  exercised
before applying methyl bromide, because it reacts with a number of organic
materials to produce foul-smelling odors.  These materials  include furs,
faather pillows, leather goods, and a variety of woolens.

    The final chemical registered for  this site/pest  combination is sulfuryl
fluoride.  Two products are registered, both containing  99%  active ingredient.
This chemical, like methyl bromide, is released as a  gas through copper  tubing
from a pressure tank into  the  structure being fuaigated.  The economic impacts
of EDB use on vault fumigation were not evaluated but are expected to  be  small.

         5.   Fumigation of Quarantined Plants, Soil and Grass Sod

    The Japanese Beetle Domestic Quarantine program (7 CFS 301.48)  requires the
treatment of regulated articles (including such material as  soil,  balled
plants and grass sod).  Ethylene dibromide ¥440  (20% SDB), miscible ethylene
dibromide $434 (2.5% EDB), and methyl bromide are authorized  fisnigants for
these regulated articles to control the Japanese beetle.  Because chlordane has
been cancelled for all uses other than structural termite control  and dipping
the roots and tops of non-food use plants,  it is limited only  to some uses
cited below.  Therefore, methyl bromide is the  single alternative  used for
comparison.

    Labels for these EDB-containing products do not specify that use be
restricted to certified applicators only, but do caution against inhaling,
swallowing, skin contact, vapors near eyes, or contamination  with  feed and food
stuffs.

    Applications of ZDB and methyl bromide for each of these  articles are
detailed below.

     Grass sod;  EDB #440  (20%) or *434 (miscible, 2.5%) is mixed at 20 cc £D3
to 1 gallon of water, sprayed over the sod and left to stand undisturbed  for 24
hours.  This amounts to 20 Ibs. of SDB per acre treated.

    Methyl bromide is applied to cut,  stacked,  or rolled sod, under a tarp and
left to stand for 3 hours.  The average rate of treatment is 2.5 Ibs a.i. per
1000 cubic feet or 60 pounds methyl bromide per acre under  a tarpaulin.

     3are root plants:   ZDB #440 (20%) is mixed at 12 cc per gallon of water
and the plant roots ars soaked in the solution 24 hours.   Methyl bromide
application is 2.5 Ibs/a.i. per 1000 cubic feet under a tarpaulin.

     Plants Balled or _in Containers:    £440 SDB (20%)  is mixed at 2 cc  per
gallon of water to be used as a dip or 2.5  cc per gallon of  water to be poured

                                       -80-

-------
                                      211

on.  Plants are then left undisturbed for 24 hours.  There  is  also  an  injection
method which is not ordinarily used at the present time.  At this rate,  1.65
Lbs. of EDB are used per 1000 cubic ft. of soil  in balled and  containerized
plants.  Methyl bromide is applied at a rate of  2.5  Ibs. per 1000 cubic  ft.
under a tarpaulin.

     Pottina and bench soil:   50 cc of $440 is mixed  in 3  gallons  of  water per
cubic yard of soil (55 cc in 6 gallons of water for  high organic soil).  The
soil is treated in 12 inch layers and soaked for 48  hours.  Present annual
usage anounts to less than 50 gallons of the formulation.   (USDA, 1978).
Methyl bromide is applied at a rate of 2.5 Ibs. per  1000 cubic feet for  3 hours
under a tarpaulin.

    The economic impacts of the alternatives have not been  evaluated.
Fumigation chambers or technology are not generally  available to nurserymen who
are involved in this treatment on only an occasional basis.
                                          -81-

     418-574 0-83-15

-------
                                      212

IV. Identification of Regulatory Options

    A.   Introduction

    As explained in Part I, FIFRA mandate's  the  Agency  to  weigh  the  health  risks
and economic benefits of each use of a pesticide  in order to  achieve  a  balance
between these competing considerations.  In Chapters II and III,  the  Agency
identified the risks posed by exposure to SDB and the  benefits  derived  from
each of its registered uses.  To achieve a  risk/benefit balance,  the  Agency  has
identified a number of regulatory options,  and  has evaluated  each option for
its impacts on both sides of the risk/benefit equation.

    In addition, the Agency has evaluated the risks of alternative
pesticideswhich have been identified as viable  options for the  various  uses  of
EDB.

    This section identifies the regulatory options available  to the Agency to
reduce the risks from the registered uses of EDB.  The Legal  authorities for
these options as provided in FIFRA are also discussed.  In Chapter V of this
docunent, each option wj,ll be evaluated for its impact on the risks and
benefits of each registered use of EDB.  These  changes will then be compared to
determine the most appropriate regulatory option for the  specific registered
use being examined.

    B.   Legal Basis for Options

    FIFRA provides the Administrator with three broad  legal courses of action
for the registration of a pesticide.  These are cancellation,  continued
registration without restrictions, and continued registration with restrictions.

    Cancellation and continued registration without restrictions are opposite
courses of action which hinge on a determination that  a pesticide does or does
not cause unreasonable adverse effects.  To make this  determination, the
Administrator weighs the risks and benefits of each use of a pesticide.   If  the
risks of a use outweigh the benefits, and the risks for that use cannot be
lowered, the use may be cancelled.  If the benefits of a use outweigh its
risks, and any actions taken to reduce the risks result in a significant
reduction in benefits, the registration of that use may continue.   In the case
of cancellation, the Administrator issues a notice of  intent to cancel any  or
all registrations off a pesticide (FIFRA Section 6(b)(1)].   This cancellation
notice becomes effective within 30 days, unless a registrant requests a.  hearing
regarding its registrations.  In the case of continued registration without
restrictions, the Administrator may publish a final SPAR position  docunent
stating a rationale for choosing that course of action.

    Between these polar alternatives of cancellation and continued registration
without restrictions exists a whole range of possible regulatory actions.   In
fact, FIFRA requires ths Administrator to "consider restricting a  pesticide's
use or uses as an alternative to cancellation...." [FIFRA  Section  5(b)].
Although FIFRA does not specify what kinds of restrictions are to  be examined,
the Agency interprets this to mean amending the terms and  conditions of
registration in any fashion necessary to lower exposure (and consequently risk)
to the point that benefits are greater than the risks,  and the reduction in
risks has not resulted Ln a significant reduction in benefits.
                                        -82-

-------
                                      213

    C.   Data Gathering  for Identification  of Options

    In the process of varying the terms  and conditions of registration,  the
Agency can create an array of regulatory options intermediate to those of
cancellation and continued registration  without  restriction.   In order to
facilitate the development of viable regulatory  options,  FIFRA provides  the
Agency with certain legal "tools" to be  used  in  the  process of gathering
information on available options.  The legal  "tools"  which FIFRA provides for
developing intermediate options are:

         1.   Additional data

    Under FIFRA section 3(c)(2)(3), the  Administrator may require a  registrant
to submit any additional data necessary  to  maintain  an  existing registration.
When reviewing an RPAR pesticide, the Agency  may identify areas in its risk and
benefit analyses where data are inadequate  or totally lacking.   The
Administrator may then require a registrant to provide  specific data  within a
certain time; if the registrant fails to  comply,  his- registrations may be
suspended.

         2.   Information-Gathering Hearings

    If the Administrator is uncertain whether to  cancel or restrict a
pesticide, s/he may publish a notice of  intent to hold  a  formal hearing  to
resolve the issue [FIFRA Section 6(b)(2)].  At the end of  this  hearing,  the
Administrator makes a decision based on  the hearing  record.   As  an alternative
to a 'formal hearing, the Agency may conduct an informal' hearing to gather
information, [FIFRA Section 21(b)l  and then  use  thac  information  co propose  a
recommended decision.

    D.   Leaal Options Available Under FIFRA

    The following regxilatory options are  available to the  Administrator  under
FIFRA to reduce the risks from the registered uses of a pesticide:

         1.   Classification for'Restricted Use

    FIFRA Section 3(d) enables the Administrator  to restrict a  pesticide's
use(s) if sxich use(s) cause unreasonable  adverse  effects.  This classification
means that the pesticide may only be applied  by or under the supervision of a
certified applicator, one who is trained  and  approved by a state as competent
to use restricted pesticides.

    In classifying a pesticide for restricted  use, the Administrator may also
propose other restrictions specific to each use of the pesticide.  The
additional restrictions are aimed at reducing  the risk to  the general
population and applicators.  These other  restrictions are discussed in Section
E of this chapter.

         2. Amend the Terms and Conditions  of Registration

    Under FIFRA Section 6(b)(1), the Administrator may offer  the registrant the
option of voluntarily amending the specific registered uses of a pesticide for
which unreasonable adverse effects exist.   Rather than to cancel all  uses of
the pesticide, the voluntary change in classification would permit tha
                                        -33-

-------
                                      214
registration to be continued for certain uses which do not cause unreasonable
adverse effects and it would be cancelled for those uses where unreasonable
risks are present.

         3.   Labelling

    FIFRA Section 3(c)(5) specifies that the label and labelling of a pesticide
must comply with the requirements of FIFRA.  That is, the label must not have
use directions which result in unreasonable adverse effects.  Consequently, the
Agency nay require a registrant to amend the label in order to keep the label
in compliance with FIFRA.  Examples of label changes include requiring
protective clothing and/or respirators, warning applicators of the
toxicological hazards of the pesticide, altering the application fates or
methods, and other use or handling directions which are designed to reduce
applicator exposures.

         4.   Establish or Re-evaluate Tolerances in Con-junction
              with FDA

    Section 408 of the Food, Drug and Cosmetic Act requires that the
Administrator, under recomnendation from recognized experts qualified to
evaluate pesticides, establish a tolerance for residues of a pesticide on raw
agricultural commodities.  Also, the Administrator may require that the
registrants submit data in the form of a petition for tolerance according to
the recommendations of an advisory committee.

    The data required under Section 408 to establish a tolerance are:

    (A)  the name, chemical identity, and composition of the pesticide
         chemical;

    (B)  the amount, frequency, and time of application of
         the pesticide chemical;

    (C)  full reports of investigations made with respect to the
         safety of the pesticide chemical;

    (D)  the results of tests on the amount of residue remaining,
         including a description of the analytical methods used;

    (S)  practicable methods for removing residue which exceeds
         any proposed tolerance;

    (F)  proposed tolerances for the pesticide chemical if tolerances
         are proposed; and

    (G)  reasonable grounds in support of the petition.

    Should the Administrator cancel certain uses of a  pesticide,  s/he may
revoke the established tolerance for that pesticide on tha basis of the  data
supporting cancellation [40 CFR 180.32].

         5.   Cancel After a Specified Period of Time

    If the Administrator determines that  certain uses  of a pesticide chenical
cause unreasonable adverse effects, then  s/he would issue a notice  of intent  to

                                        -84-

-------
                                      215

cancel those uses.  IS s/he also determines  that  there  are  no  effective
pesticide or nonchemical pest control methods currently available  as effective
alternatives to EDB, the Administrator could delay cancellation  for  a specified
period of time to allow for development of alternatives.  Such a decision would
be implemented by issuing a notice of intent to cancel  [FIFRA  Section 6(b)(91)]
and by conducting a formal hearing if requested.

    In summary, FIFRA provides the Administrator  with several  legal  authorities
for regulating RPAR pesticides.  With this legal  foundation established,  the
Agency next must explore possible technical  use restrictions (i.e.,  amendments
to the terms and conditions of registration), and evaluate  the extent to  which
these restrictions would reduce the risks of SDB.

    E.   Risk Reduction Measures for EDB

         1.   CJse Restrictions for All CTses  of EPS

    This section presents the use restrictions which could  apply to  all uses of
EDB.  The restrictions listed here are intended to reduce exposure to
applicators, other persons, and animals during and after fumigation.   This
section is intended only to present all of the potential options developed by
the Agency, and not to indicate a decision to impose specific  options.
Sections following this one present additional restrictions specific  to each
use pattern.  These additional restrictions  are intended to reduce both
applicator exposure and food residue exposures.

              a.   Classify SDB for Restricted Use

    Due to the chronic hazards posed by SDB, the  Agency could  classify  this
pesticide for restricted use.  Under this classification, only certified
applicators or persons directly under their  supervision would  use EDB.  A
standard statement to this effect would appear on the label.

              b.   Protective Clothing

    To prevent dermal contact with EDB, applicators could be required to wear
boot covers, disposable gloves and an apron  during transfer of the pesticide
(i.e., pouring), during calibration or maintenance of application equipment, or
during clean-up of spills.  These protective articles must  be  made of a
material resistant to EDB penetration.  Butyl rubber is recommended  for use in
gloves.  Nitrile and polyethylene, while less resistent, are sufficient for use
in aprons, coveralls and boot covers where close  body contact  does not occur.
Articles made of natural rubber or leather should not be used.    Contaminated
clothing should be removed immediately and exposed skin washed with  soap and
water.  Gloves and boot covers would be thrown away immediately  if
contaminated, or after each use (not to exceed one hour) in any  event.  Other
clothing vjould be commercially laundered daily, or aired for at  least 24
hours.  All of these requirements would appear on the label.

              c.   Respirator

    A respirator could be required to be worn at  all times  during application
and when reentering the treated site throughout the specified  reentry period.
A full-face  black canister gas mask approved by IJIOSK/MSHA  for removing organic
vapors  could be used.  Applicators would be  trained to  properly  wear, maintain.
                                         -85-

-------
                                      216

and store a respirator.  The black canistor or charcoal  filter  is  to  be
replaces after each day or use.  All of these requirements would appear on  the
label.

              d.   Reentry Period

    Before application, signs would be required to be posted at entrances to
the treated area (i.e., treated field or building) bearing the signal word
"DANGER," the product name, date of treatment, treatment period (if
applicable), the date when reentry is allowed, and what  protective equipment  is
required.  The sign would remain posted during application and throughout the
period before reentry is allowed.  The label would also  contain this
information.  Sample warning statement might read:

                            FOR OUTDOOR FUMIGATION;

                                    DANGER

                     SOIL FUMIGATED WITH (PRODUCT NAME)  ON
                     (DATE).  DO NOT REENTER TREATED AREA UNTIL
                     (REENTRY DATE) OR^UNTIL EDB AIR LEVEL IS
                     LESS THAI! 0.4 PPM .  WEAR A FULL FACE
                     BLACK CANISTER RESPIRATOR IF YOU MUST REENTER
                     THIS AREA BEFORE THE POSTED REENTRY INTERVAL.

                            FOR INDOOR (SPACE) FUMIGATION;

                                    DANGER

                    PREMISES FUMIGATED WITH (PRODUCT NAME)
                    ON (DATE).  TREAMENT PERIOD LASTS UTIL
                    (DATE).  WEAR A FULL FACE, BLACK CANISTER
                    RESPIRATOR BEFORE REENTSRING PREMISES.  AFTER
                    TREATMENT PERIOD,  OPEN ALL WINDOWS,  DOORS AND
                    VENTS.  REENTRY WITHOUT A RESPIRATOR IS ONLY
                    PERMITTED AFTER 24 HOURS OF AERATION OR WHEN
                    EDB AIR LEVEL IS BELOW _0._4. PPM.*
* 0.4 ppm is the lowest level at which an IR detector alarm can sense the
presence of EDB.  These detectors are available from the Foxboro Company,
Wilkes Infrared Center, S. Norwalk, CT.

              e.   Warning of Chronic Risks

    The label would indicate that EDB poses a risk of cancer and mutagenic
effects to humans in addition to the acute effects already described on the
label.  A sample sentence is:

                      "EDB HAS CAUSED CANCER, MUTAGENIC
                      EFFECTS AND REPRDUCTIVED DISORDERS
                      IN LABORATORY ANIMAL TESTS AND MAY
                      PRESENT TIIES2 HEALTH RISKS TO HUMANS".
                                        -86-

-------
                                      217

               -'   Other  Safety  Precautions

    The label  would  also  include statements expressing the following
requirements:

     - Fumigation stay  cake  place only when personnel
       will not be in  the area (i.e., treated field
       or building)  during  the treatment  period and the
       reentry period.  Humans and animals must be  vacated
       from treatment  area  before fumigation;

     - EDB may not be  used  or stored in domestic dwellings;

     - Soap and water  must  be available for washing skin
       or flushing eyes in  case of contact  with  EDB;

     - Hands must be washed before eating  or smoking;

     - Transferring activities (i.e., pouring)  must be
       conducted outdoors,  unless a closed  system of
       transfer is used;

     - Spills  must be  cleaned up  immediately.   If a spill
       is on soil, cover with six inches of soil.   If a
       spill occurs on other surfaces, remove EDB with an
       EPA-approved cleaning agent;

     - Pesticide containers are disposed of by  either
       burying in a State-approved pesticide disposal
       site or by returning them  to a container
       reprocessor;

     - Containers or equipment made of aluminum,
       magnesium, or their alloys must not  be used,
       because EDB will corrode  them.

         2.    Additional Requirements for Soil Fumigation Uses

               Reentry Period.  Reentry without  full-face,  black canister
respirator into treated area will not be permitted until 24 hours after
fumigation is  completed.  This information would appear on the label and on
warning signs posted around treated areas.  This requirement '*ould supersede
the minimum use restriction mentioned previously in that it proposes a specific
reentry time interval.   The reentry interval is based on data indicating that
EDB level are below the level of detection at 24 hrs. after fumigation.

         2.   Additional Requirements for Stored Grain Fumigation

              a.   Restrictions for Reducing Applicator and
                   Reentry Exposure

                   1)   Protective Clothing

    EDB-resistant gloves (butyl rubber),  boot covers and apron (butyl  rubber,
nitrile,  or polyethylene)  would be required at all  times during fumigation.
                                         -37-

-------
                                      218

This requirement would supersede the minimum  use  restriction  of wearing
protective clothing only during mixing and loading, etc.   The label  would
describe this additional requirement.

                   2)   Use Directions

    Before fumigation, all leaks and vents in the grain storage structure
(e.g., bins, flats, elevators, boxcars or other storage facilities)  would be
required to be sealed with masking tape, caulking or plastic.   Humans  would be
vacated from the structure to be treated.  Only devices which apply  the
fumigant quickly, evenly and with minimal vaporization would  be permitted (such
as a sprinkling can without a nozzle or a coarse  stream power sprayer).   After
fumigation is completed, the treated structure would be sealed and all
entrances locked.  The label would describe these requirements.

                   3)   Reentry Period

    After the treatment period has elapsed, a respirator could be required to
be worn upon reentry, all doors and windows opened, and aeration systems
activated (if available).  Reentry without a  respirator would be permitted only
after 24 hours of aeration or until an IR detector indicated  a concentration of
EDB in air no more than 0.4 ppm.  The label and posted warning signs would bear
this information.

                   4)   Placarding- Treated Grain

    Treated grain transported to another location would be placarded with  a
sign indicating the date of last treatment.   This sign could  be removed at the
grain's destination.  The purpose of this use restriction  is  to warn grain
inspectors and other workers of the previous  use of a funigant in a  particular
grain shipment.  The label would include this restriction.  As a further  safety
measure, grain inspectors would use an IR detactor to determine whether
transported grain contained levels of EDB gas in excess of 0.4 ppm.  If levels
were to exceed 0.4 ppm, grain would be held in quarantine  at  the elevator  until
EDB gas measurements were below the 0.4 ppm level.

              b.   Restrictions for Reducing  Food Residues

                   1 ) Reo-uire a Holding Period for Treated Grain

    EDB slowly dissipates from treated grain held in storage.  To allow EDB
residues to dissipate, the Agency will consider requiring holding periods  of
30, 50, or 90 days for treated grain.  If a holding period were imposed, it
would be stated on the label and on the placard that accompanied the
transported grain.  Tolerances would be established, or changes in Labelling
would be made to enforce this requirement.

                   2)   Prohibit "Off-Farm" Fumigations (Country Elevators,
                        Terminal Elevators, 3oxcars, Trucks,  ,Mill Storage.
                        etc.).

    Because "off-farm" use of EDB contributes about 3/4 of the  dietary  risk
from grain, the Agency could prohibit these uses.  The only uses on  grain  which
would be permitted under this restriction would be "on-farm"  fumigation of
                                          -88-

-------
                                       219

grain stored in warehouses, bins, or under  tarpaulins.   If this restriction
were imposed, it -would be  required  to  appear  on the label.  Tolerances would be
revised or changes in labelling could  be  imposed to enforce this measure.

                   3)   Prohibit  "Off-?am" Use and Prohibit"On-Fanr." Treated
                        Grain  from  Being Used for Human Consumption

    To eliminate direct dietary exposures of  EDB through treated grain, the
Agency will consider prohibiting  the use of treated grain for direct human
consumption.  To ensure that this restriction is followed, registrants would be
required to add an FDA-approved dye to their  products;  grain treated with EDB
would then be discolored and identifiable as  illegal for hunan consumption.
This requirement would have to appear  on  the  Label.  Tolerances would be
revised or changes in labelling could  be  imposed to enforce this measure.

         4.  Additional Requirements for Soot Treatment of
            Milling and Cereal-Handling Machinery

              a.   Use Restrictions for Reducing Applicator
                   and Reentry Exposure

                   1 )   Protec-tive  Clothing

    SDB-resistant gloves (butyl rubber), boot covers and apron (butyl rubber,
nitrile, or polyethylene) would be  worn at  all times.   The label would state
this.

