ETHYLENE DIBROMIDE (EDB)



Scientific Support and Decision Document for Grain



         and Grain Milling Fumigation Uses
           OFFICE OF PESTICIDE PROGRAMS




   UNITED STATES ENVIRONMENTAL PROTECTION AGENCY






                 FEBRUARY 8, 1984

-------
             ETHYLENE DIBROMIDE (EDB)



Scientific Support and Decision Document for Grain



         and Grain Milling Fumigation Uses
           OFFICE OF PESTICIDE PROGRAMS




   UNITED STATES ENVIRONMENTAL PROTECTION AGENCY






                 FEBRUARY 8, 1984

-------
                             EXECUTIVE SUMMARY



    Agency  has determined  that  further Federal regulatory  actions for

pesticide uses of  ethylene dibronide  (EDB)  are necessary to reduce the

risks  to human healthTj These actions include:

     "linniediate emergency suspension of  all  uses  of  EDB to fumigate  stored
       grains and  grain milling equipment?]

     0 guidelines  for maximum permissible levels of EDB  in raw grains and
       consumer products made fron  grains;

     0 initiation  of the rulemaking process to revoke the  exemption from
       the  tolerance requirements for EDB on  grain products which is
       currently applicable; and

     0 establishment of appropriate federally enforceable  maximum residue
       levels for  EDB.

    exposure can pose risks to  human  health which  include  cancer,

reproductive disorders  and heritable  genetic  damage?)  Consequently,

residues of EDB occurring  in consumer products raise  public health

concerns.   The actions  the Agency is  taking will reasonably reduce the

present  risks to human  health from  the uses of EDB to fumigate stored

grains and  to fumigate  grain-milling  equipment.



In September 1983  EPA announced a series  of regulatory decisions for each

of the pesticide uses of EDB.   The  Agency issued a Notice  of Intent to

Cancel registrations for the uses of  EDB  to fumigate  grains and milling

equipment.  Pursuant to §6(b) of FIFRA, the cancellation of pesticide

products which are the  subject  of a Notice  of Intent  to  Cancel does not

become effective when adversely affected  persons file a  request for a

hearing  for particular  registrations  and  uses. Hearing  requests were

submitted.  Subsequently two hearing  requests have been  withdrawn.

-------
                                  ii






As a result of these challenges,  the  sale  and  distribution of several



EDB-containing pesticides  for use on  stored grain  and  grain-milling



equipment have continued to  be  legal  under FIFRA.   Historically,



adjudicatory cancellation  hearings  have  lasted about two  years and may



extend for longer periods.   The Agency has concluded that this extended



period of time is unacceptable  in light  of the additional risks to human



health which would be expected  to result as use of EDB continues.  In



its 1983 decision, the Agency stated  that  it would consider emergency



suspension of the grain registrations of EDB  if additional monitoring



data confirmed widespread dietary residues in  consumer grain products.







Recent results fron numerous testing  programs  of consumer products



confirm that there are residues of  EDB in  the  nation's grain-based food



supplies.  EPA's monitoring  efforts were aided by  collaborative efforts by



the Food and Drug Administration and  the United States Department of



Agriculture.  Several states have supplied the results of significant



numbers of samples of grain  products  analyzed  for  EDB.  In addition, the



American Bakers Association  and the Grocery Manufacturer's Association



and certain of its member  companies have conducted extensive EDB analyses



and submitted those results  to  EPA.  The Agency has evaluated the results



which have been provided by  these public and private organizations, and has



concluded that it would be appropriate to  recommend EDB residue guidelines



below which grain products should pose no  extraordinary risks to public



health.  These recommendations  are  being made  available to the public as



well as to state governments and  to the  various sectors of the grain

-------
                                       iii
 products industry.  The Agency will  initiate  the  rulemaking  process



 necessary to revoke the current exemption of  grain products  containing  EDB



 from the tolerance requirements of the Federal Food, Drug, and Cosmetic



 Act, and will seek the establishment of federally enforceable maximum



 residue levels.








 The Agency has further concluded that the uses of EDB to  fumigate grains



 and milling equipment are responsible for residues of EDB in the nation's



 grain-based foods and such use must be stopped immediately to eliminate



 continued contamination of grain-based consumer products.








 The Agency has evaluated the economic impact of these actions and



 reconnendations and has determined that there could be sane economic



 disruptions,  but that they are not unreasonable in light of the reduction



 in  the  hazards to human health which are expected to result.








 This document summarizes the Agency Staff evaluation of exposure, risk  and



 economic  consequences of implementing alternative maximum permissible



 residue levels of EDB in grain and grain products and,  at the same time,



 placing an  immediate  emergency suspension upon further use of EDB for



 fumigation of grain and grain milling equipment.   The consequences of a



 number  of optional  levels  of  maxium permissible EDB residues were



 evaluated, for  raw grain,  cooked/ready-to-eat grain products and uncooked



grain products.   The  results  of the analysis are  summarized in Table 1,



 including the option  selected, which was  to  recommend against human



consumption of products  with  EDB residues  above 30 ppb  (parts per billion)

-------



Hsnrussicde


[to.