                   2)   Use Directions

    Before fumigation, milling equipment would be run dry to  minimize the
amounts of flour and grain parts  left  in the  machinery.   Persons not  involved
with the fumigation would  be vacated from the building.   Only a  "closed system"
hand-held applicator gun or an automatic dispensing system would be used  to
apply EDB.  Open pouring or pouring into bottles would  be prohibited.   After
completing fumigation, all windows, doors,  vents  and entrances would  be  locked
shut.  These requirements would appear on the  label.

                   3)   Reentry Period

    After the 24 hour treatment period, a respirator would be required  to be
worn to reenter the building.  All  doors, windows  and vents would be opened  at
this time.  Before anyone could reentar the treated building  without  a
respirator, the premises would be aerated 24  hours  or until an IR detector
shewed that £03 air levels are below 0.4 ppm.  The  label  and  posted warning
signs would bear this information.

              b.   Restriction for  Reducing Food Residues

     Prohibit a Quantity of Contaminated Flour from 3eina tJsed for Human
Consumption.  After spot treatment  a mill could  require several hours to  start
producing flour of acceptable quality.  This  first batch of flour is normally
separated and then slowly  fed back  into the flour  stream.  Because a
significant portion of the EDB residues resulting  from spot treatment occur  in
this start-up batch, the Agency would  require  that  the batch of  flour be
discarded.  This requirement would appear on  the label.


                                        -89-

-------
                                      220

         5.   Additional Requirements for Pose-Harvest
              Fumigation of Citrus and Tropical Fruits

              a.   Engineering and Operational Changes  for
                   Fumigation Facilities.

    The Animal and Plant Health Inspection Service  (APHIS) of the  U.S.
Department of Agriculture (USDA) regulates all fumigation centers  in  the  U.S.
and two centers in Mexico.  APHIS regulations and its treatment manual would be
revised so that each fumigation center meets a standard of a maximum
permissible level of EDD concentration in air of 0.4 ppm  (15 min.  avg.).  To
meet this standard, APHIS could implement any of the following risk reduction
measures to achieve the air standard in addition to those which appear on the
label (See Section IV. S. above):

    -Closed dispensing system (EDB-resistant tubing with  soldered
connections), including an air coupling (with which to purge the line
before opening it), and a pressure relief valve.  Valves would be  designed  for
handling SD8.

    -Air monitoring device with an I'R detector connected  to an alarm and
calibrated to be triggered if the EDB1air levels in the hallway exceed 0.4 ppm.

    -Ventilation system would be adj usted so that there is greater
pressure in the hallway than in the fumigation chamber.

    -Training of applicators in all aspects of fumigation procedures,
protective equipment, safety precautions and emergency plans.

    -Testing of fumigation chambers for leaks before ths USDA inspector
approves a fumigation permit.

    -Restarting exhaust fan after opening chamber door to move EDB vapors away
from workers.

    -Monitoring of personnel to detect EDB exposure levels on a scheduled
basis.

    -Posted operating instructions and emergency procedures.

              b.   Requirements for Warehouses Holding
                   Treated Commodities.

    APHIS is responsible for reducing exposures that result from treated fruit
as it is transported or stored.  Because high exposures have been detected
during warehouse workers' handling of treated fruit, APHIS could require these
warehouse futilities to meet the standard of 0.4 ppm EDB in air.  This would
involve:

    -Monitoring of personnel to detect EDB exposure levels on a scheduled
basis.

    -Installation of ventilation systems in warehouses adequate to ensure that
exposure does not exceed 0.4 ppm.
                                          -90-

-------
                                       221

     -Installation of  detection/alarra devices in warehouses calibrated for 0.4
open.

     -Requiring  full-face,  black canister respirators for all personnel who are
exposed to EDB  air levels  higher  than 0.4 ppn.  (For instance, the "stickmen"
who  unload treated citrus  from  trucks would have to wear a respirator unless
the  cardboard containers are  eliminated.)

     -Replacement of cardboard citrus boxes with a bulk loading system made of a
non-SDB-absorbent material.   (This  requirement  would be optional.  However,  a
bulk loading system may reduce  exposure  to less than 0.4 ppa and thereby
eliminate the need for workers  to wear respirators.)

     -Training of all  warehouse  personnel about  hazards of EDB, methods for
reducing exposure, and emergency  procedures.

     -Signs would be posted in warehouses stating:

   WARNING - FUMIGATED FRUIT  MAY  SHIT HARMFTE, STHYLSNE OIBRCMIDS
   VAPORS.  IF  AIR LEVELS  BECOME  HAZARDOUS AND  ALARM SOUNDS,  PUT CM
   FULL-FACE BLACK CANISTER RESPIRATOR IMMEDIATELY.

              c.   Restriction  for  Reducing Food Residues

     Require treated  fruit to be  stored  for a specific period.  One approach to
reducing SDB residues in treated  commodities is to  lengthen  the  treatment-to-
market interval by requiring  treated fruit to be stored for  a certain period
after treatment.  Based on the  available data the Agency has  estimated a
degradation formula for EDB residues in  treated fruit.   Using this  formula  for
citrus, SDB residues  are estimated  to  decrease  from 8 ppm to  1.42 ppm during a
six-day treatment-to-market interval.  If  this  treatntent-to-market  interval
were lengthened, EDB  residues could  be further  reduced.   TO  explore  the
feasibility of  this approach  the  Agency  could require APHIS  to hold  treated
fruit from commerce for a  certain time period.   This requirement  would be
stated on the label.

         6.   Additional Requirements  for  Fumigation of Beehive  Supers

              a.   Use Restriction  for Reducing Applicator and
                   Reentry Exposure

                   1)   Protective Clothing

     The label vould state  that  EDB-resistant gloves  (butyl rubber) , boot covers
and  apron (butyl rubber, nitrile or  polyethylene) would  be worn  at all times.

                   2)   Use Directions

    Hives and/or combs (supers)  would be placed  in a gas-tight room or under a
gas-tight covering such as a polyethylene tarpaulin  held down with sand-filled
"snakes".  All  windows, doors and vents would be sealed.  All persons not
involved with the funigation would be vacated from the building.  The label
would state these directions.
                                         -91-

-------
                                      222

                   3)   Reentry Period

    Following the treatment period, a full-face black canister respirator would
be worn when reentering.  All doors, windows and vents must be opened at this
time.  Aeration for 24 hours or until EDB air levels are less than 0.4 pom,
measured by IR detector, shall be required before reentry is permitted.  The
label and warning signs would contain this information.

              b.   Restriction to Avoid Food Residues

     Limit_fumigation to beehive supers that are clean and in storage.  To
avoid possible contamination of honey or honeycombs, the label would prohibit
treatments other than on clean supers in storage.  This restriction is in
accordance with current practice.

         7.    Additional Requirements for Vault Fumigation of
              Stored Clothing.

              a.   Use Restrictions for Reducing Applicator and
                   Reentry Exposure

                   1)   Commercial Use Only

    Vauiz funigation would be conducted only by certified commercial
applicators in commercial clothing storage .vaults.  The label would stats this.

                   2)   Use Directions

    Before fumigation all doors and leaks in the vault would be sealed.  All
persons not involved with the fuaigation would be vacated from the building.
This requirement would appear on the label.

                   3)   Reentry Period

    Following the treatment period, a full-face, black canister respirator
would be required to be worn during, reentry.  All doors and vents in the
treatment area and connected buildings would be opened at this time.  Premises
would be aerated 24 hours or until the EDS air level is less than 0.4 ppm
increased by an IR detectoc, before reentry is allowed.  The label and posted
warning signs would contain this information.

         8.   Additional Requirements for Fumigation of
              Subterranean and Drywood Termites

              a»   Qse Restrictions for Reducing Applicator and
                   Reentry Exposure

                   1)   Certified Applicators

    Fumigation would be carried out only by certified commercial applicators.
                                          -92-

-------
                                       223

                   2)   Protective Clothing

    EDB-resistant gloves  (butyl rubber), boot  covers  and apron (butyl rubber,
nitrile, or polyethylene) would be required at all  times.  The label wou!4
state this.

                   3)   Use Directions for  Dryvood  Termites

    Persons not involved with the funigation would  be required to be vacated
from premises.  Holes of size and spacing specified by the  registrant should be
drilled and the recommended dosage of  EDB inj ected  into these  holes  using a
calibrated syringe or like device.  The  label  would state these requirements.

                   4)   Use Directions  for  Subterranean Termites

    Persons not involved with the fvmigation would  be  vacated  from premises
before fumigation and subsequent aeration.  Use would not be permitted  under
slabs which contain air ducts, electrical conduits, or the  like.   Recommended
dosage would be inj ected through holes drilled in concrete  slabs which  support
infested structures.  All holes would  be plugged temporarily to  prevent  the
escape of SDB during treatment, and would be sealed permanently immediately
after treatment.  The label would state these  requirements.

                   5)   Reentry Period

    During.and after fumigation all doors,  windows  and  vents in  the  treatment
area and connected buildings would be opened.  Premises  would  be  aerated 24
hours or until EDB air levels are below 0.4 ppm before  reentry would  be
permitted.  The label and posted signs would contain  this information.

         9.   Additional Restrictions  for Fumigation of  Bark Beetle

              a.   C7ae Restrictions for Reducing Applicator and
                   Reentry Exposure

                   1)   Protective Clothing

    EDB-rssistant gloves (butyl rubber), boot covers and apron  (butyl rubber,
nitrile or polyethylene)  would be required at all times.  The  label will state
this.
                   2)   Use Directions

    SOB would only be applied to infested,  failed logs by spraying recommended
dosage onto stacked logs,  immediately covering the stack with heavy plastic,
and sealing this cover with sand snakes.  Signs would be posted on the treated
stacks stating when  the cover could be removed.  Spraying logs  in the open
without an approved covering would be prohibited.
                                         -93-

-------
                                      224

         10.  Additional Restrictions for Fumigation of
              Japanese Beetle Under USDA Quarantine Program

              a.   Restrictions for Reducing Applicator and
                   Reentry Exposure

    Under the authority of the Organic Act (7 C.S.C. 147a), the APHIS program
of USDA has established a Japanese beetle quarantine program which requires
fumigation of certain regulated articles (such as infested plants, grasses or
soil) before they can be shipped out of a quarantined area.  The APHIS Japanese
Beetle Program Manual specifies the approved pesticides and treatment methods.
A review of this manual indicates that it could be revised to include the
a-forementioned minimal, use requirements.  Further, a review of the two
currently registered labels for this use (registration numbers 1681-6 and 3743-
263) shows that these labels could be upgraded to describe all the usa
restrictions and directions for this use pattern.  Following are the additional
use restrictions or revisions in the APHIS Manual developed by the SPA for each
regulated article:

                   1)   Grass Sod

    Treated sod would be required to be covered with a heavy plastic or tarp
during the treatment period to ensure an sfficacious treatment and to minimize
air exposures.  The tarp would be required to be aired before re-use.

                   2)   Bareroot Plants

    The proper method for disposing of the used EDB-water mixture would be
specified.

                   3)   Balled or Containerized Plants

    The "enclosed place" into which plants are placed after treatment would be
posted with signs indicating the minimal reentry information described
previously.  Proper disposal methods would be described.

                   4)   Potting and Bench Soil

    Soil would be covered with plastic or heavy tarp for the duration of the
treatment period.
                   5)   Beds and Other Uses

    Soil would be covered with plastic or heavy tarp for the duration of the
treatment period.

                   6)   For All Above Uses

    The APHIS Manual states for balled or containerized plants that treatment
is approved only when beetle grubs (larvae) are present.  This requirement
would be imposed for all regulated articles to avoid unnecessary applications.
                                          -94-

-------
                                      225

V. SELECTION OF REGULATORY OPTIONS

    A.   Introduction

    The purpose of this chapter  is  to determine the most appropriate regulatory
options the Agency should propose for each  registered use of EDB.   to
accomplish this, the risks to hunan  health  and  the  environment for each
registered use are systematically compared  with the benefits which result from
that registered use.  If this comparison  demonstrates that "unreasonable
adverse effects" to man or the environment  exist for particular uses, then
specific regulatory options are  developed to  reduce the  risks for  each of those
uses.  The Agency then evaluates the extent to  which these regulatory options
will change the risks and benefits of each  use  of EDB.   Finally, by comparing
the changes in the risks and benefits the Agency selects and proposes the nost
appropriate regulatory measure for  each use of  EDS.

    In previous chapters of this docunent,  the  Agency has evaluated both the
risks and the benefits which exist as a result  of the currently registered use
practices of EDB.  In addition,  the  Agency  has  sumnarized the regulatory
options which are available under FIFRA.

    As discussed in C'.-.apter II of this docunent,  the Agency has examined all of
the available health studies on  EDB.  These studies  include investigations of
possible carcinogenicity, mutagenicity, and reproductive disorders.   After a
careful review and evaluation of these studies,  qualified Agency scientists and
outside consultants have reached several  conclusions regarding  the  possible
hazards which may result from exposure to EDB.

    Based on the accumulation of evidence,  the  Agency considers EDB  to  be a
potent animal carcinogen and a probable human carcinogen.  Because the
pesticidal uses of EDB result in hunan exposure during application,  through
contamination of some foods during application,  and  during  transport of  treated
commodities, the Agency concludes that EDB  poses  a  risk  of  cancer to  hunans.
The Agency also concludes that EDB causes reproductive disorders in bulls,  rats
and chickens, suggesting that repeated exposure to EDB may  cause reproductive
disorders in humans.  Finally, the Agency concludes  that  adequate raultitest
evidence exists to demonstrate that  EDB also causes  ntutagenic effects in
several organisms.  This evidence indicates that  exposure to  EDO poses a
potential risk of genetic damage to  man.  Collectively,  these three conclusions
confirm qualitatively that hunan exposure to EDB  poses substantial risks  to  the
public health.

    The Agency has quantitatively estimated the cancer risks  to the public
which exist due to current registered use practices.  Animal toxicity data  and
the human exposure estimates were used to evaluate the lifetime probability of
EDB-induced cancers to the general population and to occupationally-exposed
workers.  Probability estimates, based on total dietary exposure resulting  from
pesidues persisting in EDB-fumigated commodities, range  from approximately  10
  to 10  .   Applicator risk estimates range from  10   for soil fumigation
operators to 0.4 for some persons involved  in the fumigation ofcitrus.

    The Agency has determined that human exposures to EDB also pose a risk of
nutagenic effects and reproductive disorders to exposed individuals.  The
Agency has not quantitified the risk of mutagenic damage to exposed
individuals, but has concluded that these risks exist and must be taken into
consideration in determining the most appropriate regulatory actions.  Based on

                                          -95-

-------
                                      226

the NCSL for reproductive disorders observed  in  rats  (as  well  as the repro-
ductive disorders observed in bulls) and based on  the  applicator exposure
levels, the Agency has also determined  that an adequate margin of safety does
not exist.  Although the Agency does not plan to perform  a  quantitative risk
assessment for reproductive disorders,  it will propose regulatory options that
should provide an adequate margin of safety by significantly reducing exposure.

    Benefits resulting from the use of  EDB, summarized in Chapter 4  of this
docunent, are estimated to be approximately 60 million dollars per year.   Of
the estimated fifteen million pounds of EDB used annually,  over eleven million
pounds are used for soil fumigation.  However, the largest  economic  benefits of
EDB use do not result from its use as a soil  fumigant  but result from a
significantly lower volume use - the USDA APHIS quarantine  program.   This
program requires that in certain geographic areas  citrus  and certain other
commodities be funigated with EDB prior to transport in order  to  prevent  the
spread of fruit flies into geographic regions where they  are not  established.
While this program accounts for less than one percent  of  the annual  EDB usage,
it can account for over 30% of the benefits of ED3, an estimated  24  million
dollars annually.

   It is for reasons such as this that  each major  use  category is system-
atically analyzed in the following pages using these and  significant factors.
For a given use category, the risks and 'benefits are reviewed  and the data  gaps
identified.  The significant hxinan health risks are then  weighed  against  the
corresponding benefits and a judgment is made as to whether or not the risks
outweighed the benefits.  If regulatory actions are required to reduce these
risks, an analysis is then made of the  extent to which each available
regulatory action might reduce the risks.  The most reasonable regulatory
measure is identified and proposed for  that use category.

    3.   Fumigation of Stored Grains

             1.   Summary of Risks and  Benefits

    The Agency has determined that the  general population,  grain  fumigation
applicators, and individuals handling EDB-fumigated grains  are at risk to
health hazards as a result of exposure  to 2DB.  The risks include cancer,
nutagen'ic damage, and reproductive disorders.  As  discussed in Chapter II,  the
Agency has concluded that residues of SDB persist  in flour, baked breads,  and
other baked flour products made from wheat fumigated with EDB.  The  range of
expected residues which the Agency has  determined  occur in  bread  baked from
this flour vary from 0.07 ppb to 38 ppb.  Based on these  residue  levels,  the
Agency has estimated that the lifetime  risks of EDB-induced cancer^to  the_4
average U.S. citizen on a typical diet  vary from approximately 10   to 10
for this registered use.  Although the  Agency has  not  quantitatively estimated
the public's dietary risk of mutagenic  damage and  reproductive disorders,  it
has concluded that these combined risks warrant consideration  of  additional
regulatory actions.

    The Agency has no monitoring data on exposure  to EDB  for individuals  who
fuaigate stored grains.  However, based on available exposure  data from similar
use prac-.ices for other grain fumigants, the  Agency has concluded that grain
fumigant applicators are exposed to EDB through dermal contact and the
inhalation of ECB vapors.  Mixing, loading, use, and storage of SDB-containing
grain fumigants can expose applicators  and individuals in the  immediate
vicinity to SDB vapors.  Commonly occurring spills of  these fumigants on  skin

                                         -96-

-------
                                       227

or clothing also result  in  dermal  contact: for applicators.  As a result of
these exposures, about 50,000  eo 60,000 applicators and "non-involved" persons
are potentially at risk,  to  the  health  hazards of EDB.

    In addition to the above risks,  persons involved in the transport or
handling of ED8-fumigated grains as  they move through  commerce (dock workers,
grain inspectors/ etc.)are  potentially at risk to the  health hazards of EDB.
These risks result in people being exposed to vapors which have slowly
dissipated from treated  grains  and which have been trapped in storage or
transport systems.  The  Agency  has no  exposure data for this occurrence and has
not estimated the number of people exposed in these situations.  The Agency has
determined that the health  risks resulting from these  exposures,  should be
considered in the selection of  an  appropriate regulatory action for this use.

    Bystanders in the immediate area of use or transfer of EDB grain storage
fumigants can also be exposed  to SDB vapors.  The Agency has not evaluated the
exposure to these individuals  in its quantitative risk estimates because of a
lack of needed data.  There are no estimates of the number of persons exposed,
but the Agency recognizes that  such  exposure can occur and would be attended by
health risks which should be considered in the development of a proposed
regulatory position.

    The Agency has evaluated the economic benefits of  grain fumigation with SDB
and has determined that on  a national  basis there would be no  adverse economic
impacts if EDB were cancelled;  in  fact,  users would save money by using
alternative pesticides.  As discussed  in  Chapter III,  EPA has  determined that
certain available alternative  funigants,  if used properly,  are more economical
and that, on a national basis,  farmers  would save approximately 0.5 to 2
million dollars by using them.  These  alternatives,  which will be examined in
detail later in this section, are  aluminum phosphide and a mixture  of carbon
tetrachloride and carbon disulfide.

    These savings tend to obscure  the  fact that under  certain  conditions there
are advantages in using EDB.  Grain storage  facilities,  which  allow furaigant
vapors to dissipate easily  because the  storage facility is not well-sealed, may
be berter suited to EDB due to  its characteristic persistence.   However,  the
Agency has concluded that,  in most cases,  the  alternatives  will be  more
economical.

    If EDB were cancelled,  the most likely substitutes  would be a mixture  of
carbon tetrachloride and carbon disulfide  at on-farm storage facilities, and
aluninium phosphide at off-farm storage centers.   Because on-farm and  off-farm
facilities vary widely in airtightness, the  substitutes  also vary —  the
tighter the facility, the more likely  it  is  that  the substitute would  be
aluminum phosphide.   This grain fumigant is  less costly but requires a well-
sealed storage facility and a longer exposure  period to  be  effective.

          2.    Risk/Benefit Analysis

    Based on the above risk and benefit information, the Agency has concluded
that the risks exceed the benefits for EDB funigation of stored grains.  Thus,
risks of unreasonable adverse effects to the public health axist and will
continue to exist  unless regulatory actions are implemented to reduce
oreliminats these  risks.  This conclusion  is based mainly on potential dietary
                                         -97-
    418-574 O - 83 - 16

-------
                                       228

risks and the estimated negative economic  benefits.   It  is supported by the
potential health risks to grain funigant applicators,  individuals who handle
£C3-fumigatad grains, and non-applicators  who  may be exposed to EDB vapors
during storage, transfer, or use.

    Accordingly, the Agency has evaluated  several regulatory options which
night reduce these risks.  Table 29 displays the  impacts  which each regulatory
option would have on human health risks resulting from exposures to EDS,  the
health risks of the likely substitutes for  this use  of SDB,  and the expected
economic inpacts.

    All actions short of immediate cancellation would  result in continued
dietary risks.  Options 2 and 3, "Classify  for Restricted Use"  and "Restricted
Use Classification Plus Changes in the Terras and  Conditions  of Registration"
•would result in some reduction in applicator and  dietary  risks  because  some EDB
users might voluntarily switch to alternatives.   However,  only  immediate
cancellation would eliminate all EDB dietary and  inhalation  risks.   Because
unreasonable dietary affects continue  to exist for all options  less restrictive
than immediate cancellation, the Agency has concluded that immediate
cancellation of this registered use of EDS  is  necessary.   Immediate
cancellation is further supported by the availability of  alternative funigants
which are less expensive and which the Agency  considers will  pose  less  risk to
applicators and the public when compared to EDB.

    To reach determinations about the  risks of the alternatives, available
health effects-related studies on these chemicals were reviewed with these
results:

    Ose of aluminum phosphide in not expected  to  result in unreasonable risks
to applicators or the general public.  Aluninun phosphide  has an established
tolerance for many food products including  grains.   This  chemical  rapidly
reacts with water in air resulting in  the release of phosphine gas  and  a
residue of aluminum hydroxide.  It is  this  gas which actually destroys  insect
infestation.  Because phosphine is a highly reactive and  fugitive gas,  residues
on commodities or treated materials are not expected.  Phosphine  is acutely
toxic and poses a risk to unprotected applicators.   If the precautions
specified on the label are followed explicity, applicators,  should  be
adequately protected.  Aluninum hydroxide, a common  ingredient  in antacids,
does not appear to pose a health hazard, as determined by  an FDA review of
toxicology data.

    Carbon disulfide (CS ) is an acutely toxic, highly flammable and
penetrative liquid.  It LS formulated  in a mixture with carbon  tetrachloride
and other fvmigants to reduce its flammability.   Adequate  toxicological data
are* not available to determine if carbon disulfide poses  chronic hazards  to
hunans-  Likewise, no studies have been conducted to determine whether  residues
of CS  might occur in food.  Though CS  is exempt from tolerances for
grains because residues were not expected to occur in food or feed,  the Agency
considers it prudent and is initiating a study to determine if any  residues  of
CS  can be detected in grain or flour.  If residues are found, registrants of
CS  may be required to submit chronic  feeding studies.

    The third major alternative pesticide is carbon  tetrachloride.   *Tow under
RPAR review, CC1  has been presuned to present a  risk to hunans of  cancer,
chronic liver effects and chronic kidney effects.  CC1, is also suspected of
inducing mutagenic and teratogenic effects  in laboratory tests.

                                        -98-

-------
                                                Tab la 29 lapacta ot HayuUtory Optlonu on Stored Grain*
ID
Ho>jul<>tory Option
1. Cuiitlnuutl
Hdijlatratlon
EI)U Human llo* 1th
K*l>oa\>r* tout*
Inhalation, Our mil
Dietary
Hlaka Kink of Uubutltutau
Mlukit lUiann/EnvlroniBontAl
Boo Table* IB and 20 Not applicable
(or concur rlaka and
dlucuiulon in Chapter II
toe other po««lblo rlaka
Economic Impact*
Uuera
Hone
Conauaara
Nona
Cuauaenla
Unreasonable
dietary, duruul
and Inhalation
rl*ka cuntlnua
                 (or
       HoutrtutuU Uu« and
            A. Prutactiva  E
-------
                                      230

    However, the cancellation of EDB  is  not  expected  to  increase  the health
risks of CC1 , because exposures to CC1  are not  expected.   The ECB  mixture
which is applied to grain, is formulated* with CC1.  and other liquid
fumigants.  The concentration of CC1, in these mixtures  usually ranges
between 70% and 90%.  If ECB is cancelled, the predominant  substitute is a
mixture of 30% CC1,, and 20% CS_.  Thus,  roughly the same amount of CC1.
is contained in the substitute mixtures.  As a result, exposure  to CC1.  would
not be expected to increase.

    In STjnmary, from the available information, the Agency  concludes that,
aluninum phosphide and carbon disulfide  used according to labeled safety
precautions, will not pose the range  or  magnitude of  adverse health  effects to
users and to the public that the use  of  SDB  poses.  Once cancellation is
effective, the Agency will initiate action to revoke  the current  exemption for
grain from tolerances for EDB.

    The Agency must act to reduce the unreasonable  risks posed by SDB.
Cancellation is the most effective and viable option  in  light of  the  high  risks
of EDB and the fact that the alternative fumigants, which are less expensive,
are just as efficacious and will not  pose the range or magnitude  of  adverse
health effects to users or the public as demonstrated for EDB.  This decision
is enhanced by the recent development of carbon dioxide,  nitrogen and
combustion product gas as effective means of controlling pests in stored
grain.  If these gasous insecticides  are exempted from tolerances and prove to
be technically and economically feasible, they will be additional alternatives
to EDB.

    C.  gjARANTTSE FUMIGATION
         1 ..   Summary of Risks and Benefits

    As discussed in Chapter II, over 99% of the EDB  used for quarantine
purposes in 1977 was applied to citrus and tropical  fruits  (35,000  Ibs.) and
the remaining amount (300 Ibs.) was applied to miscellaneous fruits and
vegetables.  The Agency has determined that both the risks  to human health and
the economic benefits which result from this registered use may be
substantial.  The Agency is concerned that the estimated benefits may
substantially overstate the economic value which results from current use
practicas for this registration.  This issue and the fact that there are
currently no registered chemical or approved nonchemical alternatives for this
use have complicated the Agency's decision as to what regulatory action to
propose for the use of EDB in quarantine fumigation.

    In Chapter II, the Agency established that significant  human health risks
are present as a result of current EDB use practices in quarantine fumigation
programs.  These include risks to the health of the general public through
exposures to EDB residues persisting in treated commodities, and health risks
to occupationally-exposed individuals, including those working at or around SDB
fumigation centers and those working in warehouses where fumigated commodities
are temporarily stored.  The health risks are cancer, mutagenic effects, and
reproductive disorders.

    The Agency has quantitatively estimated the lifetime cancer risks to the
general public and has established that significant cancer risks result from
current practices at quarantine fumigation centers.  The residue estimates
forming the basis of estimated risks are supported by chemical analyses of EDB
residues which were found in commodities obtained from retail grocery stores.

                                        -100-

-------
                                       231

As discussed  in Chapter  II,  che  Agency has determined chat the lifetime risks
of EDB-induced cancer  to  the general  public range from approximately 10   for
citrus to 10-  for  tropical  fruits.

    The Agency has  also  determined  tha lifetime risks of cancer for persons
occupationally exposed,  including fumigation center employees, truckers
involved in the transport of fumigated commodities, and persons involved with
the movement and handling of fumigated commodities  into and  out of warehouses.
As discussed  in Chapter  II,  the  Agency has estimated the cancer risks to
fumigation center employees  to vary from  approximately TO   to 10   ,  to
warehouse workers,from approximately  0. 1  to 0.4,  to outdoor  operators of
approximately 10  , and  to truckers of approximately 10  .   These  cancer
risks are among the highest  risks the  Agency has  ever confronted.

    In addition to  these quantified risks,  the  Agency has also concluded that
adequate evidence exists to  demonstrate that SDB  causes mutagenic  effects  in
several organisms and  reproductive disorders in exposed animals.   Though the
numerical range of  the risks of  these  health hazards is not  available,  the
Agency has concluded qualitatively that humans  exposed to SD3  are  also  at  risk
to genetic damage and  reproductive disorders.   These additional risks,  together
with the cancer risks, demonstrate that the use of  EDB in APHIS quarantine
programs poses unusually high health hazards to the general  public and  to
occupationally exposed workers.

    The benefits resulting from  this  use of SDB,  discussed in  Chapter III, may
amount to as much as 25 million  dollars annually.   This  estimate is based
largely on the assumption that if this registration were cancelled, the
grapefruit export market to  Japan would be  eliminated,  resulting in a loss to
Florida growers of  about 29  million dollars in  revenue.   However,  this  loss
would be partially offset by an  increase in the value of domestically produced
mangoes and papayas, and also by the decrease in  treatment costs.

    It should be noted, however, that  the  estimated benefits from  this use may
be substantially overstated  because they are based  on the assumption  that  the
grapefruit export market to  Japan will be  lost  if this registered  use is
cancelled.  The Agency recognizes that this export  market was  established on an
understanding between  the U.S. and Japan that grapefruits and  certain other
citrus destined for Japan from Florida will be  treated with ED3 prior to
shipment as a quarantine measure against the spread of fruit flies.  The loss
of this export market rests  on the assumption that  if  this registered use of
SDB were cancelled, the Japanese would no  longer  accept  Florida's citrus
because it had not previously been treated  with SDB.   One possible option is to
cancel the use of EDB on crops destined for consumption  in the Q.S., and allow
its continued use on crops designated  for  export  to  Japan.  However, it may
happen that the Japanese government will decide that  SDB  fumigation is
unacceptable and will not allow EDB-treated grapefruit and citrus to be
imported.  Thus, the actual  impacts of cancelling this use of  SOB depend on
whether or not the Japanese  will continue  to require EDB treatments.

    The underlying assumption that there are no viable alternatives to SDB for
this use also has direct bearing on the estimated economic impact of cancelling
SDB.  There are no  registered chemical alternatives  for  this use of SDB, nor
are any -approved non-chemical treatment methods presently available.  Heat
vaporization ccoks the fruit, giving it an  "off  taste; furthermore, there are
                                         -101-

-------
                                       232

no known  available  commercial vapor-heat  chambers.   Cold treatment damages
grapefruit and other  citrus; it  cannot be used on mangoes or papaya, and causes
a loss of about  10  percent  in other miscellaneous fruits and vegetables.

    However, gamma  irradiation,  a  non-chemical method,  is a  promising
substitute for £DB  post-harvest  funigation.   In many countries,  irradiating
foods with gamma radiation  has been shown to  be an  economical,  effective and
safe method  for eliminating pests  from fresh  foods.   A  number of foreign
countries, including  Japan, have approved and are using gamma radiation for
food preservation and/or  insect  disinfestation.   Appendix 2  lists the  status by
country of the acceptability of  food  irradiation.  In the U.S.,  irradiation is
subj ect to approval of  the  Food  and Drug  Administration (FDA) under the
authority of the Food,  Drug and  Cosmetic  Act.   In fact  FDA has  already approved
its use on for use on white potatoes  and  wheat (21  CFR  179.21).   The Agency has
had numerous meetings with  both  FDA staff and food  irradiation  experts to
determine if irradiation  of citrus should be  considered as a  potential
substitute for this use of  EDB.  As a  result,  the Agency has  concluded that
irradiation  is a potential  substitute  and should  be  a factor  in  developing a
regulatory decision.  This  conclusion  is  based both  on  the Agency's  recent
discussions  (Williams,  1980a and b; Savinski  1980; Welt,  19SOa and  b;  Seeder,
19SOa) and on recommendations made by  the Interdepartmental Committee  on
Radiation Preservation of Food (ICRPF)  in their report  entitled,  "Food
Irradiation  in the United States"  (December,  1978).

    Thus," the impact of cancellation of this  use  of  EDB also  depends on  FDA
approval of  the use of gamma radiation  as an  insect  disinfestation method  for
citrus and tropical fruits, as well as  on Japan's acceptance of  this substitute
method.   If  the FDA approves and the Japanese  accept  irradiation of  citrus  *s
an alternative to EDB, then the  current estimates of the  benefits for  this  use
of SDB may be significantly reduced.   However, the Agency has not thoroughly
studied the  economic  impacts of  the use of irradiation.

    2.    Risk/Benefit Analysis

    The Agency has evaluated the risks  to  human health  and the economic
benefits which result from current use practices  and  has concluded that this
use causes unreasonable adverse  effects to humans.   Regulatory options
available to the Agency to reduce these risks  are listed in Table 30.  This
table shows  the expected  changes in risks  and  benefits  which result  from the
implementation of each regulatory option.

    Option 1, continued   egiatration,  would result in the indefinite
continuation of unreasonable adverse effects both to the general population
front dietary exposure and to workers applying EDB or handling treated  fruit.
The benefits of this use of EDB,  approximately 25 million dollars annually,
would remain unchanged.

    Under Option 2,  the Agency would impose restrictions to reduce risks for
truckers,  employees at fumigation centers and warehouse  workers.   Various means
of reducing dietary exposure to  the general population were explored in Chapter
IV,  however,  none were considered to be effective or practical.   Consequently,
this option would only reduce risks to persons occupationally exposed to EDB.
Risks to the general public would continue indefinitely  because  this option
                                        -102-

-------
                                       233

 would have no impact on residues in citrus and tropical fruits.  The benefits
 resulting from implementation of this option would remain essentially the same
 as in Option 1,  because costs to impose ehfl restrictions are small, about
 13,000 cer facility.  (See Table 30).

     Option 3 would permit SDB use to continue until a specific date.  During
 the period of continued use,  the restrictions specified in Option 2 would be
 imposed.   Because there are currently no effective registered alternatives to
 EDB,  selection of this option would allow for the development and
 implementation of alternative safer methods for use on quarantined commodities.

     With  implementation of this option the risks and benefits would-change as
 the restrictions imposed change.  The immediate impacts on risks and benefits,
 which would last until cancellation became effective,  would be the same as
 those under Option 2 (i.e.,  reduced risks to applicators,  continued high
 dietary risks to the general  public,  and essentially unchanged high economic
 benefits).  Once cancellation became effective all SDB health risks would be
 eliminated, both to exposed applicators and to the general public from dietary
 exposure  to EDB residues in treated commodities.  The economic impact of
 cancellation, if gamma irradiation or another alternative became available,
 would be  the capital investment costs, amortized over an appropriate period of
 time, plus the operating costs/ minus the costs of treatment with EDB.   The
 Agency has not examined these costs in detail, but has obtained some
 preliminary estimates of the  operating costs for an irradiation facility with
 the capacity for effectively eliminating fruit flies on about 4 million pounds
 of packaged citrus per day (Welt,  T9SOa).   At a minimum dosage level of 25
 Krad, not to exceed 50 Krad,  the estimated cost of processing is between 552
 and $90 per 40,000 Uss of citrus.   This cost estimate  is based-on 100,000 cases
 of grapefruit containing 40 Ibs per case.   Costs of the implementation and use
 of gamma  radiation as a pesticide  have been studied in great detail  in  a report
 entitled  "Food Irradiation in the  United States",  prepared by the
 Interdepartmental Committee on Radiation Preservation  of Food (December,
 1980). Conclusions from this report were that "Use of chemical pesticides such
 as ethylene oxide, methyl bromide  and ethylene dibromide could be reduced or
 eliminated."

     Option 4 is immediate cancellation of the use of EDB to treat commodities
 destined  for consumption in the U.S.,  while permitting its  use  to continue for
 fumigation of commodities destined for export, and requiring the  same
 restrictions as  would be imposed by Option 2 to reduce applicator risks.  This
 option has three possible outcomes.  The impact on risks and benefits depends
 on the possible outcomes.  Under this option,  dietary  risks to  the U.S.  public
 would be  eliminated and the risks  to fumigation center and  warehouse employees
'would be  reduced.  The economic impacts of this option could be highly variable
 depending on which outcome results.  One outcome  is that the Japanese
 government could continue to  accept imported citrus which as been treated.   If
 this outcome were to occur,  the reductions in benefits could be small under  the
 first outcome,  amounting to an estimated $1.6 million.   A  second  outcome of
 Option 4  is that the Japanese might decide to refuse EDB-treated  citrus, in
 which case the economic loss  to growers and the U.S. balance of trade would  be
 large ($25 million and S27 million, respectively).   A  third possible outcome is
 that the  U.S. could substitute gamma irradiation  for EDB and still retain the
 Japanese  export market.  In this case, the economic impact  on growers would  be
 a change  in treatment costs which  the Agency has  not evaluated.
                                         -103-

-------
                                      234

    Option 5 is  immediate cancellation of  all  EOB  used  for  post-harvest
fumigation.  This option would  have  significant  impacts on  both the risks and
benefits of EDB.  All risks of  EDB would be  immediately eliminated.  Because
the export market to Japan would  be  lost,  the  Agency  estimates that citrus
growers would lose S2S million  in revenue  losses and  that there would be  a net
$13.4 million loss in the U.S.  balance of  trade.

    Upon evaluation of the impacts each regulatory option has  on the risks and
benefits, the Agency has concluded that Options  1, 2  and 4  should be eliminated
from further consideration and  that only Options 3 and  5 should be examined in
more detail.  The reasons for eliminating  the  three options are as-follows:

    Option 1—continued unrestricted use—would result  in the  continuation of .
unreasonable adverse health effects to the general public,  to  the occupation-
ally exposed worker including fumigation center employees,  warehouse workers,
and truckers, and to those individuals living  around  fumigation centers.

    Option 2—classifying for restricted use and requiring  fumigation center
and warehouse modifications—would reduce  the  risks to  workers through the
implementation of these measures.  The benefits would remain essentially
unchanged.  However, risks of unreasonable adverse effects  to  the U.S.
population through dietary exposures to EDB would continue.

    Option 4—immediately cancelling the use of EDB to  fumigate  commodities
destined for consumption by the U.S. public, while retaining its  use  to
fumigate commodities intended for export to  foreign markets and  imposing
measures to reduce U.S. worker  risks—would eliminate all dietary risks to  the
U.S. public and  would reduce U.S. workers' risks.  However, this  regulatory
option would announce to the international community that the  Agency  has not
attempted to reduce the potential dietary residues and  health  risks  which  may
be faced by countries importing U.S. treated commodities.  The Agency views
this option as a poor approach  to the solution"of an  international
environmental problem.  The Agency rejects this option as an acceptable
alternative.

    The Agency has conducted an intensive comparative evaluation  of  the risks
and benefits resulting from the implementation of Options 3 and 5.

    Option 3—cancellation on a specific date of all uses of EDB  to  fumigate
quarantined commodities, and in the interim requiring measures to reduce worker
exposure—would  allow the benefits of this use to continue until  the date when
cancellation becomes effective.  The EDB health risks to workers would be
reduced in the interim, while the benefits of approximately $25 million
annually in revenue would be reduced only slightly by the costs to install
worker exposure  reduction measures.  In addition, positive balance of trade
payments of about $13 million annually would be retained.  With no effective
approved chemical or nonchamical alternatives to SOB currently available for
this use, the interim period would allow for the replacement of SDB with a
safer alternative.

    However, until cancellation becomes affective,  residues of EDB would
continue to occur in fumigated food commodities,  and the U.S. public would be
at risk to unreasonable adverse affects resulting from dietary exposures to
these commodities.  Persona in countries  that import commodities treated in the
U.S. would be also potentially at risk to adverse health effects.  The Agency
has determined that the lifetime probability (70 yr)  of EDB-induced cancer to

                                         -104-

-------
                                             Jabln 30.   Impacts of Regulatory Options on Post-Harvest
                                                      Ftaigatlon of Cltrua, Tropical Fruits, and Vegetables
                                          EM Hi man Health Riaka
                                                                         Rlaka of Subatitutea
                                                                                                     Economic Tmoaota
O
in
Regulatory Ootfop Bvnoaure Route
1. Continued Inhalation
Registration Dietary
2.
3.
Classify for Same as Option 1
Restricted Use
and require
Amigatlon
center and
warehouse nodl-
flcatlon (see
Chapter 1 for list
of restrictions)
Option 12, plus Sana as Option 1
cancel after a
spec If lo tine
period.
Rlaka fluman/Envlronmental
Sea Tables Ifi Not Applicable
i. 20 for cancer
risks. See
discussion
in Chapter 5
for other
possible
risks.
Reduced Not Appllctble
fumigation
center and
warehouse
employee In-
halation risks;
dietary risks
unchanged.
Same risks as Not Applicable
for Option 12
except that
all risks
are eliminated
after selected
data whon cancellation
would become effective
Uaepa Consumers
None Nona
4500.00 per Hlntaal
each fumigation
chamber $13,000
per fumigation
center for
monitoring and
analytical
devices. I/
Option 12 Minimal
plus unknown
but probably
moderate coats
for instal-
lation of
gamina irradlatora
or other alternative
treatment methods
which become
	 available 	
Cunnenta
Unreasonable
dietary and
Inhalation
risks con-
tinue.
Unreasonable
dietary risks
continue.
Thla option
could allow
time for alter-
natives to be
developed and
registered.
      1  Per chamber cost includes expenses for installation of a olosed EDB delivery (system, ventilation equipment and aafety devices (based on
      personal couununlcatlon, J. Uiltealdea to J. Panetta 7/16/80).