1

2

3

4

5

6

7

a

9

level

FEBa/

N3 Oaair^p
Limit
9U/250/
5,000
30/150/
3,000
30A50/
1,000
JU/L50/900

24/100
2,000
10/50/
1,U»
5/25/500

Witt)



(Estimated Actual
Intake)
HB in
Oain Diet

2.9

2.1

1.8

1.8

1.8

1.6

1.1

0.8

0.5

Kj^/t&y
ChUd/Milt

1.3xlO-^/
6x10-"
9x10^
4x10-*
axio-fy
3x10-^
SxlO-V
3xlC^*
8xlO-5/
3x10^0
TxloA/
3xlO-°
SbdO^6/
2x10-6
3K10-6/
2x10^
2X10-6/
9x10-''
10 ^ SlldlTtJ ar-alo 1.2 SxlfMV



a/ Permissible residua levels
Vo
h/ L*a
if we
:rt.ct to the
alCLc .IcVeL
AlO-o

Canoer
Risx UILULU
aiuiu CE. oj/i\*jf£-i
S^ J&ry; in cancer risk usiry dietary cancer nrdels presented
d/ Aoretant exD
jeuee vas ass
Ljicd tor calc
ilal-mn ori'1^1^.
UUU IULU-MHU Ly (J
enndTKTIC OTUUHJHIt:!
U_ U. J1J/3U/±,UU1
JRD.

Ifew Grain
P&lXEXt.
ftOK
Level
(pet)

NA

0.2

0.2

0.6

0.7

0.2

0.6

1.4

6.3

0.2/0.6
value
Arrve
Level
(ma.)

S ryfc

26

3S

73

97

26

73

193

858

13/86
a 5-fold increase from the finished
JIB a 6-7- told increase fron uncooked predict to grain) .



in the ED 2/J and SO 4.
EXnosjce is nred
ictsd to cbclire after the susuens
ion is in effect
f


antes:  See TfebLes VT-1, VIII-B, 9, 15 and 18.

-------
in ready-to-eat products, 150 ppb  in  prepared/uncooked grain products and



900 ppb in raw grain.  Enforcement of this  option would mean that generally



1 to 2 percent of grain and  related products would exceed these levels,



with such stocks having a value of nearly 5150 million.  Economic impacts



would be most significant in the immediate  future, and would dissipate



as EDB-contaminated stocks of grain and  related  products are eliminated



from the food system  in a period to three to five years.

-------
                         ACKNOWLEDGEMENTS







Arnold Aspelin, Economist



Gary Ballard, Economist



Anne Barton, Statistician



Gary Burin, lexicologist



Jim Cogliano, Statistician



Roger Holtorf, Economist



Cara Jablon, Attorney



Richard Johnson, Team Leader



Robert Lenahan, Economist



Robert Mclaughlin, Attorney



Sami Malak, Residue Chemist



Marcia Mulkey, Attorney



Richard Schmitt, Residue Chemist



Linda Stallard, Secretary



Peggy Stuart, Secretary



Edward Zager, Residue Chemist







Consultants







S.B. White, Statistician (Research Triangle Institute)



Andy Clayton, Statistician (Research Triangle Institute)



Barbara Petersen, Nutritionist

-------
                            TABLE OF CONTENTS
Executive Summary
I.      Introduction	   1



II.     Background	   4



III.    Use and Usage	   7



IV.     Residue Data	  10



V.      Consumption Pattern and Dietary Burdens	  33



VI.     Cancer Risks	  51



VII.    Toxicity of Likely Chemical Alternatives	  57



VIII.   Economic Impacts	  63

-------
I.  Introduction







On September 28, 1983, EPA  issued a Notice of  Intent  to Cancel



registrations for the pesticidal uses of EDB which were determined  to  be



responsible for residues occurring in consumer grain  products.   These  uses



include spot fumigation of  grain milling equipment, and fumigation  of



stored grains.  These cancellation notices were  the result of an evaluation



of the long-term risks to public health and the  economic  consequences  of



cancellation.







FIFRA §6(b) provides that adversely affected parties  may  challenge  these



cancellation actions through requests for adjudicatory hearings. For  the



specific products for which hearings nave been requested, registrations



remain unaffected by the cancellation actions  until the cancellation



proceeding is concluded for these products.  The products subject to the



cancellation hearing, therefore, can continue  to be sold, distributed, and



used until the hearings end and  the Agency  announces  its  final regulatory



decision.  Historically, adjudicatory hearings average 2  to 3 years or



more.







In announcing its decisions to cancel registrations of EDB for fumigation



of stored grain and milling machinery,  the  Agency stated  that further



federal action might be necessary  if the cancellation was delayed by a



hearing request and additional residue  data increased the Agency's  concern



for human health risks.  At the  time the notice of cancellation was issued,



there was very little data  on  the actual residues of  EDB occurring  in



grain-based consumer products.  The cancellation action  was instead based



primarily on residue levels which the Agency predicted would result from



these two uses of EDB.

-------
The  information now available on residues occurring in consumer products

is greatly  expanded,  and clearly demonstrates the need for additional

federal  actions.  These actions include:

      1.   Emergency suspension of the uses of EDB for spot fumigation of
          milling equipment and fumigation of stored grain;

      2.   Revocation of  the exemption from tolerances of residues of ECB in
          fumigated grains, and;

      3.   Steps  to establish appropriate federally enforceable maximum
          permissible residue levels.

Claims that this contamination is the result of other sources of EDB such

as gasoline or  a naturally occurring source of EDB have been rejected.  The

Agency knows of no scientifically sound studies which can reasonably

explain  that these or other possible sources are responsible for this

extensive contamination of the nation's food supply.



The comprehensive scientific and technical analyses and extensive data

discussed in this document and the PD 2/3 and PD 4 provide the basis for

the Agency's decision to emergency suspend the stored grain and milling

uses  of  EDB.  Additional analyses provided are the basis for the Agency's

reconnendation  of the maximum permissible residue levels during the period

of time  required for  treated grains to clear the food supply.  Included in

this  document are:

-------
      1.   a survey of the extent of use of EDB to fumigate grains and to
          fumigate grain  milling equipment;

      2.   a detailed evaluation of all available data on residues of EDB
          occurring in grains,  and grain-based consumer products before and
          after  cooking;

      3.   an evaluation of dietary consumption patterns to determine
          "typical" consumption of grain-based products and levels of intake
          for subgroups of the  population with high grain consumption;

      4.   dietary  burdens or  residue consumption levels for dietary patterns
          developed in 3  above, both for best judgement estimates of EDB
          contamination levels, and also for hypothetical maxmimum permitted
          residue  standards;

      5.   estimated human health risks for dietary burden levels developed
          in 4 above;

      6.   economic impacts expected to occur for hypothetical maximum
          permitted residue levels and impacts of the use of alternative
          fumigation methods;

      7.   economic impacts of  immediate suspension of the grain and milling
          registrations during  the next six months and two years; and

      8.   summaries of readily  available information on the toxicity of
          alternative chemicals.
This document reflects  the  information upon which the Adminsitrator's

February 3, 1984 Suspension Order  and  recommended maximum residue levels

were based.  A  final February  3, 1984  version of this document has been

maintained by EPA.  This February, 1984 document reflects additional

explanatory material and some  additional analyses, but does not materially

differ from the February 3  document.

-------
II.  Background








On September 28, 1983, the Administrator signed a Notice of Intent  to



Cancel the registrations of pesticide products containing EDB and labeled



for use as a fumigant of grain stored in bulk, and as a spot treatment  of



grain milling equipment.  The cancellations of 16 registrations  for the



stored grain uses, and 4 registrations  for spot treatment of milling



equipment did not go into effect because requests for a hearing  on  the



proposed cancellation action were filed by 9 parties.  One registrant has



since withdrawn and voluntarily cancelled 2 registrations for the stored



grain use.  A second registrant has voluntarily cancelled a registration



for grain fumigation and spot treatment of milling equipment.








The Agency's notice of cancellation concluded  an extensive analysis of  the



risks and benefits of the use of EDB as a pesticide  under the Agency's



Rebuttable Presumption Against Registration  (RPAR) process.








The Environmental Protection Agency issued a Notice  of Rebuttable



Presumption Against Registration and a Position Document 1  (EDB PD  1)



concerning the pesticide products containing ethylene dibrcmide (EDB)  in



the FEDERAL REGISTER of December 14, 1977  (42  FR 63134).  The Agency took



this action in response to test results which  showed that EDB  induced



cancer, mutations, and adverse reproductive effects  in  test  animals.  The



EDB PD 1 discussed the relevant test data and  the Agency's concerns with



the continued use of EDB as a pesticide.  The  document  also  solicited



public comments and additional data.

-------
Comments and data in response to the BOB PD  1 were  submitted  to  the  Agency



by registrants, user groups, and other members of the public.  The Agency



carefully evaluated the comments and additional data vihich were  submitted,



and in the FEDERAL REGISTER of December 10,  1980  (45 FR 81516),  the  Agency



announced the preliminary notice of determination for EDB and the



availability of the EDB PD 2/3.







In the EDB PD 2/3, the Agency concluded that EDB poses  a significant risk



of oncogenic, mutagenic, and adverse reproductive effects in  the human



population.  In addition, the EDB PD 2/3 contained  analyses of the  risks



and benefits of continuing the various uses  of EDB  and  presented the



Agency's proposed regulatory decisions.







The EDB PD 2/3 was submitted to the Scientific Advisory Panel (SAP)  and the



U.S. Department of Agriculture (USDA) for  review  and ccmment.  The  SAP held



public hearings concerning the EDB PD 2/3  on March  25-26, 1981,  and  heard



presentations by the Agency, registrants and other  interested members of



the public.  Both SAP and USDA submitted comments on the EDB  PD 2/3 to



EPA on April 22, 1981 and April 8, 1981, respectively.







The Agency again received numerous comments  fron registrants, user  groups,



and other members of the public.  All comments were reviewed  and evaluated



by EPA in arriving at the final regulatory positions concerning the uses of



EDB as a pesticide.  Comments which  included significant information were



analyzed and discussed in the EDB PD 4.  The Notice of  Cancellation was



published in the FEDERAL REGISTER on October 11, 1983  (48 FR 42668).

-------
Since the Notice of Cancellation was issued,  an administrative hearing has



commenced to consider, among other  things,  the cancellation of pesticide



products containing EDB  for use  as  a fumigant of stored grain and a spot



treatment for milling equipment.  The first prehearing conference was held



on December 1, 1983.  A  second prehearing conference is scheduled to be



held on March 5, 1983.   The date of commencement of the testimony in the



hearing is not yet set,  but it  is estimated that least 2 years may elapse



before a cancellation order could be issued at the close of the hearing.

-------
                                      7





 III.   EDB Usage Estimates







 An estimated 452,000 gallons of formulated products containing 471,000



 pounds of EDB active ingredient are used annually for insect control in



 grain stored in bulk and flour milling  machinery (Table III-l).








 For grain storage  uses,  425,000 gallons of formulated products were



 applied to an estimated  202  million bushels of grain in 1983.  Usage on



 wheat is believed  to comprise about 75% of annual usage;  corn 10%;  oats,



 rye,  rice,  and barley 10%, and grain sorghum 5%.   In total,  the



 estimated bushels  treated comprise  about 2.4 percent of 1983 production



 of  these commodities,  and 1.4 percent of the total supply of these



 commodities.   It should  be noted that these estimates do not include



 mixing or blending of grain  as it moves through conmerce.







 The estimated  159.4 million  bushels of  wheat treated in 1983, represent



 about 6.6 percent  of the 1983 production and about 4 percent of total



 wheat supply  for 1983  (beginning stocks,  plus production and imports).