-------
                                        Table TO.  Impaota of Regulatory Optiona  for Poat-Rarveat
                                        (Cent.)  fumigation of Oltrua, Tropical Prulta, and Vegetables
                                     EDB Iktman Health Rlaka
 Regulatory Option
                      Exposure Route
     maka
Rlaka of Bubstltytea
Ruaan/Envl ronmental
    Economic Impaota
 UsersConsunara      Ccamenta
 4. Immediately cancel uaa of EDB on crops consumed In the U.S., but allow continued uao on exported coanodltlea, with Option f 2 reatrlotlona imposed.
o
at
                            Inhalation
Poaalble Outccneat

 A. Export market
    to Japan la
    retained
    because IDE-
    treated citrus
    la acceptable
    to Japan.
       B. Export market        Nona
          to Japan
          la lost
          because EDB-
          treated fruit
          la unaccept-
          able to Japan*
       0. Export market        None
          la retained
          through  the
          substitution
          of Irradiation.
          or other alternative
          treatment method
Reduced
occupational
rlekj U.S.
public dietary
rlaka elimi-
nated.  Rlaka
to Japaneae
unknown.
                                            All EDB
                                            rlaka elimi-
                                            nated.
                                            All EDB
                                            rlaka elim-
                                            inated.
Not applicable
                  Not applicable
                  B>e occupational
                  rlaka associated
                  with use of irra-
                  diation In the
                  U.S. are un-
                  lauvn, but likely
                  to be negllgltle.
$1.6M
(coat of
using gamma
Irradiation
not
Included).
Minimal
•AaBuaea Japan
will continue
to accept EDB
treated fruit.
                          $25M to     Conauoer    iaauoa citrus
                          uaara.      prlcea      previously
                          $16H loaa   would       exported will
                          In balance  be lowered  be abaored by
                          of trade.   due to ex-  U.S. market
                                      ceaa supply through price
                                      of oltrua   reductions.
                                      In the U.S.

                          Coat of       None      Japan la already
                          Implementing            using irradatlon
                          and using               for use on some
                          Irradiation.            fooda; however, ra-
                                                  diation could not
                                                  be adopted inmodlately
                                                  In the U.S.
                                                                                                                                                          05

-------
                                       Table ffi.  Impacts of Regulatory Options  for Poat-llarveat
                                       (Cont.)   Fuaigatlon of Citrus, Tropical Fruits, and Vegetables
                                         pt)B
                                                     Blaka,
                                              pf
Regulatory Potion
Exoosure Route
                                                          Riaka
Human/Environmental
5. Immediately cancel use on both domestic and export oropa.

      Possible outcomest
Uaera
 I
*-•
o
 I
      A. Export market         None
         to Japan Is
         retained
         because
         Irradiation or
         other alternative
         treatment la
         acceptable to Japan

      B. Export market         None
         to Japan Is
         lost because
         Irradiation la
         unacceptable to
         Japan
                                                      All EDB rlaka
                                                       eliminated
                                                      All  EDB rlska
                                                      eliminated
                                        Occupational
                                        risks of using
                                        Irradiation
                                        In the U.S. are
                                        unknown, but
                                        likely to be
                                        negligible
                                        Not Applicable
                         Same as "1C.
                                                                                                          Tmr>anta
Consumers
                                                                                                                          Commenta
Minimal   Japan and other countries
          are already using irradia-
          tion as a food preservation
          technique. But irradiation could
          not be implemented Immediately.
          No protection would be offered
          against the fruit fly until
          implementation of irradatlon.

          Irradiation
          will likely
          not be used.
          Outcome results
          eame as Option
          H, Outcome B.
                                        Consigner
                                        prices
                                        would be
                                        lowered
                                        due to
                                        excess
                                        supply of
                                        cltrua in
                                        the U.S.

-------
                                       238

                                               -4                  -5
the average U.S. citizen  is  in  the  range of 10   for citrus and 10   for
tropical fruit.  Thus,  for example, if the  Agency chose 70 years as a data when
cancellation of this use  would  become  effective,  one person in every 10,000
might incur cancer as a result  of EDB  residues in citrus,  and one in every
100,000 from residues in  tropical fruits.   In  addition to  these risks,the
public would also be at risk  to reproductive disorders and nutagenic effects.

    The Agency considers  these  high levels  of  risk resulting from lifetime
exposure to be unacceptable,  even in comparison with the considerable benefits
of SDB for this use.  To  allow  the  use of EDB  to continue  for seventy years
isunacceptable.

    Option 5 —would eliminate the risks of  EDB entirely, and for this reason
it has received serious consideration.  Without appropriate alternative treat-
ment methods however, this option would totally eliminate  the benefits of
ZDB for quarantine use.

    There is one vital  consideration which  the Agency has  extensively
investigated:  whether gamma  irradiation presents a viable alternative to ZDB
in the near future.  As indicated previously,  this method  is effective and
approved for use in many  countries, but has not yet been approved for use in
the U.S.   The USDA APHIS has been  researching this method for a decade,  and
since 1974 has been attempting  to receive approval of its  petition to FDA for
use of gamma irradiation  on  mangoes and papays.  USDA testing has also shown
good results for citrus.

    Recently, the EPA. has met with  the USDA and FDA to determine if USDA's
petition for the use of irradiation as'an alternative to SDB can be resolved.
Based on these discussions FDA  has  accelerated its review  and it appears .that
FDA could approve the use of  gamma  irradiation within 18 to 24 months.  Eased
on the Agency's contacts  with manufacturers who market irradiation systems,  it
appears that about 3 months  would be required  to  install such systems (Welt,
1980a).

    Based on this factor  alone—that irradiation technology can be implemented
within about 24 =o 36 months—the Agency believes that Option 3 presents  the
best solution.  With this option  the dietary risks to the  U.S.  population
would continue for 2 to 2 1/2 years.   The estimated cancer risks are based on
a 70 year lifetime exposure,  the  cancer risks  associated with this interia
periodqf 2 years would be about  2/70  of that  risk, or about 10   for citrus
and 10   for tropical fruit.  In  light of the  large benefits of EDB that
would continue for this period, the Agency  finds that these lower risk levels
would be acceptable, but  only for the  2 to  2 1/2  yr period.   Any further
continuation of risk would be considered unreasonable.   Regarding risks to
applicators and warehouse workers,  the Agency  finds that implementation of the
safaty measures previously described would  significantly reduce the risks  to
these persons without significantly effecting  the economic benefits.

    In conclusion, the  Agency selects  Option 3—phase-out  of EDB after no  more
than 2  1/2 years  from now, and  immediate implementation of applicator and
worker protection measures specified in Chapter IV or other measures to ensure
that the EDB air  standard of 0.4  ppm is met.  A reasonable date for the
cancellation to become  effective  -.vould be July 1, 1983. After that date,  no
use of  EDB will be allowed  for  post harvest commodity fumigation.
                                          -108-

-------
                                      239

    D.  S?CT FUMIGATION OF GRAIN  MILLING MACHINES?

    1.  Summary of Risks and Benefits

    In Chapters II and III, the Agency  evaluated the risks and benefits
resulting from spot, fumigation of milling machinery and cereal handling
equipment:.  This periodic treatment  with EDB  of critical spots in grain milling
machinery, ducts and conveyors is specifically  timed to break the life cycle of
invading pests, thus preventing the  tremendous  increase in pests which could
result from the development of a  second generation.   The Agency has determined
that this use of EDB, amounting to 465,000 pounds per year,  causes health risks
to the general public, mill workers, and fumigators.   An investigation of
alternatives revealed that the only  effective alternative for this use of EDB
is general space fumigation of the entire mill.

    Recent studies which investigated the potential for residues of EDB to
contaminate wheat flour following spot  treatment confirm that flour does absorb
residues and that they may persist for  two weeks following treatment.   The
amount of E3D which contaminates  flour  varies greatly and depends on numerous
factors including the length of time betveen  spot treatment  and wheat
processing, the concentration of SDB in  incoming wheat,  and  the construction of
the flour mill.   The Agency has determined that the  average annual dietary
burden resulting from this use of EDB ranges  from 0.7x10   to 2.77 x 10
mg/kg/yr.  The Agency's Carcinogen Assessment Group  used these  residue levels
and estimated that the lifetime cancer  risks  to  the  U.S.  public range  from
approximately 10   to 10

    These residue levels and the corresponding  dietary  cancer risks  are.viewed
by the Agency as upper bound estimates of the actual  cancer  risks.   The
estimated residue levels in bread include the assumption that all flour mills
use EDB as the method of treatment of milling equipment  for  pest  prevention.
The number of mills using this process  is unknown,  as is the  amount  of wheat
flour processed annually through these mills.   In  the absence of  this
information, the Agency has assumed  that all  mills  use  EDB.

    The Agency has determined that spot  treatment  funigators  and  mill  employees
are exposed to SDB vapors.  These conclusions are  based  on findings  from  actual
EDS air concentrations measured during and after  spot treatment.  To estimate
the risks to fumigators of EDS-induced  cancers,  the Agency has  considered  both
typical exposures and, potential "worst-case exposures"  (involving a  fumigator
who moved or adjusted his face mask).  The estimated  lifetime risks  range  from
approximately 10   to 10   for applicators and  10   to  10   for mill
workers.

    2DB, used as a spot funigant in  flour mills, also presents  potential
risksof mutagenic effects and reproductive disorders  to  the general public,
mill workers and fumigators.  Although  these  risks have  not been quantified,
they must also be considered along with  the cancer risks 'in weighing the risks
and benefits of this general use of  EDB.

    The Agency estimates the economic benefits of spot fumigation to be in  the
range of 4.6 to 7.5 million dollars  annually.   These benefits are based on the
assumption that, if this use of EDB  were cancelled, the alternative would be
general space fumigation with methyl b.romide or aluminum phosphide.  The costs
ara estimated to vary from 13 to 21  thousand dollars  per year for individual
wheat and rye mills,  and from 9 to 14 thousand dollars per year for dry corn

                                          -109-

-------
                                      240

mills.  The resulting annual consumer  impacts -would  vary  from  an  estimated 1,
to 2.8 cents per capita per year for four general space funigations,  to  2.3
cents if five such fumigations are required.
aluminua
131°C),
    The Agency has reviewed available data on the two  substitute chemicals,
methyl bromide and phosphine, the active  ingredient and  degradation  product  of
    jinum phosphide.  Unlike 2DB, which is a liquid at  room  temperature  (b.p.
        tha alternatives (methyl bromide  and phosphine,  the active  ingredient
of aluminum phosphide) are gases and would not be expected  to  leave  significant
residues on treated surfaces and commodities.  Both phosphine  and methyl
bromide are acutely toxic, posing health hazards to unprotected applicators.
However, available toxicology information and consideration of the physical
properties of these chemicals lead the Agency to conclude that because of  the
low expected residue level, neither would cause unreasonable adverse chronic
effects.  Methyl bromide is known to cause mutations in  some laboratory test
systems.  Application and ventilation practices which  protect  applicators  from
acute hazards should also provide significant protection to  guard against  the
potential mutagenic hazards of methyl bromide.

          2.  Risk/Benefit Analysis

    Based on the above risk and benefit information, the Agency has  determined
that the risks of adverse health effects exceed the economic benefits which
result from the use of SDB in spot fumigation of grain mills.  Therefore,
unreasonable adverse effects to the public health exist  and  will continue  to
exist unless the Agency adopts regulatory options to reduce them.

    This determination has led the Agency to evaluate  several  regulatory
options which might reduce these risks.  Table 31 presents  the impact of each
option on the risks and benefits of spot fumigation.

    Option  1, continued registration, would result in continued risks of
adverse health effects to the general population through dietary exposure, and
to applicators and flour mill employees through dermal and  inhalation exposures.

    Option  2, imposing certain use restrictions  see details in Chapter IV)
would reduce applicator and general mill worker inhalation  exposures to the
UIOSH recommended level of 0.4 ppm.  Applicator dermal exposuras would also be
reduced.  However, this option would not reduce the residues of EDB  which
contaminate flour processed through grain milling equipment  fumigated with SDB.

    The intent of Option 3 is to achieve  the same exposure  reductions as Option
2, and also to lower the EDB-contamination of flour.  By requiring that the
first batch of flour, produced after a mill has been fumigated, be discarded,  a
portion of  the EDB residues would be eliminated.  However,the  available
information is not adequate to determine the degree of reductionin residues
that might be achieved by this measure.  Further, it would be  difficult to
enforce such a restriction or to expect a high degree of voluntarycompliance.
Consequently, Option 2 would not likely result in a significantreduction of the
dietary residues of EDB from spot fumigation.

    The impacts of Option 4, cancelling use after 5 years and  restricting  use
for the interim period, is the same as Option 3 until after  the 5th  year,  at
which time  the impacts would be the same as Option 5, i.e.,  cancellation.  The
purpose of  this option would be to allow tine for flour  mills  to upgrade their
facilities  to use methyl bromide or aluminium phosphide.

                                          -110-

-------
                      Table 31  Impacts of Regulatory Options for Spot Fumigation
EDB Itaan Health Risks
Rlek of Substitutes
Economic Impact
                                                                                                        Coomenta
Regulatory Options Exposure Route
TT
2.
T~
4-
b.
Continued Dietary
Registration Inhalation
and Dermal
Classify for Same as Option 1
Restricted Use;
impose uae restrictions
listed in Part IV
Option 2 flame as Option 1
humfjn consumption of
flour allied during
the first several
hours
after H)B-treatnient
Cancel after £ years Same as Option 1
and impose Option 3
for the interim
Cancel Immediately No Exposures
Risks
fee Tables 16 and 20
for cancer risks and
discussion In
Part II for other
possible risks.
Dietary Risks
continue.
Reductions In
applicator and
non-applicator
risks.
Reduced applicator
and non-applicator
risks. Dietary
risks should be
reduced but the amount
reduction Is not known
Risks same
ea In Option 3 for
5 years; no risks
after 5 years.
All B)B risks
eliminated
ILnan Environmental
Hot Applicable
Hot Applicable
Not Applicable
of
After 5 years, the risks
of alternatives are the
same as in Option 5«
Aluminum phosphide JAP)
and methyl broulde (MB)
are acutely toxio
KB is a. suspect
•utagen
Users Consumers
None Nona
Respirators Nona
and clothing
cost about $100.
label and reentry
impacts are
negligible.
Coat of Negligible
discarding
flour woa
not analysed,
but is
probably minor.
None None
J5.1-I6.4H 1.6 to
par year 2.8 cents
for 293 per capita
wheat and
rye nllla;
$1.3M/yr.
for 93 corn
mils, total
max.inpuot •
57.? H/yr


Applicators would
probably comply
with restrictions.
Flour mills might
not comply with the
holding restriction
and enforcement
would be difficult.
During the 5 year
period, some mills
could modify their
facilities for use
of ice thy 1 bromide
or aluminum


-------
                                       242

     Immediate cancellation,  Option 5,  would force EDB users to switch
 immediately  to  alternative  funigants  at a total increased treatment cost of
 $7,700,000 per  year  for  all  391  grain  mills,  or 2.8 cents per consuner.  While
 this  is  a "worst-case"  estinate  (because it assumes all mills now use EDB), the
 potential cost  of  upgrading  some mills to allow raaxiaan effectiveness from
 space  fumigation has not been  included.  In any event,  this option would
 eliminate the risks  of EDB and '^ould  result in use of apparently safer
 substitute fumigants.

    As the data presented in Table 31  demonstrate, any action short of
 immediate cancellation will  result in  continued dietary risks.  If the Agency
 wera  to  classify EDB for restricted use,  or amend the terms and conditions of
 registration, some users nay continue  to  apply EDB rather than switch to
 alternatives.   Although  occupational  risks would be reduced,  this option would
 not result in a reduction in the risks to the general public  resulting from
 dietary  exposure.

    In conclusion, the Agency  believes that the health  risks  of EDB are serious
 and unacceptable when considered with  the benefits,  and will  only be reduced
 through  immediate  cancellation.  Cancellation will not  result in unreasonable
 economic costs  to  grain  mills  or consumers.   There are  alternative chemical
 fumigants for this use which the Agency believes will not cause the public and
 properly protected applicators to  be at risk  to the  range or  magnitude of
 serious  adverse health effects that exist with use of EDB.

    E.   PRE-PLANT  SOU. FUMIGATION

    1.   Summary of Risks and Benefits

    The  Agency  has evaluated the hunan health  risks  and economic  benefits  of
 the use of EDB  as a  pre-plant  soil  fumigant.   The  health  risks  of  concern  are
 oncogenicity, mutagenicity and reproductive effects.  These risks  exist  for
 fumigation appl-icators and other individuals  directly involved  in  application
 operations, and might exist  for  the general public potantially  exposed to  EC3
 residues in crops grown  in firrigated soil.

    The  maximum hypothetical risks  to  humans  resulting  from potential  dietary
 exposure were estimated  for  the  16  soil-fumigated  crops listed  in  Table  8  of
 Chapter  II.  As discussed in Chapter II,  no detectable  residues of EDB have
 been found at a level of detection of  0.001 ppn.   Although the  analytical
 method used in  these tests has not been completely validated,  the  Agency
 assumes that residues will not exceed  this level.  To determine an  upper-bound
 lifetime probability of  EDB-induced cancer resulting  from dietary  exposure, ths
 Agency conservatively assuned  that SDB residues  are present at  this  level of
 detection.  Based on the  assumption that  a maximum of 0.001 ppm EDB  is present
 in each soil-funigated' crop, the maximum  lifetime  probability of EDB-inrtuced
 cancer for the  general public  varies from  approximately  10   to 10
 However,  because actual  levels of EDB  are probably below  0.001  ppn,  the actual
 cancer risks are also expected be  lower than  these maximum risk estimates.  The
 Agency has also concluded that if residues do occur,  they also  pose a potential
 risk of genetic damage and reproductive disorders  to  man.

    Soil  fumigation applicators  and those  individuals involved  in the transfer
of soil  fumigants are likely to  incur  inhalation and  dermal exposures during
 SDB soil  application and  transfer.   Soil  funigation operators are exposed to
EDB  ranging from 0,7 ng/yr  for  peach crops to between 303.3 and 522.4 mg/yr

                                          -112-

-------
                                      243

for pineapples.  The Agency used  these  exposures  to  estimate that thg life time
risk of EDB-induced cancer to soil  fucxigators  is  in  the  range of 10   to 10
    As discussed in Chapter II, the Agency has also  concluded that exposures
to SOB may result in hazard of genetic  damage  and  reproductive problems to
exposed individuals.

    In addition to the potential  for residues  occurring  in  crops,  the Agency is
concerned that SDB might leach into groundwater as a  result of current use
practices.  Although no data are  available on  the  leaching  potential  of EDB,
residues of the soil fumigant dibromochloropropane (DBCP),  which has  similar
use sites and use practices and chemical properties,  have been found  in ground
water.  The Agency views this data gap  as a significant  issue  which must be
resolved before final regulatory measures can  be determined for this  use.   As a
result, the regulatory actions proposed for this use  should be considered  as
interim measures which will be finalized when  adequate information  is available
to evaluate this issue.

    The Agency has evaluated the  economic impact of cancelling soil funigation
with EDB and has determined that  this use has  substantial economic  benefits.
All alternative treatment methods would be more expensive to farmers.
Depending on the crop treated, the total increased costs to  farmers of  using
alternative fumigants would vary  from S3,500 for citrus  to  33,000,000  for
vegetables.  The total user impacts of  cancelling all preplant  soil funigation
uses would approach $21,000,000.

    Numerous alternative pesticides are available  for use as soil  funigants in
place of EDB.  The major alternatives in most  cases are equally  as  efficacious
as EDB.  The alternatives for most crops are DO, -Telone  II  and-Vorlex.   For
pineapple uses, OBCP must be used in combination with DO or Telone.   There  are
both substantial health risks as  well as a number of  data gaps associated  with
these substitutes, as shown in Table 26.  A recent tICI study shows  Telone  to be
oncogenic.  DD and Telone have been demonstrated to be cmtagenic in laboratory
tests-.  DBCP is currently suspended for all uses except on  pineapples due  to
its oncogenic, rautagenic and reproductive hazards.  The  remaining alternatives
(Vorlex, Vemacur, Mocup, and Dasanit) have data gaps  for most  chronic effects
hazards.

          2.  Risk/Benefit Analysis

    Based on the above risk and benefits information, the Agency has concluded
that there are significant health risks to soil fumigation  applicators due  to
exposure to EDB.  The risks of cancer,  mutagenic effects and reproductive
disorders to the occupationally exposed exceed the economic  benefits  for EDB
soil funigation.  Therefore, it is necessary that the Agency consider
appropriate regulatory options to reduce these potential "unreasonable adverse
effects to the public health."  The Agency considers  the potential  for EDB
residues to occur in water supplies and crops  grown in funigated soils to be
data gaps which must be resolved.  The  regulatory measures  which are developed
below are interim actions intended to reduce the health risks of EDB use as a
soil fumigant.  When the data gaps are  resolved, the Agency  will reevaluate
this use to determine if additional regulatory measures are  needed.

    Table 32 identifies three possible  regulatory options.   They are  in order
of increasing restriction: continued registration without restrictions,
continued registration with restrictions, and  cancellation.  Changes  in  the
risks and benefits of EDB expected to result from each option are also shown.


                                         -113-
     418-574 0-83-17

-------
    table 32  Impaota of Regulatory Optlona on Soil fumigation
H)B Human Health Rlaka              Riak of Substitutes
Econottlo   Impaota
Connenta
Regulatory Optlona
1.
2.
3-
Continued
Registration
Continue Registration,
but require Applicator
Use Restrictions, Set
Tolerances for EDB In
Fbod >opa, Require
Studies of Leaching
Potential and Pood
Residues
Immediate
Cancellation
Exposure Route
Inhalation and
Dermal for
applicators.
Potential
dietary and
drinking water
for the general
public.
Inhalation and
Dermal for
appl lea tore.
Potential
dietary and
drinking water.
for the general
public.
Rlaka
See Table IB
and 20 for
cancer rlaka
and discussion
in Chapter 5
for other
possible risks.
Dermal and
Inhalation
exposure risks
reduced
All EDB rlaka eliminated
Human/Envl ronmental
Not applicable
Hot applicable
DO and Telona are
oncogenlc and suspected
mutagena. Vorlex la
acutely tonic, but
haa no chronic effeota
data.
Users Consumers
None None
$175.00 llone
to
$600.00
per
applicator
$3,500 Nona
to
S3.6H,
depending
on use-site.

Unreasonable
applicator
health risks
continue.
Data gaps exist
for grounduater
contamination and
food residues.
Monitoring and
enforcement will
be necessary.
Monitoring data
needed for potential
of groundwuter
and food contamination.
EDB tolerances
must be established.
All substitutes
present potential
risks to human
health.
                                                                                                                      to

-------
                                       245

     Option  1,  continued  registration  without restrictions, would result in
unreasonable  adverse  effects  to  applicators, as concluded above.  Until more
definitive  data  are provided  by  registrants, the Agency also assumes that risks
to  the  public  may  result from potential exposure to EDB in food and/or drinking
water.

     Option  2,  continued  registration  with  restrictions, would significantly
reduce  risks  to  applicators,  but would have  no  impact on the risk from
potential residues in food  or drinking water.   As a component of this option,
the  Agency  will  require  a study  of  the potential of EDB to contaminate food
crops grown in EDB-fumigated  soils, and laboratory and field studies of the
leaching potential of EDB.  The  food  residue study will investigate if residues
of  EDB  in crops  grown in EDB-fumigated soils (at maximum application rates)  do
not  exceed  the current limit  of  sensitivity  of  analytical methodology of .001
ppcu  The studies  on  the leaching potential  of  EDB will be subject to the
Agency's review  and evaluation prior  to commencement.

     Immediate  cancellation, Option  3,  would  result in significant economic
costs to growers,  but would eliminate all  health risks from EDB.   Losses to
growers would  amount  to  almost 21 million  dollars.  In. addition,  if this use
were cancelled,  users would substitute other soil funigants which nay also pose
significant health hazards  and have nunerous data gaps.  If this use were
immediately cancelled, the  Agency fcould have no assurance that  the net risks to
public  health  would decrease.

     The Agency selects Option 2  as  the appropriate regulatory option for thrae
reasons:  First, this option  will establish  risk reduction measures  which  lower
applicators-' risks at a  reasonable cost.   The specific use restrictions for  the
protection  of  soil funigation  applicators  include all those measures  specified
in  Chapter  IV: classifying  EDB for restricted use by certified  applicators
only; raquiring  applicators to wear a  NIOSH/MSHA-approved  full  face  black
canister respirator during  application;  establishing a 24-hr  post-application
period  iuring  which any  person reentering  the treated area must wear  a
respirator; requiring EDB-resistant (butyl rubber)  gloves  ar.d boot covers
(butyle rubber,  nitrile  or polyethylene) be  worn during mixing  and loading of
EDB  as  well as during the maintenance  of application equipment; and  requiring
signs be posted  on the treated site to notify readers of the  treatment  with  ED3
and  bearing further warning information.   These restrictions  will result  in  a
significant reduction in worker  exposure,  while maintaining the economic
benefits of SDB.

     The second reason for selecting Option 2 is that the risks  to public health
{pending resolution of the data  gaps)  which  might result from the  undetected
but  possible exposure to EDB,  are outweighed by the  economic  benefits.   In
determining the  risk  of  cancer due to  consumption of  SDB-funigated crops,  the
"Agency  assumed that a maximum  of 0.001 ppm of EDB was  present on  each crop.
The ;astimate derived  from this assumption  is  presuned  to be a worst-case
estimate because 2DB  has not been detected in analytical tests on  such  crops  at
a level of  detection  of  0.001  ppm.  To ensure that any  EDB  residues remain
acceptably  low (i.e.,  <  .001 ppm), the Agency proposes  to  establish a tolerance
for  residues at  the level of  detection of  EDB per se  in  or  on food crops grown
in EDB  Jvnigated soils.

     The third  reason  for choosing Option 2 is that it  will  enable  the Agency  to
resolve the data gaps concerning the potential  for SDB  to contaminate food
crops and/or to  leach into water  supplies.   Until these  data are  available, it
                                        -115-

-------
                                       246

 is not  possible  to  estimate  quantitatively the risks for these health hazards
 of EDB.  The Agency therefore  proposes to  use Section 3(c)2(B)  of FIFKA to
 quickly obtain these additional  data.  When results of these studies are
 available and have  been evaluated,  the Agency will reevaluate whether
 additional regulatory actions  are needed  for this use.

    Following the suspension of. DECP as la  preplant for  soybean nematode control
 in 1979, several states have claimed there are no other registered pesticides
 that are economical or effective against neraatodes.   Based  on the available
 test results showing EDB  to be as efficacious as DBCP,  six  states applied for
 temporary permits (e.g.,  exemptions from FIFBA)  to allow this use under section
 18(c) of FIFKA.  These exemptions were granted with the requirement that
 numerous restrictions be  implemented to reduce exposures to EDB and the
 associated health risks to applicators. These exemptions were granted for the
 period  beginning April 24, 1980  and ending July 31,  1980.   Table 3 lists these
 restrictions, those applicants who  applied for and received exemption,  the area
 and acreage to be covered, target pests, and the SOB products approved  for each
 use.
    F.  Minor Jses—Beehive Supers,  Vault  Fumigation. Termite Control,  Bark
        Beetle Control, and Japanese Beetle Control

    1.  Summary of Risks and Benefits

    The Agency has  examined the  potential  human health  risks and the  economic
 benefits for four of the  five  known lower   volune uses  of EDB.   A risk/benefit
 evaluation of the fifth use—subterranean  and drywood termite control—will be
 deferred until a cluster  analysis for  all  pesticides used for termite control
 is conducted.  The  uses which  are evaluated in this section include fmigation
 of stored beehive supers  and honeycombs (to control  greater wax noth  larvae),
 vault fumigation of stored clothing and furniture (to control various pests),
 treatment of felled logs  (to control bark  beetles),  and treatment  of  APHIS
 quarantined plants,  soil  and grass  sod (to control Japanese beetles).   The
 health  risks include cancer, genetic damage,  and reproductive disorders.

    As  discussed in Chapter II,  quantitative estimates  of the risks are not
 available for noat  of these uses.   However,  as discussed below,  an  examina-
 tion of the use  practices for  each  of  these uses reveals that applicators  are
 likely  to be exposed  to EDB during  use  and that  they may be  at  risk to
 unreasonable adverse  effects.

    Vault fumigation  involves  the pouring  of EDB  into a heating  device, which
 evaporates the EDB  to penetrate  items being treated.  The formulation used in
 vault fumigation is 5% EDB and the  period  for  funigation is  generally 16
 hours.  Thus, vault fuaigators can be  exposed  dermally  during the pouring of
 th^ formulation  into  dispensing  containers and through  inhalation during
 pouring and when re-ntering the vaults.  The Agency  currently has no infor-
mation on the amount of EDB used annually  for  vault  funigation,  how many of
 these facilities exist in the  U.S.,  how many vault fumigation employees there
 are in the U.S.,  or the levels of EDB to which vault funigation can be exposed.

    No  data exist on  th<» exposures of beekeepers  to  EDB.  However, use
practices for the fumigation of stored beehive supers suggest that beekeepers
nay be  exposed both  through inhalation  and  dernal  contact but less frecuantly
 than applicators  for all other registered uses of EDB.  Because this firrigation
 procedure is an annual practice among certain  beekeepers, exposures only occur
once a yeer when  stored hives and supers are treated during the off season.

                                        -116-

-------
                                       247

Generally, one to two tablespoons of 33% SDB  formulation are placed on
absorbent material which is set on stacked  supers;  the stack is then covered
with plastic sheeting for 24 hours.  Both dermal  contact: and inhalation of
vapor can occur during use.  But the Agency has no  estimate of the air levels
of EDB that could result from  such use practices,  nor of the frequency or
quantity of EDB dermal contact which could  occur.   However, the number of
beekeepers who use SDB could be large,  because  20,000 Ibs.  of EDB are estimated
to be used annually for funigation of  supers  and  hives.

    A separate but related issue for this use of  EDB  is  the possibility that
EDB residues could occur in honey as a result of application to stored honey-
combs.  As discussed in Chapter II, the  Agency has  no data  on EDB in honey,  but
views this as a very unlikely possibility.  The Agency considers it as a data
gap which must be resolved to determine  the total  health risks which may exist
for this use.

    No information exists on the EDB air levels to  which individuals are
exposed during the treatment of felled logs.  However, current  use  practices
suggest that both dermal contact and inhalation exposures will  likely occur
during this use.  Generally, a formulation  of 23% EDB is applied to  felled logs
in one of two ways:  spraying to the point  where  runoff  of  Liquid occurs, or
spraying at a lower rate to stacks of  wood  which are  subsequently covered under
a plastic sheet.  Applicators thus may be exposed  to  EDB der-r.ally and through
inhalation of vapors during loading of sprayers as  well  as  during actual  use.
The use of plastic sheeting appears to offer  some protection against  inhalation
exposure both because it captures vapors and  because  of  the reduced  quantity of
formulation needed.  Though the Agency has  estimated  that 20,000 Ibs.  of EDB
were used in 1977, enough to treat an  estimated 63,000 trees,  there  is no
information on the nunber of users or  the frequency of use.

    For funigation of quarantined plants, soil and  grass sod.  the  Agency has no
information on applicator exposures, nuaber of applicators,  frequency of  us*» by
applicator, or the amount of EDB used  for this site.  This  use  is required by  7
CFR 201.48 (APHIS Japanese Beetle Domestic  Quarantine Program).   Current use
practices suggest that users may be exposed both through dermal  contact and  to
inhalation of SDB vapors.  For each of these  uses,  EDB comprises either 2.5% or
20*% of the fontrulation (see Benefits Chapter).  For all  applications,  the
formulation is mixed with water and applied in very low  concentrations.  During
the mixing of the formulation with water, users may contact  SDB  dermally or  via
inhalation.  During application of the diluted solution,  exposure of  users to
EDB will also occur.  Because this solution contains  very small  quantities of
EDB, exposures are likely to be very Low.

    The four minor uses of EDB each have economic benefits.   For be-hivp
supers the benefits are considerable,  amounting to  nore  than S1C million
annually.  EDB is more effective than  either of the alternatives, paradichloro-
benzene (PD3) and carbon dioxide.  ?DB  is generally much more practical to use
than CO.,, which requires elaborate monitoring equipment  to  ensure efficacy.
The substitution of PDB would result in  increased treatment  costs of  about $2.7
million annually (approximately 5 cents  per super)  due to its higher cost and
to the additional treatments required.   Because the alternatives  are  not as
effective as EDB, the quantity of domestic  honey produced would be reduced by
an estimated 6 percent annually, resulting  in a loss  to  beekeepers of  36.5
million.  A production loss of this magnitude uould shift seine of the  cost
impact to the consumer, but the extent of this impact cannot be  estimated.
                                         -117-

-------
                                      248

    Use of EDB on infested, felled trees to control the pine bark beetle  in
western states, primarly benefits private land-owners.  Although the  extent  of
EDB use does not appreciably affect the value of commercial timber harvests,
control of the pine bark beetle on residential lots may prevent tree  mortality
losses of 51,500 to $3,000 per lot.  The use of the alternative endosulfan
could actually reduce total costs by as much as 336,000 annually, in  spite of
the fact that it may not be as effective as EDB.  The other alternative,
lindane, could reduce treatment costs by up to ?16,000 annually.  However,
because lindane is in RPAR and is recommended for cancellation for thus use, it
was not evaluated as a viable alternative.

    The extent of the use of EDB in both vault fumigation of stored clothing
and furniture, and in the APHIS/USDA Japanese beetle quarantine program is
unknown.  The economic benefits therefore, could not be quantified.   A
registered acceptable alternative in the APHIS/USDA quarantine program is
methyl bromide.  Four registered alternatives are available for vault fumi-
gation: a carbon tetrachloride/ethylene dichloride mixture, chloropicrin,
methyl brosiide, and sulfuryl fluoride.  However, methyl bromide requires
extreme care  in use, as described in Chapter III.

    2. Risk/Benefit Analysis

    The Agency has concluded that sufficient information exists to propose
cancelling the use of EDB to fumigate felled logs.  The basis for this decision
lies in the fact that applicators are at risk to unreasonable adverse effects
of EDB, and because there are no known benefits for this use.  Use of endo-
sulfan in place of EDB would result Ln a total reduction in treatment costs of
about $36,000 annually, even though it is not as effective as EDB.  The Agency
concludes that the potential health risks to applicators through both dermal
contact and inhalation exposures to SDB outweigh these negative benefits.

    In view of the lack of.risk and benefit information for the remaining three
uses, the Agency is unable to perform an adequate risk-benefit balancing
analysis.  However, because applicators could potentially experience  unreason-
able adverse effects of EDB, the Agency will require registrants to submit
certain data necessary to adequately evaluate the risks and benefits of each of
these lower uses.  Because time will be needed to obtain these data,  evaluate
it, and perform a thorough risk-benefit balancing, for the interim the Agency
has examined a number of actions designed to reduce applicator risks at a minor
cost.  Summarized below by use are actions the Agency is proposing and data
which the Agency will require of registrants.  These are organized on a use-by-
use basis for three of the four minor uses: fumigation of beehives and supers,
APHIS Japanese Beetle Control Program, and vault fumigation.

    For fumigation of beehive supers and honeycombs, the Agency is proposing
the following interim label changes for such EDE products:

    Certified Applicators — Fumigation may be conducted only by .or under the
supervision of a certified applicator.

    Protective Clothing — EDB-resistant gloves (butyl rubber only),  boo-.
covers and apron (made of butyl rubber, nitrile, or polyethylene)  will be worn
at *11 times.
                                         -113-

-------
                                      249

    Use Directions — Supers and honeycombs must  be  placed  in  a gas-tight
room or under a gas-tight covering such as a polyethylene tarpaulin  held down
with sand-filled "snakes."  All windows, doors and vents must  sealed.   All
persons not involved with the funigation must be  vacated from  the building.

     Reentry Period — Following the treatment period, a full-face,  black
canister respirator must be worn when reentering.  All doors,  windows  and vents
must be opened at this time.  Aeration for 24 hours  or until EDB air levels  are
less than 0.4 ppm measured by IS detector shall be required before reentry is
permitted.  The label and warning signs (See Chapter IV) must  contain  this
information.

    Limit fumigation to beehive supers that are clean and in storage.  — 7o
avoid possible contamination of honey or honeycombs, the label  will  prohibit
treatments other than on clean supers or comb mounted in frames in storage.
This restriction is in accordance with current practice.

    In addition to these label changes, the Agency is proposing to require the
following data from registrants:

    1. ZDB air levels to which applicators are exposed during  typical
conditions of use.

    2. Number of users.

    3. Residue studies to determine if EOB is detectable in honey from
fumigated supers and combs.

    For vault storage, the Agency is proposing to require the  following  label
changes and use restrictions:

     Protective Clothing — Require Che sane protective clothing listed  for
beehive supers.

     Certified Commercial Applicators — Vault funigation may be conducted only
by certified commercial applicators in commercial fumigation vaults.

     Use Directions — Before ftanigation, all doors  and leaks  in the vault must
be sealed.  All persons not involved with the funigation must be vacated  from
the building.

     Reentry Period — Following the treatment period, a full-face, black
canister raspirator aust be worn during reentry.  All doors and vents in  the
treatment area and connected buildings must be opened at this  tims.  Premises
must be aerated 24 hours or until EDB levels are  less than 0.4  pan as measured
by-O detector before reentry is allowed.  Posted warning signs {See Chapter
IV) must bear this information.

    In addition to these label changes, this Agency proposes to  require the
following information from registrants:

     1. .'tonitoring studies of the SDS air levels to vhich applicators may  be
exposed, both during pouring .EDB into dispensing  containers and also during
reentry into  the vault after fumigation.
                                         -119-

-------
                                      250

    2. Survey data on the number of vault funigation operations using EDB and
the nvraber of applicators involved in this practice; and

    3. Data on the amount of EDB applied annually for this use.

    For the APHIS Japanese Seetle Control Program, the Agency is proposing  to
require the following label restrictions and revisions in the APHIS Manual:

     Grass sod — Treated sod must be covered with a heavy plastic or tarp
during the treatnent period to ensure an efficacious treatment and to minimize
air exposures.  The tarp must be aired before re-use.

     3areroot Plants — The proper method for disposing of the used EDB-vater
mixture must be specified.

     Sailed or Containerized Plants — The "enclosed place" into which plants
are placed after treatment nust be posted with signs indicating the minimal
reentry information described previously.  Proper disposal methods must be
described.

     Potting and Bench Soil — Soil must be covered with plastic or heavy tarp
for duration of the treatment period.

     Sees and Other Uses — Soil must be covered with plastic or heavy :ar? for
duration of the treatment period.

     For All Above Uses — The APHIS Manual states for balled or containerised
plants that treatment is approved only when beetle grubs (larvae and pupae) are
present.  This requirement must be imposed for all regulated articles to avoid
unnecessary applications.

    For this use, the Agency is proposing to require the following information:

    1.  Monitoring studies of the air " evels to which applicators are exposed
during mixing and during use.

    2.  Data on the amount of EDB used annually, the nunber of applicators and
the frequency of use.

    In addition to these actions, the Agency is proposing to require certain
interim measures for the use of EDB in the control of termitas.   As discussed
earlier, the Agency is proposing to defer a risk/benefit balancing decision for
this use and will include it in a cluster analysis of all pesticides used for
temite control.  In the interim, however, the Agency recognizes that both
applicators and persons in the immediate treatment area can be axposed and are
therefore at risk to unreasonable adverse affects.  To reduce exposures of
applicators and non-involved persons to EDB, the Agency proposes the following
use restrictions be included on the label:

     Certified Commerieal Applicators — Fumigation may be conducted only by a
certified commercial applicator wearing a full-face, black canister respirator;

     Protective Clothing — EDB-rssistant gloves, boot covers and apron will be
required at all tiaes;
                                         -120-

-------
                                      251

     Use Directions for Drywood Tarnites — Persons not  involved  with  the
fumigation must leave the premises.  Holes of size and spacing specified by the
registrant must be drilled and the recommended dosage of EDB  inj ected  into
these holes using a calibrated syringe or like device.

     Use Directions for Subterranean Termites — Persons  not  involved  with ths
fuaigation must leave the premises before fumigation and  subsaquent aeration.
Use is not permitted under slabs which contain air ducts, electrical conduits,
or the like.  Recommended dosage must be injected through holes drilled in
concrete slabs which support infested structures.  All holes must be plugged
temporarily to prevent the escape of EDB during treatment, and must be sealed
permanently immediately after treatment*  The label must  stats these
requirements.

     Reentry Period — During and after funigation all doors, windows  -ind
vents in the treatment area and connected buildings would be opened.   Premises
must be aerated 24 hours or until EDB air levels are below 0.4 ppra before
reentry is permitted.  The label and posted warning signs must contain this
information.

    In addition, the Agency is proposing to require registrants to provide data
which would be needed in the cluster analysis.  These data are:

    1.   Air concentration of EDS to which applicators are exposed, their
         annual time spent using SDB, the number of applicators using EDB in
         the U.S., and the amount of formulation sold annually; and

    2.   Air monitoring studies which investigate the potential SDB air levels
         to which persons not involved can be exposed.  These -would include
         indoor places at or near the site(s)  of application.
                                         -121-

-------
                                       252

                                 Bibliography

Alper, M.D. and B.N. Ames.  1975*  Positive  Selection of  the  gal-chl Region of
   the Salmonella Chromosome as a Screening  Procedure for Mu tag ens  that Cause
   Deletions.  J. Bact., 12(1}:259-266

Amir, 0.   1973.  The Sites of the Spermicidal Action of Sthylene  Oibromide in
   Bulls.  J. Reprod. Fert., 35(3):519-525

Aair, 0. and E. Ben-David.  1973.  The Pattern of Structural  Changes Induced
   in Bull Spermataasa by Oral or Inj ected Ethylene Oibromide (EDB) .  Ann.
   Biol. Anim. Biochem. Biophys., 13(2):165-170

Amir, 0. and V. Lavon.  1976.  Changes in Total nitrogen,  Lipoproteins  and
   Amino Acids in Epididyatal and Ej aculated  Spermatozoa of Bulls  Treated Orally
   with Ethylene Dibromide.  G^ Reprod. Fert., 47(1): 73-76

Anderson, Elizabeth L.  1977.  Letter to Edwin L. Johnson.  Be:   Potential
   Oncogenicity of Ethylene Dibromide  (EDB) .  1 March 1976

Seeder, O. 1980. SPRD/EPA File memorandun. Be: Annual Meeting of  the
   Interpartfflental Committee on Radiation Preservation of Food (ICRPF) .  26
   February 1980.