 Corn  is  the other  major  grain crop  treated with EDB; with about 10% of



 EDB usage in 1983  on about 14.2 million bushels.   Com treated comprises



 about 0.3 percent  of  1983 production, and about 0.2 percent  of 1983



 supply.   Stored grain  treatments with EDB are most canton in the late



 summer and early fall.   EDB  usage during the year on stored  grain is



believed  to be distributed on a quarterly basis as follows:   10% First



Quarter;  15% Second Quarter;  25% Third  Quarter; and 50% Fourth Quarter.

-------
                                            8
  Table III-l.  Estimated Annual Usage of Formulated EDB as a Fumigant of Grain
        Stored in Bulk and for Spot-Treatment of Flour Milling Equipment


                    GrainFlour
Item	Storage	Milling a/	Total	

Gallons of
 formulated product
 applied:           425,000               27,277              452,277

Pounds A.I.
  Applied:          170,400              300,600              471,000

Units treated:   202 million bu.        243 mills


a/  Usage of EDB by the majority of  flour milling firms was apparently
~   voluntarily halted  in August/September 1983.  Estimates are  for  a typical
    recent year of usage.

Sources:  Centaur, 1983; Douglas Chemical Co.,  1984; and  EPA  Estimates.

-------
                                       9
States leading  in EDB usage to fumigate stored grain  include Minnesota,
North Dakota, Iowa, South Dakota, Nebraska, Oregon and Indiana.

EDB usage to spot treat flour milling machinery was apparently
voluntarily discontinued in August and September, 1983, by the major
producers in the milling industry.  Prior to that time, typical annual
usage was about 300,600 Ibs. active ingredient.  The majority (75-80%)
of the industry employed EDB to combat insect infestations in milling
machinery.  The frequency of usage varies widely but averaged 10-11
annual treatments in 231 to 243 flour mills during recent years.
Approximately 10.5 gallons are used (range 2-45 gallons) per mill
fumigation.

-------
                                         10
 IV.   Residues of EDB in Grain and Grain Products

 This  section  contains a summary  and discussion of  the  available data on EDB
 residues in grain and grain products  throughout  the United States.

 A.  Raw Grain
The available  data  indicate  a large  variation in the EDB residues in
fumigated grain.  This  is due to  several variables  including  the  use patterns,
temperature, amount of  aeration,  and time  in  storage.  Length of  time in  storage
appears to be  the predominant factor in determining  the  level of  EDB in treated
grains.

Table IV-1 contains estimates of  the maximum  likely  EDB  levels at various
post-fumigation  intervals.
     Table IV-1. Estimated EDB residue  (ppb)  in  fumigated grain at various
intervals post-treatment

Grain
Barley,
Oats,
Popcorn,
Rice, Kye
and Wheat
Corn
Sorghum

Maximum
Application
Rate/1000 bu

1.2 Ibs.
9.0 Ibs.
11.0 Ibs.

Calculated
Initial
Residues

120,000
140,000
183,000



Post-Fumigation Time
1 wk. 2 wk.

60,000 47,000
75,000 58,000
92,000 72,000
30 days 60 days
-ppb 	
27,000 10,000
33,000 11,000
41,000 14,000
90 days

3,400
4,200
5,100
Source:  Worthington, 1978

The calculated initial residues  in Table  IV-1  are derived  assuming that the
maximum registered application rates  are  used.  If lower application rates

-------
                                           11
are used, the calculated values would be  reduced proportionately.



Moreover, these calculations assume  there  is no mixing of  treated grains



with untreated grains and  therefore  no  reduction of  residues  through



dilution.







A large volume of monitoring data  for grains has been received  by the



Agency.  These data are summarized in Table IV-2.  The data indicate  a  wide



variation in residue levels, which would  be expected from  the use of  EDB on



a small percentage of the grain (5%).   Surprisingly, however, the extent of



EDB contamination of raw grain is much  greater than  5%.  About  60% of all



tested raw grain samples contained detectable residues of  EDB.  This  is



likely attributable to mixing during storage and transit of treated grain



with grain which has not been treated or  that the estimate of the percent



of grain treated was wrong.







The monitoring data described in the previous paragraph were  combined to



give an estimate of the distribution of EDB in raw grains. The combined



results of all of these studies are  presented in Table IV-3.  In  this



table, the total number of samples,  the average residue and the standard



deviation of the average residue are given in the  last three  columns



entitled "no regulatory restriction  level."  Values  are given for wheat,



corn and other grains (rice, barley, oats, rye and sorghum).  Table IV-3



also includes the effect of setting  various maximum  permissible residue



levels.  This calculation  assumes  that  enforcement results in removal from



commerce of 100% of all grains containing residues above the  maximum

-------
                          12
Table IV-2.  Residue Data for EDB in Raw Grain

Study
FDA-1

NC


USDA-1


USDA-2


GMA


Texas

Reference Grain
IV 3 Wheat
Other
IV 9 Wheat
Corn
Other
IV 14 Wheat
Corn
Other
IV 12 Wheat
Corn
Other
IV 1, 2 Wheat
Corn
Other
IV 5 Corn
Other
Number
of
Samples
31
6
2
3
3
10
10
10
197
202
1
622
69
90
6
2
Average
Residue (ppb)
33
2000
27
N.D.
N.D.
4.9
0.9
2.1
36
5.6
N.D.
43
220
5.7
1.4
N.D.
Standard
Deviation
85
4000
__
—
—
7
0.5
2.4
61
26
—
150
1200
20
1.6
—

-------
Table IV-3 Effect of Proposed Maximum Permissible Residue
Levels (MPL) on Average (mean) EDB Residues In Raw Grains


Guideline
Level In
ppb
Product

Wheat
Corn
Other
100

No.
Samples
below mpl
No. (I)
783 (90.8)
279 (96.2)
108 (96.4)

Ave rage
Residue

12
5
3.7
500

No.
Samples
below mpl
No. (Z)
8SO (98.6)
286 (98.6)
110 (98.2)

Ave rage
Residue

27
13
7.0
900/1000

No.
Samples
below mpl
Ho. (2)
858 (99.5)
289 (99.7)
110 (98.2)

Average
Residue

34
21
7.0
2000

No.
Samples
below mpl
No. (*)
862 (100)
289 (99.7)
111 (99.1)

Average
Residue

40
21
21
3000

No.
SampJ es
below mpl
No. (Z)
862 (100)
289 (99.7)
111 (99.1)

Average
Residue

40
21
21
5000

No.
Samples
below mpl
No. (Z)
862 (100)
290 (100)
111 (99.1)

Average
Residue

40
21
21
No
Restriction Level

All
Samples

862
290
112

Average
Residue

40
56
110

Standard
Deviation

130
590
950
                                                                                                                               Osl

-------
                                            14
permissible residue  level.  The extent of  reduction will depend on  the



effectiveness of enforcement.  Thus,  the averages given in Table  IV-3



should be taken as the maximum possible reduction in  average  residues,



which may not be achieved  in reality.







B.  Effect of Milling on Residues  in  Grain







Residues in processed grain products  can result  from  residues in  the bulk



grain and also from  residues resulting fron  spot fumigation of mills and



milling equipment.  While  the milling process may reduce levels (i.e.



produce lower levels in flour than in bulk treated grain), spot fumigation



of mills will contribute additional residues during milling.







          1.  Effect of Milling on Residues  in Grain







In a controlled study by Wit et. al.  (IV-16)  (1969),  treated  wheat



containing an average of 9,500 ppb EDB was milled.  After milling,



residues in the flour averaged 3,000  ppb (68% reduction) and  residues  in



shorts/bran averaged 14,600 ppb (a concentration factor of 1.5x).  In  a



similar experiment by Berk (IV-15), treated  wheat grain containing  an



average of 376 ppb EDB was milled.  After  milling, residues  in the  flour



averaged 98 ppb (74%, reduction),  residues in the shorts/bran averaged 106



ppb (0.3x) and residues in middlings  averaged 78 ppb  (0.2x).

-------
                                             15
 Recently (1983), in a controlled experiment by the USDA (IV-18), fumigated



 wheat grain containing an average of 3,690 ppb EDB was milled.  After



 milling the average residues in the flour was 520 ppb (after subtracting



 contribution from spot treatment),  a reduction of 86%.








 In a study carried out by the Grocery Manufacturers of America (IV-1),



 fumigated wheat containing 17,816 ppb EEB was milled.  After milling,



 residues in the flour were an average of 967 ppb (95% reduction).








 It  is concluded from these studies  that residues in the treated grain are



 retained in the processed fractions.  Residues in the flour are generally



 lower than the  residues in the grain.   EDB residues may concentrate in



 selected milled fractions other than flour.








 The  studies reported above are all  controlled experiments and are presumed



 to be broadly indicative of the changes in residue levels that would occur



 during  commercial milling,  tto data on the effect of processing on



 residues  in milled  products are available for grains other than wheat.








           2.  Contribution from Spot Fumigation








 In a  study conducted by Litton Bionetics (IV-17), it was shown that spot



 fumigation with EDB resulted  in residues in flour that generally follow a



 loy-linear curve.   Initial EDB residues in the wheat-flour in the first



half  hour  after start-up  of the milling process was determined at 2,750

-------
                                         16
ppb.  This level declined to 900, 170 and 30 ppb  after  1.5  hrs.,  3  hrs.,



and 1 week, respectively.  A time-weighted average contribution  to  the



wheat flour from spot treatment was estimated  in 'the Agency's PD-4



(September 27, 1983) at 34-68 ppb (based on a  4-week time weighted



average).  Actual  levels  in flour ranged from  54  to 2,750 ppb, some of



which may have been attributable to bulk grain fumigation.








In a controlled study conducted by the USDA (IV-18),  it was shown that



contribution to the initial wheat flour from spot  fumigation averaged 240



ppb.  Because the  wheat was treated and processed  in a  laboratory study,



contribution frcm  spot treatment in this test  may  not be representative of



grain in commerce  that received multiple applications.   It  is  also  noted



that this study was run for only a short time  after  fumigation,  commercial



mills are normally treated only once every 2-4 weeks, and residues  in flour



fall to lower levels with time.  Thus the 240  ppb average residue observed



in this study is higher than the average residue  expected  in a conmercial



mill.








C.  Intermediate (Uncooked Grain Products)







Monitoring data for uncooked grain products were  reported by the Grocery



Manufacturers of Anerica, Inc., FDA, Midwest Research  Institute, the States



of Florida, Georgia, North Carolina, California and  Texas.   The  data for



each of these studies are summarized  in Table  IV-4.  The data on uncooked



grain products were analyzed as described previously for the raw grains.



The results of this analysis are presented  in  Table  IV-5.   This table

-------
                                 17
Table IV-4.  Residue Data for EDB In  Uncooked Grain Products

Study
FDA-1

FDA-2
Georgia


Florida

Texas


N.C.