Berck, Ben.  1975.  Fumigant Residues of Carbon Tetrachloride, Ethylene
   01 chloride, and Ethylene Dibromide  in Wheat, Flour, Bran,  Middlings,  and
   Bread.   J. Agr. Food. Chen., 22(6):977-984

Berenblun.  1974.  Careinogenesis as a Biological Problem.  North Holland
   Publishing Company, Aasterdam.  Series:   Frontiers of  Biology, BC268.5.B47

Bontoyan Warren. 1980  Memorandun to Gary Otakie, Analysis of Selected  Flour
   Samples for Grain Fualgants, April  3, 1980

Boyersmith, J.  1977.  Oral communication Record on fumigation of cereal
   milling machinery with «thylane dibromide.  Conversation with  Boger
   Hoi tor f, Criteria and Evaluation Division, EPA.  Representing  Centennial
   Mills, Portland, Oregon, dated 29 August  1977

Brem, H., A.fl. Stein, and H.S. Rosenkranz.   1974.  The Mutagenicity  and  DMA-
   Modifying Effect of Baloalkanes.  Cancer  Res., 34:2576-2579

Buselmaier, »., G. Rohrborn, and P. Propping.  1972.  Mutagenicity Studies  with
   Pesticides in the Host-Mediated Assay and the Dominant-Isthai  Test in  the
   Mouse (in German) Biol. Zbl., 91:311-325

Buesel, J. and S.S. Xanburov.  1976.   Ethylene Oibromide  Fwigation of Citrus
   Fruit to Control the Mediterranean  Fruit  Fly, Caratitis capitata  (Wied.).
   J. Amer. Soc. Hort. Sci., 101(1):11-14

Butler, William A.  1975.  Petition to the Honorable Russell  Train,
   Administrator-SPA (7-14-75).  RE:   EDB as a Grain Fomigator.   Environmental
   Defense Fund

-------
                                      253

Carcinogen Assessment Group,  (GAG) .   1980a.   Risk Estimate Utilizing the
   Revised Dietary Exposure to Ethylene  Dibromide (EDS).   Memorandun froa Roy
   E. Albert, M.D., CAG, to Peter  McGrath,  Ph.D., HED.   28 March 1980

CAG.  198Qb.  Review of Inhalation Study of Ethylene Dibromide (EDB) on Rats.
   Memorandum to Christine F. Chaisson from Robert McGaughy,  CAG, dated 30
   April 1980,  1pg.

CAG.  1980c.  Risk Estimate Utilizing the  Revised Non-Dietary Exposure to
   Ethylene Dibroaida (EDB).  Memorandum to  Peter McGrath,  Ph.D., from Roy E.
    Albert, M.D., CAG. 16 May 1980

CAG.  1979.  The Carcinogen Assessment Group1 s  Updated  Quantitative Risk
   Assessaent on Ethylene Dibromide  (EDB).   Report dated  22 April 1979,  12
   pp. U.S. Environmental Protection Agency,  Carginogen Assessment Group

CAG.  1978.  The Carcinogen Assessment Group's  Risk Assessment for Ethylene
   Dibromide (EDB).  Report dated  "21 April  1978,  11  pp

CAG.  1977.  The Carcinogen Assesanent Preliminary Report on  Ethylene
   Dibromide (EDB).  Report dated  2 December  1977,  26 pp

CAG.  1977.  Preliminary Risk Assessaent for  Ethylene Dibromide.   Memorandum
   to Edwin L. Johnson dated  30 September  1977,  2 pp.   Updates memorandum  of 7
   September 1977

CAG.  1977.  Potential Cncogenicity of Ethylene Dibromide (SDB) .   Memorandun to
   Edwin L. Johnson dated 26  August  1977,  1 pp

CAG*  1977.  Preliminary Risk Assessment for  Ethylene Dibromide.   Memorandun
   plus appendices to Edwin L. Johnson,  DAA,  from Roy E.  Albert,  CAG,  dated  7
   September 1977, 2 pp

Chaisson, Christine.  1978.   Sthylene Dibromide;   Review  of Rebuttal.
   Section 3—-Mutagenicity.   Toxicology  Branch,  Hazard  Evaluation  Division,
   U.S. Environmental Protection Agency.  3 August:  1973,  18 pp

Clive, D.  1973.  Recent Developments with the  LS178Y7X  Heterozygote  Mutagen
   Assay System.  Snvir. Hlth. Perspect., 6:119-125

Da Serres, F.J., and H.V. Mailing.  1970.  Genetic Analysis of ad-3 Mutants of
   Neurospora crassa Induced  by Ethylene Dibromide - a  Commonly Csed
   Pesticide.  Newsl. Snvir.  Mut.  Soc.,  3:36-37

Developnent Planning and Research  Associates, Inc.   1978.   Economic Impacts,
   Cancellation of Registration of Ethylene Dibromide for Use  in  Fumigation of
   Gcain Storage.  Study for  U.S.  Environmental Protection  Agency, approx.
   100 pp

Sd wards, K., H. Jackson, and  A.R.  Jones.  1970.   Studies  with  Alkylating
   Estars-II.  A Chemical Interpretation Through  Metabolic  Studies of the
   Antifertility Effects of Ethylene Dimethanesulfbnate and Ethylene
   Dibromide.  Biochem Pharmacol.  19:1783-1789

-------
                                       254

Ehrenberg, L., S. Ostasaan-Golkar,  0.  Singh,  annd U.  Lundquist.  1974.
   On the Reaction Kinetics and Mutagenic  Activity of Methylating and [beta] -
   Halogenoethylating Gasoline Additives.   Had. Bot.,  15(3):185-194

Epstein, S.S., E. Arnold, J. Andrea, W. Bass,  and Y.  Bishop.   1972.
   Detection of Chemical mutagens by the dominant lethal assay in the mouse.
  Toxicol. Appl. Pharaacol. 23:288-325

Equitable Environmental Health, Inc.   1977.   A Retrospective  Study of Inferred
   Fertility of Married Men Occupationally Exposed to  Ethylene Oibrcmide
   (EOB).  22 November 1977.  Submitted to Ethyl  corporation.

Fanrig,  R.  1974.  Comparative mutagenicity Studies with Pesticides.
   International Agency for Research on Cancer, Sci.  Publi., No.  10:161-181

Federal  Register.  1977.  SUB Rebuttable Presumption  Against  Registration  Wed.
   Dec.  14, 1977  V.42 No. 240 p.63134-63161

Federal  Register.  1978.  Rules and Regulations:   Pesticide Use Restrictions.
   Thu., Feb. 9, 1978.  p. 5788-5791

Federal  Register.  1979a.  Rules and Regulations. Wed.  Aug. 1,  1979.   V. 44
   No.  149. p. 45131-45133

Federal  Register.  1979b.  Proposed Rules  Wed. Aug.  1,  1979, V.  44 No.  149. p.
   45218-46304

Ferguson Fumigants, Inc.  1978.  Letter to Anthony Inglis, OSPR, discussing
   Litton Sionetica, Inc. Project #20927 (athylene dibrcmide residues  in wheat
   and wheat products).  Dated 22 September  1978,  2pp.  plus attachments.
   FR »22S (30000/25)

Ferguson Fumigants, Inc.  1978.  Letter to Federal Register Section enclosing
   correction to page 12 of their Rebuttal dated  3 April 1978.   6 April  1978,
   1 pp. plus enclosure.  FR #22C (30000/25)

Ferguson Fuaigants, Inc.  1978.  Letter to  Federal  Register in  rebuttal of
   Ethylene Dibromide REAR.  3 April 1978.   1 pp.  plus  attachments.  FR  #22B
   (30000/25)

Farguson Fumigants, Inc.  1978.  Letter to Anthony Inglis, OSPR, enclosing
   preliminary Litton Bionetics, Inc. study on EDB residues in  funigatad flour
   milling machinery.  Dated 24 January 1973,  1 pp. plus enclosure.  FR  #22A
   (30000/25)

Ebod and Drug Administration, OHEW.  197S.  Compliance  Program  Evaluation.
   FY 78.  QLbromochloropropane (DBCP)  in  Raw Agricultural Products.  Accepted
   7-25-78 by William Randolph

FredricJcson, A.S. 1979 Determination of EDB in Crops, Soil Water, Bark, and
   Leaves.  2 January 1979.

Generoso, W.M.  1978.  Personal communication to  Dr. W.R. Lee;  data cited in
   Lee,  1978

-------
                                      255

Going jet al.  1976.  Sampling and Analysis  of Selected Toxic Substances.
   Task IV:  Sthylene Dibromide.  Office of "toxic Substances, 0.3.
   Environmental Protection Agency.   EPA-560-6-76-021,  July 1976

Griffith, J., Heath, R. and Davido, F.  1973.   Spermatogenesis in
   Agricultural Workers Potentially Exposed to Ethylene Dibromide (BOB) .
   An Interim Report by the Epidemiologic Studies Program,  Hunan Effects
   Monitoring Branch, Technical Services Division,  OPP,  OTS, SPA.  Dated
   June 8, 1978

Hazel ton Laboratories.  1978.  Technical Background Information Report on
   Car cino genesis Bio assay of 1,2-Dibromoethane (EDB).   OS  - DHEW, NIH, NCI.
   11-14-78

Hinkla, Maureen K.  1977.  Latter to  the Honorable  Douglas  M.  Cbstle,
   Administrator-EPA (8-26-77).  Re:  EOF Petition  for  Suspension and
   Cancellation of  Registrations of Ethylene  Dibromida  (EDB) Environmental
   Defense Fund
Uinkle, Maureen K.   1976.  Letter  to  the  Honorable  Russell  E.  Train,
   Administrator-EPA (9-30-76).  Re:   Ethylene  Dibromide (EDB)  as  a Grain
   Fumigant.  Environmental Defense Fund

Hinkle, Maureen K.   1976.  Letter  to  the  Honorable  Russell  E.  Train,
   Administrator-EPA (1-21-76).  Re:   July 14,  1975 EOF  Petition Requesting
   Suspension of Ethylene Dibromide (EDB)  as a  Grain  Funigant.   Environmental
   Defense Fund

Holder, J.W., Ph.D.  (RGB), 1979.   Memo to  Murcia Williams (SPRO),  Concerning
   update of EDB residue levels in commodities  as a result of  SOB  fuaigation.
   4 December 1979.

Holder, J.W., Ph.D.,  1980.  Re-evaluation  of the "total Dietary Burden of
   Ethylene Dlbromide Used as a Soil,  Grain, Grain  Machinery,  and  Fruit
   Fumigant.  14 August  1980.

Holladay, L.E.  1977.  General Health Survey of Personnel Warking  with
   Ethylene Dibronide.  Great Lakes Chemical Corporation.  October  19, 1977

Holtorf, Roger C»   1980.  Memorandum  to Richard Johnson  (SPRO)  concerning
   EDB Use Pattern  Estimates.  Benefits and Field Studies Division, SPA.   12
   August 1980

Interdepartmental Committee on Radiation  Preservation of Food.  1978.  Food
   Irradiation is the United States.   December  1978

Industrial Health Engineering Associates ,  Inc.  1978.  Fumigation Air
   Monitoring for Ethylene Oibrcmide.  By  Gerhard W.  Xhutson.   IHEA Project,
   29 March 1978

International Trade  Obmmission, 1978.  Synthetic Organic Ciemicals:
   United States Production and Sales,  1978.  USXTC Publication  1001
     418-574 0-83-18

-------
                                      256
Jagielski, J., K.s. Scudaoore, and S.G.  Heuser.   1978.   Residues of  Carbon
   Tetrachloride and 1,2-Oibromo«thane in Cereals and Processed Stood3 after
   Liquid Fvmigant Grain Treatment for Pest Control.  Pesticide Science 9:
   117-126

Kale, P., and J.W. Baum.  1978.  Environmental mutagen  Studies Utilizing a
   Drosophila  Test System.  Unpublished report of work supported by EPA
   Research Grant No. SPA-IAG-07-01207 and DOS Contract EY-76-C-02-0016:  to
   b« presented at the Tenth Annual Meeting, Environmental  Mutagen Society,
   March 8-12, 1979, New Orleans, LA

Kenichi, U.  1978.  Letter to Geraldine  Werdig, OSPR.   Rebuttal Submission  $80
   (30000/25).  Japanese study on ethylene dibrcmide (SOB).  Ministry of
   Agriculture and Forestry, Plant Protection Division. Transmitted by
   Embassy of Japan, Washington, O.C., 4/12/78.   1 p. plus  enclosure

Kr 1st offer son, U.  1974.  Genetic Effects of Same Gasoline  Additives.
   Hereditas 78:319 (abstract).

Lee, W.R.  1978.  Assessment of Risk  from Studies on the Mutagenicity of
   Ethylene Dibrcmide (EDB).  Consultant's draft  statement  submitted to  EPA,
   December 29

Litton Bionetics, Inc.  1978.  Letter to Ferguson Fumigants, Inc., enclosing
   second study on ethylene dibrcmide (LSI Project No.  20927), by Harold J.
   Paulin.  11 May 1978, 1 pp. plus enclosures.   FR #22D (30000/25)

Litton Bionetics, Inc.  1978*  Determination, of Residues of Ethylene Dibromide
   in Wheat Products Final Report.  Litton Bionetics Project Number  20927,
   May 1978.  10 pp

Litton Bionetics, Inc.  1977.  Determination of Residues of Ethylene Dibromide
   in Bread and Flour Final Report.   Litton Bionetics Project Number 20980,
   4 November 1977.  9 pp. FR #22A (30000/25)

Luttner, M.A.  1980.  Revised EDB Data:  Citrus.  Memorandum from Mark A.
   Luttner, EAB, to Linda Zygadlo, EAB.  30 July  1980

Ma, T.H., A.H. Sparrow, L.A. Schairer, and A.F. Naunan.  1978.  Effect of
   1,2-dibromoethane (DBS) on Meiotic Chromosomes of Tradescantia.
   Mot. Rea. 58:251-258

Maddy, K.T., S.  Uash, A.S. Fredrickson,  S. Edmiston.  1979a.  Analysis of
   Crops treated in California During 1978 for Residues of  Ethylene
   Dibromide .California Department of Food and Agriculture.  HS-41-576.
   22 January 1979

Maddy, K.T., W. Cusick, A.S. Fredrickson.  1979b.  Analysis for Residues
   of Ethylene Dibrcmide (EDB) In Selected Fruit  Which  Had  Been Funigatad
   Prior to Entry into California.  California Department of Food and
   Agriculture.  HS-628; 10 June 1979

Maddy, K.T. _et _al.  1979c.  A Summary of Studies  in California During 1973  and
   1979 concerning the Potential Hazard  to Applicators  and  Other Persons
   Exposed During Field Applicators of Ethylene Dibromide.  HS-593.  15 November
   1979

-------
                                       257

Mailing, H.V.   1969.  Bthylena  Dibroaide:   A Potent Pesticide with High
   Mutagenic Activity.  Genetics  61:539.  (abstract)

Marshall, John W., July  18,  1980.   Memorandum to  Jeff Kempter

Martin, H. 1971.  Pesticide  Manual, British Crop  Protection Couacil

Mauer, Irving.   1979.  Preliminary  Qualitative Assessment of the Genetic Risk of
   Ethylene Dibronxide (EDB)  to  Humans.  Toxicology Branch, Hazard Evaluation
   Division, U.S. Environaental Protection  Agency.  15 February 1979,  18 pp

McCann, J., S. Choi, E. Yamasaki, and 3.N.  Ames.   1975.   Detection of
   Carcinogens as Mutagens in the Salnynella Microsome Test:   Assay of 300
   Chemicals.  Proc. Nat. Acad. Sci., USA 72(12):5l35-5139

Meneghini, R.   1974.  Repair Replication of Opoasun  Lymphocyte DMA:   Effect of
   Compounds that Bind to DMA. Chem.-Biol.  Interactions 8: 113-126

Merck Index.  1976.  9th ed.; M. Windholz ed.,  Rahway,  NJ

Mishra, Lakshmi  C., Ph.D.  1978a.   Ethylene Dibromide:   Review of  Rebuttal.
   Section 1-Oncogenicity, 8-24-78.  Hazard Evaluation  Division, EPA

Mishra, Lakshmi  C., Ph.D.  1978b.   Memo to  Tony Inglis,  Special Pesticide
   Review Division.  Re:  Hunan Risk Assessment of  EDB  Alternatives  ( including
   Toxicity Assessment of EDB Alternates, with  tables).   3 Oct.  1978

Mitre Corporation.  1978.  Control of Fugitive  Ethylene  Dibromide  Emissions
   from Citrus Fumigation Facilities.  Mitre Corporation Contract  468-02-2526,
   Task No. 20 for the U.S.  Environaental Protection  Agency.   5  July 1978,
   47 pp

Mittelman, Abraham.  1980.   Non-dietary Exposure  Analysis for EDB  Uses:
   Environmental Fate of EDB.  Dated 20 March  1980.   Environaental Fate
   Branch, Hazard Evaluation Division, U.S.  Environmental Protection Agency

Mann, G. J.  1971.  Microorganisms as Test Systems for Mutagenicity.  Arch.
   Toxicol.  28:93-104 (as cited in Buselmaier  £t al.,  1972)

Monro, H.  1969.  Manual of  Fuaigation for  Insect Control,  2nd ed.
   FAO/WHO, Rome

Nachtomi, S.  1970.  The Metabolism of Ethylene Dibromide In  the Rat.  The
   Enzymatic Reaction with Glutathione in vitro and  in  vivo.
   Biochem.Pharaacol.  19:2853-2860

National Cancer  Institute (NCI).  1974.  Memorandum of  Alert  on  EDB.
   16 October 1974

      1980.  aioassay of 1, 2-dibromoethane  for  Possible  Car ci no gen i city.
   U.S. Dept of  Health and Human Services,  Public Health Service, NIH, DHHS
   Publication No. 80-1766

-------
                                      258

Nauaan, A.P., A.a. Sparrow,  and  L.A.  Schairer..   1976.   Comparative Effects of
   Ion zing Radiation and Two Gaseous  Chemical Mutagena on Somatic Mutation
   Induction In One Mutable  and  Two Non-mutable  Clones of Tradescantia. Mut.
   Res.  38(1): 53-70

NIOSH.  1977.  Criteria for  a  Recommended Standard:  Occupational Exposure to
   Ethylene Oibronide.  No.  77-221, National  Institute for Occupational Safety
   and Health, MS, U.S.,  DHEW.   208  pp

Nitschke, R.J., et al.  1980.  13-Week Repeated  Inhalation Study on Ethylene
   Dibrcmide (EDB) in Male and Female Rats.   Qow Chemical OSA.  11 February
   1980

Olson, W.A., R.T. Habermann, E.X. Weisburger, J.M. Ward/  and J.H. We'isburger.
   1973.  Induction of Stoaach Cancer in  Rats and Mica by Halogenated
   Aliphatic Fumigants.  J. Nat. Cancer  Inst.  51(6):1993-1995

Ott, M., H.C. Schamweber, R.R.  Langner.   1977.   Prepublication  Draft:   The
    Mortality Experience of  161  Employees  Exposed to Ethylene Dibromide in Two
    Production Units-The Cow Chemical Company.   24 March  1977, 10 pp

Panetta, J. 1980. SPRD/ESA File  memorandun. Re:  Approval  of gamma irradiation
   of citrus. 28 February  1980.

Panetta, J.  1979.  Memo for the EDB  file:  Treatment-to-market  Interval for
   EDB- treated Citrus.  SPRD, EPA.   24  December 1979

Plotnick, a.  1980.  Abstract-Toxicity of  Inhaled 1,2-Qibromoethane in  Rats
   With and Without Dietary  Disulfiram.  NIOSH Contract No.  210-76-0131

Powers, M.B., R.W. Vbelkar, N.P.  Page, E.X. Weisburger and H.F.  Kraybill.
   1975.  Carcinogenicity of Ethylene Dibromide  (EDB)  and 1,2-Dibrcmo-3-
   Chloro pro pane (DBCP) After Oral Administration In Rats and Mice.   Toxicol.
   App. Pharmacol.  33( 1 ): 171-172. (Abstract)

Scrtnitt, D.  1978.  Update of Food Factor  Tables,  lexicology Branch, HEP,  OPP,
   EPA

Scott, B.R., A.H. Sparrow, S.S.  Schwemmer, and L.A. Schairer.  1978.  Plant
   Metabolic Activation of 1,2-dibromoethane  (EDB) to  a Mutagen of  Greater
   Potency.  Mut. Res. 49:203-212

Short Jr.,  R.D.,  J.L. Minor, B. Ferguson, T. Unger, and C.C.  Lee.   1976.
   Tbxicity Studies of Selected Chemicals, Task  I:  The Developnental Ibxicity
   of Ethylene Dibromide Inhaled by Rats and Mice During  Organogenesis•  Final
   Report,  EPA/-560/6-76-018

Sidhu, Jlvan S.,  et al.  1979.  A study of 1,2-Qibromoethane  Residues in
   Wheat and Milled Products.  Pesticide Science  6:451-455

Sivinski,  J. 1980. Personal Communication to J.   Panetta,  SPRD/EPA.  Re: Gamma
   irradiation. August 1980.