GMA


MRI

Reference Grain
IV 3 Wheat
Other
IV 3 Wheat
IV 8 Wheat
Corn
Other
IV 6 Wheat
Corn
IV 5 Wheat
Corn
Other
IV 9 Wheat
Corn
Other
IV 1, IV 2 Wheat
Corn
Other
IV 10 Wheat
Corn
Numbe r
of
Sample
13
3
10
4
9
1
39
107
40
50
32
51
43
1
473
74
9
8
20
Average
Residue (ppb)
4.0
1.2
1.8
3.7
3.2
0.75
2.9
19
21
8.5
5.4
19
4.3
2.3
15
140
0.87
22
16
Standard
Deviation
4
—
2.8
—
5.5
—
7.4
86
20
26
23
38
7.4
—
46
250
0.35
37
25

-------
Table IV-5.  Effect of Proposed  Maximum Permissible  Residue  Levels
MPL on Average (mean) EDB Residues  in  Uncooked  Products

Guideline
Level In
ppb
Product
Wheat
Uncooked
Products
Corn-
Uncooked
Products
Other
Grains-
Uncooked
Products
5
No.
Samples
below mpl
No. (Z)
425 (66.6)
193 (63.7)
43 (93.5)

Average
Residue
1.6
1.3
0.89
25
No.
Samples
below mpl
No. (Z)
S58 (87.5)
257 (84.8)
45 (97.8)
Average
Residue
3.8
3.7
1.3
50
No.
Samples
below inpl
No. (Z)
598 (93.7)
267 (88.1)
45 (97.8)
Average
Residue
6.1
5.0
1.3
100
No.
Samples
below npl
No. (Z)
623 (97.6)
279 (92.1)
45 (97.8)
Average
Residue
9.2
7.6
1.3
150
No.
Samples
below npl
No. (Z)
628 (98.4)
281 (92.7)
46 (100)
Average
Residue
10.0
8.4
4.0
250
No.
Samples
below mpl
No. (Z)
632 (99.1)
286 (94.4)
46 (100)
Averag"
Residue
11.0
12.0
4.0
No Restriction
Level
All
Samples
638
303
46
Average
Residue
14.4
44.0
4.0
Standard
Deviation
42
140
19
                                                                                                                                     00

-------
                                          19
 includes the total number of samples,  the average  residue  and  the  effect of



 various maximum permissible residue levels on the  average  residues in



 wheat, corn and other grains.  The data generated  by Pains and Holder in



 1980 (IV-19) on school lunch flour were not included in  this data  analysis.



 These results were significantly higher than the current monitoring data



 and thus are not considered reflective of current  residue  levels.








 D.  Cooking Studies








 Several studies have been carried out to determine the extent of reduction



 of EDB residues upon cooking.   In this section only data on the cooking  of



 commercial  products are included.  Studies involving spiked products  are



 not included.








 In a study  by  the  Midwest Research Institute (IV-10),  32 samples of corn



meal,  hushpuppy mix,  instant and regular grits,  cake mixes, pancake and



cornbread mixes containing  from <0.5-110 ppb EDB were  prepared, as



ready-to-eat products,  following the  label  instructions.   Residues of EDB



in the cooked products  were reduced by 67-99%  and ranged fron ND «0.75



ppb) to 36 ppb  in  the final product.








In a study by FDA  (IV-11), 7 samples  of cornmeal,  cornmeal  mix,  quick grits



and hushpuppy mix obtained from Florida and  containing  from 1.9-586 ppb EDB



were prepared as ready-to-eat products following standard recipes.



Residues of EDB were reduced by 47-97% and ranged  frcm  <1 ppb-92 ppb in the



ready-to-eat products.

-------
                                          2C
In a study conducted by the Grocery Manufacturers of America, Inc. (IV-1),



95 samples of wheat flour, wheat yerm, pasta, mixes, corn meal, hushpuppy



and grits obtained fron GMA member companies containing EDB residues frcm




-------
                                            21
        Table  IV-6.   Data on the Effect of Cooking on EDB Residues*
Study Reference
Florida IV 6
FDA IV 11
GMA IV 2
MRI IV 10
Number
of
Experiments
32
8
100
13
Average
Reduction
88.4%
84.0%
80.2%
79.0%
82.0% weighted
average
*Not counting those experiments  in which  the uncooked product contained
 <3 ppb EDB.

-------
                                         22
 E.  Ready-to-Eat Grain Products








 Monitoring data for ready-to-eat products derived from grains were



 reported by the Midwest Research Institute, and the States of Florida,



 Georgia, North Carolina, California, and Texas and the Grocery Manufacturers



 of America, Inc.  The data from each of these studies are summarized in



 Table IV-7.







 The data on ready-to-eat grain products was analyzed as described for raw



 grains.   The results of this analysis are given in Table IV-8.  This table



 shows the total number of samples analyzed, the average residue levels and



 the effects of various maximum permissible residue levels on the average



 residues found in ready-to-eat wheat, corn and other grain products.







 The data on residues in uncooked and cooked grain products in Tables IV-5 and



 IV-8 were used to estimate the average residue level for EDB in ready-to-eat



 grain products.   In  this calculation the Agency assumed an 80% reduction of



 residue  on cooking.   Thus, the average residue levels shown in Table IV-5 for



 uncooked grain products were reduced by 80% and averaged with the residue data



 for cooked commodities given in Table IV-8.  The average residue level was



 calculated by  weighting the average residue by the number of cooked and



 uncooked samples.  Samples with no detectable residues were treated as having



 0.75 ppb.   The Agency utilized the uncooked sample data in addition to the



 cooked sample  data because of the much larger number of uncooked samples



analyzed.   The Agency did not use the data from Table IV-3 for raw grains



because  of the larger error involved in extrapolating from raw grains to



cooked ready-to-eat  residues.   The results of the weighted average

-------
                                         23
Table IV-7.   Residue Data for EDB in Cooked Grain Products
Study
FDA-2

Georgia


Florida


Texas

N.C.
GMA


MRI

Reference
IV 3

IV 8


IV 6


IV 5

IV 9
IV 1, IV 2


IV 10

Grain
Wheat
Other
Wheat
Corn
Other
Wheat
Corn
Other
Wheat
Corn
Wheat
Wheat
Corn
Other
Wheat
Corn
Number
of
Sample
10
12
3
6
3
2
29
9
3
6
5
243
31
65
6
20
Average
Residue (ppb)
N.D
N.D
N.D.
4.5
N.D.
N.D.
4.7
N.D.
N.D.
1.1
7.7
2.2
2.9
0.84
6.8
4.7
Standard
Deviation
_
—
__
5.3
—
__
10.0
— •
«
0.76
9.5
4.9
4.5
0.41
14
7.5

-------
Table IV-8.  Effect of Proposed  Maximum Permissible  Residue  Levels (HPL)  on
Average (mean) EDB Residues  In Ready~to-Eat-Producta


Guideline
Leve 1 In
ppb
Product

Wheat
Ready-
to-Eat
Product s
Corn-
Ready-
to-Eat
Products
Other
Grains-
Ready-
to- Eat
1

No.
Samples
below tnpl
No. (Z)
217 (79.8)



55 (64.0)



95 (95.0)




Average
Residue

0.72



0.76



0.66



5

Ho.
Samples
below rapl
No. U)
248 (91.2)



66 (76.7)



100 (100)




Average
Residue

0.91



0.96



0.73



10

No.
Sanples
below mpl
No. (*>
255 (93.8)



73 (84.0)



100 (100)




Ave rage
Residue

1.1



1.6



0.73



20

No.
Samples
below opl
No. (Z)
267 (98.2)



84 (97.7)



100 (100)




Average
Residue

1.7



3.3



0.73



30

No
Samples
below mpl
No. (Z)
269 (98.9)



85 (98.8)



100 (100)




Average
Residue

1.8



3.5



0.73



50

No
Samples
below tnpl
No. (Z)
272 (100)



85 (98.8)



100 (100)




Average
Residue

2.3



3.5



0.73



No Restriction
Level

All
Samples

272



86



100




Average
Residue

2.3



4.1



0.73




Standard
Deviation

5.3



7.4



0.46



                                                                                                                                        ro
                                                                                                                                        -fa.

-------
                                           25
calculations are given in Table IV-10.  The results given  in Table  IV-10  are



skewed due to overrepresentation of seme portions of  the diet.  The averages



are based on the samples analyzed and not corrected for  the percent of the



diet of each product type.







F.  Limitations of the Residue Data







While a large volume of residue data are available, there  are  several



limitations that must be considered  in  evaluating  these  data.   The  maximum



permissible residue levels discussed in this  paper should  be accepted with the



caveats as discussed below.







The estimates are based mostly on residue data from monitoring studies and



only a few controlled experiments.  The data  cones from  a  variety of sources



which possibly followed different sampling  and testing protocols.







In most of the submitted residue studies there is  no  significant information



available concerning application rates; methods of application; whether bulk



or spot fumigation was used;  length of  exposure, aeration  intervals,



temperatures during treatment and storage or  the interval  from treatment to



sampling.  Without treatment  histories  for  the monitoring  samples,  the Agency



cannot ensure that the maximum residue  level  that  may result from the



registered uses are found.  On the other hand, some of the values reported in



the monitoring studies may be the result of misuse and may,  therefore, be



higher than the maximum residue level that  would likely  result fron the



registered uses.  Further, residue  levels may vary seasonally due to



temperature variations during treatment and shipping.

-------
                                            26
 The tollowing caveats are of somewhat lesser importance and are not expected to



 significantly affect the overall results.  The available studies do not usually



 specify whether standardized sample collection procedures were followed; e.g.,



.whether grain tiers were uses for taking core samples of commodities in bulk



 quantities;  whether samples were composited by quartering; whether the quantity



 sampled was  adequate;  and whether the samples shipped and stored were frozen or



 in  cold storage.







 In  general,  various modifications of the Rains/Holder method (J.ADAC 64(5)



 1252-54,  1981)  were used to quantify the residue levels.  For some of the data



 submitters,  the analytical methodology used was not given.  In only a few



 instances were  the  data  cross-checked by analysis at different laboratories.







 In  no study  is  the  sample representative of all products on the market.



 Indeed,  some  studies purposely  chose samples that concentrated on products



where high residues were expected;  this practice extended to looking for



particular lots.  Many variables which are  likely to be related to contamination



are either unknown  or  known to  be nonrepresentative of the U.S. as a whole.



Sane of  these variables  are:







The state data  analyzed  to date are primarily from Southern States and



California. However, new data being submitted to the Agency by various states



appears  to be consistent with the data already analyzed.  All data indicate that



cake mix and corn products,  especially hushpuppy mix,  are often contaminated.

-------
                                            27



 But many other products have never been sampled, an samples of other than



 wheat and corn products are very limited.  Thus, an important source of



 exposure to EDB may have been overlooked.  There is no way to ensure that



 all samples taken have been reported to the Agency.








 The following limitations apply to the data on reduction of residue in



 processing.  The data are not directly comparable because they core from



 different sources.  Even within a single study,  different facilities and



 procedures may have been used.  In no study is the sample of cooked goods



 representative of dietary patterns.  Because the reduction apparently



 increases with the initial residue, the selection of which products to cook



 can bias the results.  Some studies cooked only high-residue products,



 increasing the reported reduction.  Some studies may not be representative



 of  consumer baking practices.  The reduction in residue may depend on



 whether  the product was baked, boiled,  or fried; at what temperature; and



 for how  long.   These differences cannot be estimated with the available



 data.  The observed reduction in residue levels  is a combination of loss of



 EDB by volatilization,  decomposition and dilution with unoontaminated



 ingredients.   EDB residues below 1 ppb  cannot be measured accurately, thus



 for the  cooked samples  in which cooking reduced  residues below 1 ppb, the



percentage reductions cannot be computed accurately.  The Agency



conservatively treated  levels below 1 ppb as .75 ppb which should tend to



underestimate  the actual reduction in cooking.   However,  all monitoring data



and  data  on cooking are subject to the  limitations discussed above.  The



data being  submitted to the Agency on a daily basis by various states



support  the residue estimates discussed in this  section.