-------
                                      259
 Sparrow, A.H.,  and L.A.  Schairer.  1974a.  The Effects of Chemical Matagens
    (EMS, CBS)  and Specific Air pollutants <03'S02/ NO , NO) on
    Somatic  Mutation Rates in Tradescantia .  Tal£ presented at the Symposium on
    The  Potential Genetic  Effects of  Environnental Pollutants on Man.  Moscow,
    USSR,  18-21  Febr-uary  1974
Sparrow, A.H., L.A.  Schairer,  and R.  Villalobos-Pietrini.  1974b.  Comparison
   of  Somatic  Mutation  Rates  Induced  in Tradescantia by Chemical and Physical
   Mutagens.   Mut. Res.  25(45:265-276

Spector, W.S.  1956.  Handbook of Biological Data.  W.B. Saunder, Co., Phila .
   1958. p. 584

Spencer, 3.Y.  1973.  Guide to the Chemicals Osed in Crop Protection. 6th ed.
   Agricvilture Canada.  1973

Tanada, et al.   1953.   "Bromide Residues:   Determination in fresh Fruits After
   Fumigation With Ethylene Dibromide."  J.  Agr.  Food Chem.  1(6): 453-455

Teramoto, S.t  R.  Saito ,  H. Aoyama,  Y,  Shirasu.   1980.   Dominant Lethal Mutation
   Induced in  Male Bats  By  1, 2-dibromi-3-chloro pro pane (DBCP).   Muta-jen
   Research, 77  (1980)  71-78

Ter Haar, G. 1978.  Comments on:   EPA's Rsbuttal  Preswnption Against
   Registration and Continued  Registration of Pesticide Products Containing
   EDB.  Ethyl Corporation.   23 January  1978.  Rebuttal Docunent
   if 48 (30000/25).

Thompson, W.T.  1977.  Agricultural Chemicals.  Thomson Pubs.,  Fresno,  CA.

Turner, D.  1977.  A Mortality Survey on Employees  at  an Ethylene Dibroaide
   Plant.  The Associated Octal Company Limited.   (See Ter  Haar, G. ,  1973)

Turner, D.  1976.  Appendix II  (see above).  Dibromoethane-A  Survey of
   Amlwch Records (See Ter Haar,  G. 1978)

0SDA/ STATE/EPA.   1973.  The Biologic and Economic  Aasessnent  of  the Pesticide
   Ethylene Dihromide.  USDA/ State/ E2A  EDB Assessment  Team.   June 1980

U.S.  Enviro mental Protection  Agency and U.S. Department of Agriculture.
   1979.  Abbreviated Benefit  Analysis of PesticLdal (toes of  Ethylene
   Dibrcmide.  Economic Analysis  Branch < Benefit and Field  Studies Division,
   Office of Pasticide Programs.  January, 1979,  132 pp

Vaccaro, J.R.  1976*  Exposures of Fun ig ant  Applicators  to  Ethylene Dibromida
   During Field Application in  the San Joaquin  Valley, California, May,  1976.
   Industrial Hygiene Report ,   Industrial Hygiene laboratory,  Dow  Chemical
   U.S.A.  21 July 1976, 8 pp

Vogel, E.,  and J.L.R. Chandler.   1974*  Mutagenicity Testing  of Cyclamate and
   Some Pesticides in Drosophila  Melanocfaster.  Experientia 30(6 ): 321-323

Ward,  J.M. ,  Habennann,  R.T.  1974.  Pathology of Stomach Cancer in Hats and
   Mice Induced with the Agricultural Chemicals Ethylene Dibromida and
   Oibromochloro pro pane.  DBCP- induced cancer.  Bull. Soc. Pharmacol.
   Environ.  Pathol.  74(2): 10-11

-------
                                       260

Weisburger, Elizabeth K.   1977.   Carcinogen!city studies on Halogenated
   Hydrocarbons.  Presented  at Cbnference  on Comparative Metabolism and
   Tbxicity of Vinyl-chloride-related  Compounds,  May 2,  1977,  24 pp.

Welt. M.A.  1980a.  Letter  to 0.  Seeder.  Re:  Cast Evaluation for use of
   radiation for citrus. 23  July 1980

Welt, M.A.  198Ob.  Letter  to 0.  Seeder.  Re:  Cost Evaluation for use of
   radiation for citrus. 4 August 1930

White, J.V., and 0. McAllister.   1977. Soil  Fumigant Applicator Expos-ore
   to Ethylene Dibrcmide.  Unpublished manuscript,  Great Lakes Chemical
   Company, 18 October  1977, 7 pp

White, L.V.  1977.  Letter to Suzanne  Harker,  OSPR,  summarizing results of Caw
   Chemical Company study  on spent  levels  of workers exposed to EDB.  Great
   lakes Company, 9 October  1977,  3 pp

Whitesides, J.B. 1980.  Letter to J. Panetta.  Re Cost data for citrus
   fumigation center modifications* 16  July 1980.
Whorton, Donald, et al.  1977.   Infertility  in Male  Pesticide Workers.  The
   Lancet,  17 December  1977

Williams, M. 1980a. Memorandun to Edwin  L. Johnson.  Re:  Request to contact FDA
   regarding- the use of gaoma radiation.  15  January  1980

Williams, M. 198Ob. Memorandum to Edwin  L. Johnson.  Re:  FDA requirements for
   approval of gamoa radiation for  use on  citrus.  28 Fedruary 1980

Wit, S.L.,  A.F.H. Besemer, a .A.  Das, W.  Goedkoop,  F.E. Loosjes, and E.K.
   Mepellink.  1969.  Results of an Investigation  on the Regression of  Three
   Fumigants (Carbon Tetrachloride, Sthylene Dibromide,  and Ethylene
   Dichloride) in Wheat During Processing  to' Bread.   Report from the Director
   General  of the National Institute of  Public Health to the Inspector  General
   of Public Health in  charge of Foodstuff Division, Bilthoven, Netherlands.
   April 1969, 21 pp

Wit, S.L.,  A.F.U. Besemer, H.A.  Das, W.  Goedkoop,  F.E. Loosjes and E.K.
   Mepellink.  1977.  Determination of Residues of  Ethylene Dibromide in  Bread
   and Flour, Final Report.  Litton Bionetics, Inc.  Project No. 20880

Wit, S.L.,  A.F.U. Besemer/ H.A.  Das, W.  Goedkoop,  F.E. Loosjes and E.K.
   Mepelink.  1978*  Determination of  Residues of Ethylene Dibromide in Wheat
   and Wheat Products,  Final Report.   Litton Bionetics,  Inc.  Project No.  20927

Wong} L.C.K., et al.   1979.  Study of  Carcinogenicity and Tbxicity of Inhaled
    1,2-QLbromomethane in Rats Treated  with Disulfiram.  Midwest Research
   Institute Final Report, NIOSH contract  No.  210-76-0131.  8 February  1979.
    100 pp

Wong, 0.  (1978).  A Retrospective Study  of Inferred Fertility of Married Man
   Occupationally Exposed  to Sthylene  Dibromide (EDB).  Report from Equitable
   Environmental Health, Inc., Rockville,  Maryland to Ethyl Corporation,  Baton
   Rouge.   In:  Rebuttal Docunent #48(30000/25}

-------
                                      261

Worthington, John.  1979.  Memorandum to Anthony  Inglis, SPRD, commenting on
   California food crop a thy lane dibromide residues.  Residue Chemistry  Branch,
   Hazard Evaluation Division, U.S. Environmental Protection Agency.   19 July
   1979, 2 pp

Worthington, John.  1978.  Final Dietary Exposure Analysis for Ethylene
   Dibromide Resulting From Residues in Funigated Crops and Ebod Commodities.
   Residue Chemistry Branch, Hazard Evaluation Division, 29 December 1978. EPA

Zygadlo, Linda.  1978.  Memorandum to John Worthington, HED, giving percentages
   of crops treated with ethylane dibromide.  Economics Analysis Branch,
   Benefits and Field Studies Division, U.S. Environnental Protection Agency.
   7 December 1978, 2 pp

-------
                                     262
                                  APPENDIX  1
                 RESPONSES RECEIVED REBUTTING HEALTH  TRIGGERS

Rebuttal No.                 Source                   Data of  Submission -
3C
7
13
14
15
1SA
1SB
16
22
22A
223
22C
22D
26
31
32
33
47
47A
48
64
Great Lakes Chemical Corporation
Chevron Chemical Company
University of Idaho Cooperative
Extension Service
Department of Citrus, State of Florida
Millers National Federation
Millers National Federation
American Corn Millers Federation
Colorado State Uhiversity
Ferguson Fumigant Inc.
Ferguson Fuaigant Inc.
Ferguson Fumigant Inc.
Ferguson Fvnigant Inc.
Ferguson Fumigant Inc.
Pineapple Growers Association of
Hawaii, Honolulu
Florida Agricultural Research Institute
Florida Farm Bureau
Department of Agriculture
Honolulu, Hawaii
Douglas Chemical Company
Douglas Chemical Company
Ethyl Corporation
Vulcan Materials Company,
01/23/78
01/16/78
01/19/78
01/20/78
01/24/78
04/07/78
04/07/78
01/26/78
01/31/78
02/01/78
04/04/78
04/13/78
06/01/78
12/12/77
01/31/78
01/31/78
01/31/78
02/09/78
04/13/78
01/31/78
04/03/78
                                                                         2/
                   Chemical Division
65                 Florida Citrus Commission               04/03/78

-------
                                      263

56                 Environmental Defense Fnnd              04/04/78-04/03/78

73                 Dow Chemical                            04/26/78

79                 Cooperative Extension, University of    04/1 1/78
                   California

J/  Dates given are the dates OFP's Federal Register Section received the
      submission.

2/  Three additional addenda were submitted on this rebuttal.

-------
                                 APPENDIX 2
                       WOM FOOD IRRADIATION NEWSLETTER
                       OCTOBER 79, IAEA

                            LIST OF CLEARANCES
GENERAL

COUNTRY
(Organisation)
BULGARIA






SURVEY Or IRRADIATED rOOD PRODUCTS
(Grouped according to country)
PRODUCT
pota,toea *)
potato** *)
onlona *)
garlic *)
grain »)
dry food concentrate* •)
dried fruit* *)
CLEATBD rOR HUHAN CONSUMPTION IN DirrBRENT COUNTRIES
Auguat 1979
PURPOSE Or TYPE AND SOURCE DOSE
IRRADIATION Or RADIATION (k Gy)
aprout Inhibition +
aprout Inhibition + O.I
aprout Inhibition + 0.1
aprout Inhibition + 0.1
aprout Inhibition + 0.3
Inaact dlalnfeatatlon -1- 1
Insect dlalnfeatatlon -t 1

DATE Or A

30 April
30 April
30 April
30 April
30 April
30 April

PPROVA1
1971
1972
1972
1972
1972
1972
1972
freah fruit*  ' (toaatoea,
peachea, apricot, cherry,
raapberry, grapea)
                                                                                                                    to
radurlcatton
                                                2.5
                                                                 30 April     1972

-------
COUNTRY PRODUCT
(Organization)
CANADA potatoes

onions
wheat, flour, whoje wheat
flour
MM)
poultry
cod & haddock fillets
l\ IIMl
CHILE potatoes '
CZECHOSLOVAKIA potatoes "'
»\
onions
• »
Mushrooms
PURPOSE Of TYPE AND SOURCE DOSE DATE OF APPROVAL
IRRADIATION OF RADIATION (k Gy)
60p 137- elec-
L0 CS trons
sprout Inhibition *

sprout Inhibition *
Insect dlslnfestatlon *
radlcldatlon
(Sal«onella) *
radurlzatlon +
sprout Inhibition +
sprout Inhibition +
sprout Inhibition +
growth Inhibition t
O.I wax 9 Novonber
0.15 nax H June
0.15 Max 25 March
0.75 «ax 25 February
7 max 20 June
1.5 Max 2 October
31 October
O.I Max 26 November
0.06 max 26 November
2 Max 26 November
I960
1963
1965
1969
1973
1973
1974
1976
1976
1976
                                                                                                                                                      fj\
DENMARK
potatoes
                                                        sprout
10 HaV
                                                                                         0.15 max
27 February  1970

-------
COUNTRY PRODUCT
(Organization)
FRANCE potatoes <">)
• it
onions
••I
garlic '
••»
shallot '
•| •••!
FRG deep-frozen Mais '
• t
potatoes
• •••i
HUNGARY potatoes '
potatoes
•••ii
potatoes
• •••I
onions
• ••it
onions
onions
strawberries
PURPOSE OF
IRRADIATION
sprout Inhibition
sprout Inhibition
sprout Inhibition
sprout Inhibition
radappert 1 zat Ion
sprout Inhibition
sprout Inhibition
sprout Inhibition
sprout Inhibition
sprout Inhibition
sprout Inhibition
sprout Inhibition
radurlzatlon
TYPE AND SOURCE DOSE
OF RADIATION (k Cy)
60co I37CS elac-
* 0.075 - 0.13
+ * 0.075 - O.J5
* * 0.075 - 0.15
* + 0.075 - 0.13
+ 25 -45
t O.IS MX
* O.I
* O.|5 MX
+ 0.15 MX
t
+ 0.06
* 0.06
*
DATE OF APPROVAL
a November 1972
9 Au
-------
COUNTRY PRODUCT
(Organization)
»\
IUNGAHY mixed spices (blackpepper,
(cont'd.) cumin, paprika, dried garllci
for use In sausages)
nixed dry Ingredients for
canned hashed maat ')
(wheat flour. Na caselnate.
onions and garlic powder,
paprika)
ISRAEL potatoes
onions
ITALY potatoes
onions
garlic
PURPOSE OF TYPE AND SOURCE DOSE DATE Of APPROVAL
IRRADIATION OF RADIATION (k Gy)
M- '"<» S=
radlctdatlon •»
radiation
decontamination t
sprout Inhibition *
sprout Inhibition 4
sprout Inhibition t
sprout Inhibition t
sprout Inhibition +
5 2 Apr II
5 20 November
0.15 max 5 July
O.I MX 25 July
0.075 - 0.15 DUX 30 August
0.075 - O.IS 30 August
0.075 - 0.15 30 August
1974
1976
1967
I9btt
1973
1973
1973

1





JAPAN
                        potatoes
sprout Inhibition
                                                                                                                Q.I5 max
                                                                          30 August     1972

-------
COUNTRY PRODUCT
(Organization)
it
NETHERLANDS asparagus
cocoabeans
• •«*)
strawberries
jnushrooas
nit
deep-frozen reals
potatoes
peeled potatoes
shrimps
• nit
shrimps
i)
onions
onions
PURPOSE OF TYPE AND SOURCE DOSE
IRRADIATION OF RADIATION (k Gy)
60 137 elac-
00 Cs trons
radurlzatlon *
Insect dlslnf estatlon •»
radurlzatlon *
growth Inhibition *
radappertlzatlon +
sprout Inhibition *
rttdurlzat Ion *
radurlzatlon +
radurlzatlon +
sprout Inhibition +
sprout Inhibition t
2 max
4 HeV 0.7 max
4 HaV 2.5 max
4 HaV 2.5 max
25 nln
4 HeV 0.15 max
0.5
4 HeV 0.5 - 1
1
0.15 max
0.05 max
DATE OF APPROVAL
7 Hay
7 Hay
7 Hay
23 October
27 November
23 Harch
12 Hay
13 November
15 June
5 February
9 June
1969
1969
1969
1969
to
as
1969 00
1970
1976
1970
1976
1971
1975
poultry, eviscerated
  (In plastic bags)
raiiurlzatlon
                                                                                              max
                                                                          31  Uocomber   1971

-------
COUNTRY PRODUCT
(Organization)
NETHERLANDS chicken
(cont'd.)
fresh, tinned & liquid
iiti
foostuffc
spices & condiments
1*1
spices
spices
*•)
spices
vegetable f 1 1 ling
powdered bat term! x
• «i«i
endive
PURPOSE Of TYPE AND SOURCE DOSE
IRRADIATION OF RADIATION (k Cy)
60f 137 elec-
00 Cs trons
radurliatlon.
radlcldatlon +


radappertfzatlon +
radlcldatlon t
radlcldatlon +
radlcldatlon *
radlcldatlon t
radlcldatlon +
radlcldatlon +
radurlzatlon *

3 max


25 nln
4 HaV a - 10
4 HeV 10
3 MaV 10
10
0.75
1.5
1
DATt Of APPROVAL

10 Hoy


8 March
13 Suptomber
4 October
26 June
4 April
4 October
4 October
14 January

1976


1972
1971 |
1974
1975
I97U
1974
1974
1975
fresh vegetables
  (prepared, cut,
   soupgreons)
radurUatlon
                                                                                      1977

-------
COUNTRY PRODUCT
(Organization)
NETHERLANDS fl 1 lets of haddock, „„,.
(conl'd.) coal-fish, whiting '
• •«i\
fillets of cod 1 plaice '
**)
frozen frog legs
rice and ground rlca
• it
products
PHILIPPINES potatoes *"'
SOUTH AFRICA potatoes
onions
garlic
chicken
papaya
PURPOSE OF
IRRADIATION
radurlzatlon
radurlzatlon
radlcldatlon
dlslnfestatfon
sprout Inhibition
sprout Inhibition
delayed greening
sprout Inhibition
sprout Inhibition
radurlzatlon
radurlzatlon
TYPE AND SOURCE
OF RADIATION
+
+
+
+
t
+
*
+
t
+
DOSE
(kGy)
1
1
5
1
0.15 nax
0.12 - 0.24
0.05 - 0.15
O.I - 0.20
2-7
0.5 - 1.5
DATE OF APPROVAL
6 September 1976
7 September 1976
25 September 1976
tc
13 March 1979 o
13 September 1972
19 January 1977
25 August 1978
25 August I97U
25 Auyust I97U
25 August I97U

-------
COUNTRY
(Organization)
SOUTH AFRICA
(cont'd.)


SPAIN

THAILAND
UNION OF SOVIET
SOCIALIST REPUBLICS



PRODUCT
pango
strawberries
dried bananas
avocados
potatoes
onions
onions
potatoes
potatoes
grain
fresh fruits A vegetables '
PURPOSE Of TYPE AND SOURCE DOSE DATE OF APPROVAL
IRftttlATION OF RADIATION (k Gy)
radurlzatlon *
radurlzatloh *
Insect dlslnfestatlon +
delayed ripening i
sprout Inhibition *
sprout Inhibition *
sprout Inhibition *
sprout Inhibition +
sprout Inhibition
Insect dlslnfo&tatlon +
radurUatioo t
0.5-1.5 25 Auyust
1-4 25 Auyust
0.5 max 28 July
O.I MX 28 July
0.05 - 0.15 4 November
0.06 Max
O.I wax 20 March
O.I 14 March
1 HeV 0.3 17 July
0.3
2-4 j| July
1978
1970
1977
1977
1969
1971
,973
1958
1973
1959
1964

-------
COUNTRY PRODUCT PURPOSE OF TYPE AND SOURCE
(Organization) IRRADIATION OF RADIATION
60_ |37_ el
Co Cs .
UNION OF SOVIET semi-prepared raw beef,
SOCIALIST REPUBLICS pork & rabbit products
(cont'd.) (In plastic bags) ") radurlzatlon t
dried fruits Insect dlslnf estatlon t
dry food concentrates Insect dlslnf«statlon t
(buckwheat aush, gruel
rice pudding)
poultry, eviscerated radurlzatlon +
(In plastic bags) •)
culinary prepared naat radurlzatlon +
products (tried mat,
entrecote) ,.
(In plastic bogs)
onions sprout Inhibition •*•
onions sprout Inhibition t
DOSE OATH OF APPROVAL
(kGy)
ec-
ons
6 - (t II July 1964
1 15 February 1966
0.7 6 June (966
6 4 July 1966
fi 1 February I9b7
0.06 25 February 1967
0.06 (7 July 1973

-------
COUNTRY
(Organization)
PRODUCT
PuflPOSE Of
IRRADIATION
TYPE AND SOURCE
  OF RADIATION
                                                                                60,
                                                                                  Co  '""Cs
                                                                                              trons
                                                         DOSE
                                                        (k Gy)
DATE Of APPROVAL
UNITED KINGDOM
any food for consumption
by patients who require
a sterile diet as an
essential factor In their
treatment
rodappert Izat Ion
                                                   I December   1969
UNITED STATES
Of AMERICA
wheat and wheat flour
(changed on 4 March 1966
 from wheat and wheat
 product)

white potatoes
Insect dlslnfestatlon     +
                                                        sprout Inhibition
                                                                                                5 MaV
                                                  0.2  - 0.5
                                                  0.2  - 0.5
                                                  0.2 - 0.5
                                                  0.05 - O.I
                                                  0.05 - O.I
                                                  0.05 - 0.15
                                                                                                                                                        -3
                                                                                                                                                        CO
                                                 21 Au.just     1963
                                                  2 October    1964
                                                 26 February   1966
                                                 30 June       1964
                                                  2 October    1964
                                                  I November   1965
URUGUAY
potatoes
sprout Inhibition
                                                                                                                                  23 June
                                                                                                                        1970

-------
COUNTRY
(Organization)
WORLD HEALTH
ORGANIZATION
(FAO/IAEA/HM5
Export Conn in aa)







PRODUCT
potatoes
potatoes
onions
papaya
strawberries
wheat A ground wheat
products
wheat 4 ground wheat
products
rice
chjcken
cod & red Msh *'
PURPOSE OF TYPE AND SOURCE DOSE
IRRADIATION Of RADIATION (k Gy)
60Co I37CS ?
-------
                                   APPENDIX II

                    COMMENTS RECEIVED IN RESPONSE TO FD  2/3

Conment No.                  Source

1                            W.B. McCloud & Co.

2                            Millers National Federation

3                            County of San Diego
                             Department of Agriculture

4                            Florida State Beekeepers Assoc.

5                            State of Hawaii

6                            Anheuser-Busch, Inc.

7                            The American Beekeeping Federation

8                            American Corn Millers Federation

9                            Butts County Mosquito
                             Abatement District

10                           California Farm Bureau Federation

11                           LeBost, Mr. & Mrs, Deerfield

12                           Jackson County Farm Bureau

13                           Boettcher Supply, Inc.

13A                          Reiling, Mr. Gregg

14                           Chio Dept. of Agriculture

15                           Colorado State Forest Service
                             Colorado State University

ISA                          Colorado State Forest Service
                             Colorado State University

16                           University of Nebraska-Lincoln
                             Cooperative-extension Service

17                           State of North Carolina
                             Dept. of Agriculture

18                           Crawford, Mr. Johnny L.

ISA                          University of Georgia
                             College cf Agriculture,
                             Cooperative Extension Service

                                          275

-------
                             APPENDIX IH
        FEDERAL INSECTICIDE, FUNGICIDE, AND RODENTICIDE ACT  (FIFRA)

                         SCIENTIFIC ADVISORY PANEL
                      Review of Preliminary Notice of
                  Determination Concluding the Rebuttable
                  Presumption Against Registration (RPAR)
                      of Pesticide Products Containing
                          Ethylene Dibromide (EDB)
The Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) Scientific
Advisory Panel has completed review of plans by the Environmental Protection
Agency (EPA) for initiation of regulatory action on pesticide products
containing ethylene dibromide (EDB) under the provisions of Section 6(b)(2)
of FIFRA as amended.  The review was completed in an open meeting held in
Arlington, Virginia, on March 25-26, 1981.

Maximum public participation was encouraged for the review.  Public notice
of the meeting was published in the Federal Register on February 25, 1981.
In addition, telephone calls were received from and mailings sent to the
general public who had previously expressed an interest in activities of
the Panel.  Oral, and in most cases, written statements were received from
the technical staff of EPA's Office of Pesticide Programs, and from repre-
sentatives of the U.S. Department of Agriculture;  the State of Florida,
Department of Citrus; Ethyl Corporation; the University of California;
the Hawaii papaya industry; Dow Chemical Corporation;  National Pest Control
Association; Hopes Consulting Company; National Association of Wheat Growers;
Emergent Technologies, Incorporated; and Radiation Dynamics, Incorporated.

The excellent briefings by the staff of OPP's special Pesticide Review
Division (SPRD) were of great value to the Panel.   The Panel wishes to
express its appreciation to SFRD for having prepared a Position Document
2/3 on EDB of very high quality.  Also, the Panel would like to thank
Dr. Carol Sakai, Reproductive Effects Assessment Group, EPA, for an
outstanding scientific briefing on the reproductive effects of EDB.

In consideration of all matters brought out during the meeting and careful
review of all documents presented by the Agency and other parties, the
Panel unanimously submits the following report:

                                    277

-------
            REPORT OF SCIENTIFIC ADVISORY PANEL RECOMMENDATIONS

     The Agency requested the Panel to focus Its attention upon  these
issues:

     1.  The data requirements (ground water, food residues for  pre-plant,
         exposures and survey data for minor uses applicator)

     2.  The acceptability of risks during the phase ouc of citrus

     3.  The similarities of this pesticide to dibromochloropropane  (DBCP)

         o  use sites

         o  health risks

         o  chemical structure

     4.  The acceptability of alternatives relative to the toxic hazards
         of EDB expecially for grain and citrus.

     The Panel's response to each of the above issues is as follows:

     1.  The data requirements (ground water, food residues for pre-plant,
         exposures and survey data for minor uses applicator)

         The Panel expresses considerable concern over the potential human
hazard as a result .of the spot fumigation in flour mills use of Ethylene
Dibroeide (EDB) and recommends that EPA require a thorough and critical study
of EDB residues in grain, as veil as obtain residue data in bread and other
bakery products.

         The Panel also indicates its concern over the possible appearance
of EDB in ground water, and urges that ground water be monitored closely
in high use areas.

         The Panel recommends that a reproductive study on rodents be
performed with critical structural and functional analysis of endocrine,
reproductive and fetal tissues.

         Finally, the Panel recommends that EPA monitor high risk workers
in the grain, citrus and minor use areas, both during and following exposure.

                                     279

-------
                                    280
     2.  The acceptability of risks during the phase out of citrus

         The Panel differs with the Agency's position that EDB use on
citnifl should be phased out by July 1983.  Rather, the Panel proposes
that this use be retained in a similar manner as several other uses which
the Agency is proposing be allowed, but with certain restrictions imposed,
including pre-plant fumigation of soil, stored beehive fumigation, stored
furniture and clothing, and nursery stock..  The reason for this  is that  the
Panel finds it difficult to evaluate whether it is feasible for  the citrus
industry to move to irradiation as an alternative to EDB control of fruit
flies.   Therefore, the Panel recommends  reevaluation of the risks and
benefits of irradiation as an alternative for EDB as soon as possible.

         In the meantime, the Panel recommends the use of additional
protective measures to reduce worker exposure.  It appears to the Panel
that substantial improvements can be made in EDB application technology,
and that such improvements will reduce both the number of workers exposed
and the severity of the exposure.

     3.  The similarities of EDB to" DBCP with regard to -hdmical structure,
         use sites and health risks

         The Panel finds that EDB has certain structural qualities, uses
and adverse effects which resemble those of DBCP.  However, extrapolation
from one compound to another entails the same kind of uncertainties as
does extrapolation in testing from one species to another.  Therefore,
the Panel recommends that the data be obtained on EDB directly rather than
on the basis of extrapolation from other compounds.

     4.  The acceptability of alternatives relative to the toxic hazards
         of EDB especially for grain and citrus

         The Panel finds that alternatives to EDB appear to be available
in several areas, e.g., fumigation of stored grain, termite control, bark
beetle control and other sdnor uses.  Alternatives for soil fumigation and
citrus are not well developed and clearly demonstrated to be efficient,
practical, and feasible from a cost standpoint.
     The Panel's response to the issues listed by EPA as regards EDB is
based on the following evaluation of the available data base for this
agent:

     1.  EDB is a carcinogen for multiple animal species of both sexes
producing oncogenic effects in several tissues following exposure by at
least two routes of administration.

-------
                                     281
     2.  EDB is a potent mutagen -producing point (gene) mutation, chromosomal
aberration and primary DKA damage in multiple test systems involving both
prokaryotic and eukaryotic systems (including mammals).

     3.  EDB has been demonstrated to produce adverse reproductive effects
in several species.  None of these studies definitively demonstrates a
no observed effect level (NOEL).  As indicated above, the Panel recommends
that a reproduction study in rodents be required to characterize the threshold
dosage level for EDB effects on fertility, mating,  gestation, etc.

     The Panel notes that it will be very difficult to conduct epidemiolo-
gical studies that will enable EPA to ignore the results of animal studies.
Such epidenjiological studies which have been conducted thus far do not
provide convincing evidence that animal tests do not accurately predict
potential human hazards in the area of oncogenicity and reproductive
effects.  Therefore, it is necessary to regulate on the basis of animal
studies alone.

     The Panel wishes to make a clear distinction between the apparent
hazards of citrus and grain fumigation insofar as human dietary factors
are concerned, and thus does not concur with the Agency's proposal to
cancel the use of EDB on citrus.  As regards stored grain fumigation and
spot fumigation of grain milling machinery, the Panel wishes to express
its great concern over the possible presence of EDB residues in finished
bakery products.  The evidence is far from solid, but because of the extremely
large population potentially at risk, the problem demands resolution.
Therefore, the Panel concurs with the EPA proposal  to cancel stored grain
fumigation and spot fumigation of grain milling machinery uses until such
time as convincing evidence exists that such uses present little or no
hazard to consumers of bakery products.

FOR THE CHAIRMAN:

Certified as an accurate report of findings:
PMlip HGray,
Acting Executive Sretary
FIFRA Scientific Advisory Panel

Date:  April 22, 1981

-------
                                   282


                           DEPARTMENT CF AGRICULTURE
                               orncc or TMC SCCSC-»B>

                               WASHINGTON. 0  C. 2025O
                                                           Ipril 6   3981
Honorable Walter C. Barber, Jr.
Acting Administrator
U. S. Environmental Protection Agency
Washington, D. C.  20460

Dear Mr. Barber:

This is the U. S~ Department of Agriculture's response to the U. S.
Environmental Protection Agency's Preliminary Notice of Determination
concluding the Rebuttable Presumption Against Registration (RPAR) of
pesticide products containing ethylene dibromide (EDB).

We concur in the decision to continue use of EDB for soil fumigation and
to the postponement of a decision on several other uses pending receipt
of additional data.  We do not concur that EDB should be cancelled, for
fumigation of felled logs, or for quarantine purposes.  It is our con-
clusion that there is insufficient information to adequately Indicate
that gamma irradiation 1s or will be a feasible substitute for EDB
quarantine fumigation on a broad scale basis by July 1, 1983.  We
believe that, in light of new information on use and the lack of viable
alternatives the actions on these uses of EDB should be Devaluated.

We agree that, more data are needed on residue levels of EDB on treated
and processed commodities as well as more data on the effectiveness of
worker and applicator protection techniques associated with EDB.  We
will cooperate with: EPA and industry to provide additional data as
necessary.

Our specific comments are contained in the enclosure which is an Inte-
gral part of this response.  In view of the complex issues Involved, the
additional time that your agency granted for review of this document was
very beneficial and is appreciated.

Sincerely,
Secretary



Enclosures

-------
                           SECRETARY OF AGRICULTURE'S
                              SPECIFIC COMMENTS TO
                       EDS NOTICE OF DETERMINATION  PO 2/3
  I.   Preplant Soil Funrigatlon

USDA is willing; to provide input and assistance to EPA and Industry to
develop acceptable experimental design and/or additional data to resolve
concerns associated with potential residues on crops grown 1n soil that
has been treated, with EDB.

 II,   Fumigation of Stored Grain

More recent data has been developed from an interagency study in SEA:AR,
ASCS, and FGIS, which 1s now in progress to measure the level of insect
and fungal activity in farm stored grains  This study consists of grain
samples and storage information from about 4,000 bins of wheat,. 3,000
bins of corn, and 1,000 bins of oat stored on farms in 20 States.
Storage information- obtained in this study includes a record of measures
taken by the producer to maintain the quality of the grain during storage.
Preliminary analysis of these data indicated that less than 3 percent of
the wheat bins and less than 1 percent of the corn and oat bins were
fumigated during the 1-4 year on-farm storage.  Fumigation occurred most
often during the second year of storage, generally in the late summer or
early falU  Fumigant materials identified as being used in the farm
treatment included aluminum phosphide» chloropicrin, and a liquid fumi-
gant mixture composed entirely of carbon tetrachloride (CC1.) and carbon
disulphide (CS-) (80:20) or these two used in conjunction with 1.2 to 5
percent ethylene dibromide.

An Interim analysis of EDB use in on-farm storage may be estimated from
the study now. in progress whicrr shows the following percentages of grain
being treated with EDB.
Grain in
Storage
Wheat
Com
Oats
Barley
Sorghum
Amount In Storage
(1,000 bushels)
972,054
2,494,504
388,625
240,442
39,708
% Treated
2.0
0.5
0.25
0.25
0.25
Lbs. EDB Used
52,492
4,489
349
108
54
The results of this study are expected to be tabulated in computer
Storage by May 1, 1981.

EDB liquid fumigant formulations are labelled for use in both on-farm
and commercial elevator storage.  In actual practice EDB fumigation of
stored grain is concentrated in on-farm storage.  Liquid formulations
were developed specifically for this purpose.  EDB is included in liquid
grain fumigants to improve effectiveness in the upper layer of the
treated grain.  In the event of the cancellation of EDB for this use,
fanners would have to switch to a less efficient liquid fumigant, 80
percent carbon tetrachloride/20 percent carbon disulfide or phosphine.
Carbon tetrachloride is presently under RPAR review.  The additional
applications necessary for comparable control would result in both
increased expense and exposure hazard because of lower efficiency.

                                  283

-------
                                  284

Some level of mixing with untreated grain would be expected in movement
fronr on-farm storage, in further movement from the elevator, as well as
mixing in the milling or other processing operations.  The level of EDB
(as organic bromide) in most processed products should, therefore, be
low.

Because of the changes in use patterns and the current availability of
more accurate, data on the volume of use, EPA's proposed regulatory
action regarding fumigation of stored grains should be reevaluated.

in.   Spot Fumigation of Grain Mill ing Machinery

Three types of funrigant formulations containing EDB have been used in
the fumigation of milling machinery.  The two liquid types contained
(CC14K ethylene dichloride (EDC) and EDB in different proportions.  The
third type was a. gaseous mixture of 70 percent EDB and 30 percent methyl
bromide.  Today,. virtually the entire EDB usage in this Industry involves
the application of this 70:30 mixture and only one company is currently
formulating it.  The annual use of EDB for this purpose was originally
estimated at 465,000 1bs»  Inten'm data from the study mentioned pre-
viously indicate this use at approximately 162,000 Tbs.  The original
estimate was based primarily on the use of the EDB liquid funrigant
mixture and does not accurately reflect use of the present 70:30 mixture
in the milling industry. As with grain fumigation, the exposure resulting
from the amount of EDB known to be applied is much less than originally
estimated.

The primary alternative ta spot fumigation is genera/1 fumigation of the
entire structure.  However, general  fumigation is not feasible at.the
frequency necessary to obtain control equivalent to spot fumigation.
Most mills cannot close completely for several days at a time.  To
obtain adequate insect control, mills would have to shut down four or
more times per year.  In addition, specialized mills such as those lo-
cated in breweries cannot use either methyl  bromide or phosphine in
their plants because of the corrosive effects of these fumigants on
materials in nonmilling areas of the facility.  The use of EDB does not
pose this problem.  The common use of manifold systems in spot fumi-
gations serves to remove applicators from sites of gas generation,
lessens the risk of spillage on applicators, and thus reduces hazards to
these personnel.

Good sanitation in milling areas is  necessary for effective pest control;
however, it will  not serve as an alternative to EDB use which is still
required since milling residues collect in unaccessible places in the
machinery.  The only alternative is  spot treatment with a funrigant.

Because of the changes in use patterns to the gaseous 70:30 mixture in
the past few years, the current availability of more accurate data on
volume of use, and the fact that proposed alternatives are not economi-
cally feasible and are associated with high risk, we strongly feel that
EPA's proposed regulatory action regarding spot fumigation of milling
machinery similarly should be reevaluated.  EPA should strongly consider
and evaluate the use of protective clothing and other safety options to
minimize risk as an alternative to cancelling this critical use.

-------
                                   285

 IV.   Quarantine Fumigation

The registration for post harvest coimiodity quarantine uses should be
continued for the following reasons:

     1.   As noted by EPA, there is presently no effective and practical
          alternative to EDB for these uses.  The assumption that irradi-
          ation will be available by July 1983 is unrealistic.  Although
          EPA has evidently had interaction with specialists in the
          field of Irradiation, we are unaware of data demonstrating the
          feasibility of irradiation as a near term, viable, on-line
          alternative for bulk quantities of fruit.  Data concerning the
          time-frames for getting on-line, chamber design, as well as
          design and on-line costs should be made available.  Research
          has been conducted by USOA, as well as other groups in Hawaii,
          California, Florida, Illinois, Massachusetts, and elsewhere,
          to determine if fresh fruits and vegetables will tolerate
          insect destroying rates of gamna Irradiation.  Much of this
          research is outlined in "The Proceedings of a Panel on The Use
          of Irradiation to Solve Quarantine Problems in the International
          Fruit Trade.*  (Attached)  The panel was organized by the
          joint Foreign Agriculture Organization and the International
          Atomic Energy Agency, Division of Atomic Energy in Food and
          Agriculture, and held in Honolulu, Hawaii, in December 1970.  A
          petition  (FAP 3045} from USOA to FDA requesting approval of
          the use of gamma irradiation on papaya was submitted to FDA in
          1973.  The Food and Drug Administration has not acted on this
          petition during the past eight-year period. A new petition
          relating to these uses is in preparation.  Additional research
          is needed on mango and papaya as well as citrus and other
          fruit to determine dosages necessary to control insect pests,
          potential injury to fruit, and chemical or other changes in
          fruit composition including'feeding studies on nutritional
          value.  This data collection will require considerable time
          and expense.

     2.   EPA noted in its Notice of Determination that USDA's estimated
          benefits of post harvest commodity treatments on grapefruits
          for export market to Japan may be substantially overstated.
          The benefits estimates were premised on the assumption that
          the grapefruit export market to Japan will be lost if this
          registered use is cancelled^  This assumption has been vali-
          dated. Communication with the Government of Japan since the
          publication of PD 2/3 indicates that Japan will not approve of
          gamma irradiation as an alternative to EDB fumigation.  The
          refusal is primarily on the basis that irradiation at the
          dosages used does not actually kill all fruitflies.  In
          addition, Japan has indicated that even the effective use of
          gamma irradiation may still not be acceptable.   However, they
          did state that they would continue to accept EDB treated
          fruit.  (See attached letter of February 12, 1981, from
          American Embassy, Tokyo.)  Recent studies (SAND 79-1727,
          Attachment B) suggest that dosages of 25 to 75 K-rads will
          sterilize frultfly larvae, but not kill them.  However, this
          study did not statistically evaluate the effects of varying
          dosages of gamma irradiation on fruitfly egg, larval and pupal
          sterility and mortality, when infested commodities are sub-
          jected to irradiation.  The results are of limited value in
          assessing gamma irradiation as an alternative to EDB for
          commodity treatment.

-------
                                    286

   3.   As Indicated in USOA's biologic and economic assessment report,
        citrus and tropical fruits are imported from countries such as
        Mexico, Haiti, Israel, and Morocco, thereby strengthening our
        trade positions with then.  Of the crops received from those
        countries, cold treatment can only be used for oranges and is
        limited to certain insect growth stages at the time the fruit
        is harvested. The facilities for such treatment are not available
        and must be constructed and approved before shipments could be
        Initiated. Further, in the case of Mexican oranges in addition
        to the expense involved, the delay of a minimum of three weeks
        for cold treatment would bring their fruit into competition
        with the California harvest and could virtually eliminate
        Mexican fruit from the U. S. market.  In these cases, the
        Department's quarantine requirements cannot be net without the
        use of EDB.

   4.   EDB fumigation chamber operation, construction, and personnel
        safety specifications are available to meet or exceed proposed
        QSHA safety standards.  A revised section of the PPQ Treatment
        Manual on EDB Fumigation Chambers was issued in July 1980.
        Fumigation conducted at packing sheds in Texas and Hawaii cur-
        rently meet, the proposed O.T3 ppm (time weighted average) for
        worker exposure.  Further, a study was recently conducted en-
        titled, "Behavioral Effects of Prenatal Exposure to Ethylene
        DM bromide."  This study indicates that an exposure of 6.67 ppnr
        ED& for four hours per day on alternate days throughout gesta-
        tion did not result in an observable behavioral effect to the
        study animals.  Tests: determined that maternal behavior, sen-
        sorimotor coordination* activity, passive avoidance, complex
        learning and motivation were unaffected at exposure levels in
        the offspring of exposed pregnant Long-Evans hooded rats.
        (These documents are attached.)

   5.   The proposed cancellation of EDB post harvest fumigation of
        citrus, tropical fruits and. vegetables should not occur.  This
        determination can be based upon the newly available safety
        data contained herein and the lack of effective alternatives.
        A reevaluation could be made when all the foregoing questions
        have been resolved.

        EPA should strongly consider and evaluate the use of protec-
        tive clothing and other safety options to minimize risk as an
        alternative to cancelling these critical uses.

V. Fumigation of Felled Logs

   EDB is an important use in the control of bark beetles where non-
   chemical control methods, including site selection and thinning,
   are not viable control options.  EDB 1s only used after an Infested
   tree has been cut.  If the tree is not treated to destroy the
   beetles, they will emerge and attack adjacent trees. The proposed
   alternative, endosulfan, is not specifically registered for the
   particular insect of concern which Is the mountain pine beetle
   (Dendroctonus ponderosae).  There- 1s no efficacy data comparing Its
   effectiveness with EDB.   Lindane would be the preferred alternative
   although it was not considered by EPA as a viable alternative due
   to its RPAR status.
                                  U.S. GOVERNMENT PRINTING OFFICE : 1983 0 - 418-574

-------