-------
                                          28
G.  Residues in Baby Food








fable IV-9 summarizes the available data on baby foods.  Only two of 89



samples contained detectable residues of EDB and these samples were <1.5



ppb.








H.  Residues in Meat, Milk, Poultry and Eggs








No controlled livestock feeding studies have been conducted  for EDB and



the Agency is unable to conclusively evaluate  the potential  for transfer  of



residues from grain-based feed to meat and milk.  It  is  reassuring,



however, that the monitoring to date has not shown  significant residues  in



any of the meat, milk, poultry and eggs tested to date.








I.  Additional Data







This report  includes all monitoring data available  to the  Agency  during



preparation of this report.  Additional data  is being received daily  and



will be included  in all future analyses.  The  Federal agencies, State



agencies, and trade associations response  to  the Administrator's  letter  of



January 13, 1984  requesting data on EDB  shows  an outstanding level of



cooperation.  Those data are being scanned  as  received to determine if



patterns contrary to those established by  the  statistically analyzed data



are observed.  In general, the new data  have  served to confirm previous



analyses.

-------
                                 29
Table IV-9.  Baby Foods - Data on EDB up to 1/28/84
Source
FDA
GMA (IV-1)
GMA (IV-2)


Florida

Texas (IV-5)



Georgia
(IV-8)

North
Carolina
(IV-9)

FDA (IV-3)
Ibtal
Type
infant cereal
wheat cereal
wheat cereal
rice cereal
oat meal
wheat cereal
corn cereal
barley cereal
oat cereal
multigrain
cereal
Rice cereal
multigrain
cereal

wheat cereal
rice cereal
oat cereal
wheat cereal

1 of
Samples
12
12
16
15
1
7
5
1
1

2
2

1

4
1
1
	 6
86
*
(+)
0
0
1
1
0
0
0
0
0

0
0

0

0
0
0
2
2
Average
Range Residues
(ppb) (ppb)
<1
<1
<1-1.3 <1
<1-1.5 <1

-------
          Table  IV-10.  Effect of Proposed Haulmum Permissible Residue  Levels  (HPL)  on Weighted Average
(mean)
Residue
Levels of Ready
Co Eat
Products (Table
I) and Uncooked Products

Proposed
HPL
Product
Wheat
Corn
Other Grains

1

N.D.
N.D.
N.D.

5

0.81
0.79
N.D.

10

1.2
1.1
N.D.

20

1.8
1.9
N.D.

30 SO No Regulatory Restriction

2.0 2.2 2.7
2.1 2.6 7.8
N.D. N.D. N.D.
a/*After adjusting for 80Z reduction of  residues by cooking.
N.D. - non-detectable (< 0.75 ppm).
                                                                                                                                       Usl

-------
                                           31
  J.  References

  IV(1)  Grocery Manufacturers of America,  Inc.,  tranamittal letter of 1/20/84
        Sherwin Gardner to Edwin L.  Johnson.

  IV{2)  Grocery Manufacturers of America,  Inc.,  Supplemental submission of
        1/28/84.

  IV(3)  FDA,  EDB data in grain fron  Dr.  Sanford  Miller,  Delivered  to John A.
        Moore,  D.V.M.,  November 2, 1983.

  IV(4)  USDA, January 23,  1984;  Letter from John W.  Marshall to Dr.  John A.
        Moore.   R(5)Texas Department of  Health,  Tranamittal letter from Robert
        Bernstein,  MD.D.,  F.A.CI.P., post  marked January 13,  1984  to Dr. R.D.
        Schmitt.

  IV(5)  Texas Department of Health,  transmittal  letter from Robert Berstein,
        postmarked  January 13,  1984  to Dr.  R.D.  Schmitt.

  IV(6)  Florida Department of Agriculture,  transmittal letter to Dr.  Richard D.
        Schmitt from Doyle Conner, 1/19/84.

  IV(7)  California  Department of Food and Agriculture,  Letter from Clare
        Broyhill to Dr.  Richard  D. Schmitt, January  20,  1984.

  IV(8)  Georgia Department of Agriculture,  Letter from J.R.  Conley to Dr.
        Richard D.  Schmitt,  January  23, 1984 and supplemental  submission of
        1/27/84.

  IV(9)  North Carolina Department of Agriculture,  Letter from Leonard P.
        Blanton to  Mr. Richard J. Johnson, January 10, 1984  and  supplemental
        submission  of 1/27/84.

IV(10)  Midwest Research Institute (MRI), EPA Contract #68-02-3938,  Sampling
        and Analysis for Ethylene Dibromide.  Preliminary data submitted January
        20, 1984.

IV(ll)  FDA Cooking Study, Data  verbally transmitted to Dr.  R. D.  Schmitt,
        1/20/84  and a letter  from Jerry A. Burke,  FDA  to Dr. R.D.  Schmitt dated
        1/26/84.

IV(12) USDA, Janurary 31, 1984,  Letter from Kenneth A. Giles  to Dr.  R.  D.
       Schmitt.

IV(13) USDA, January 30,  1984 letter  from Ronald  E. Engel, Deputy
       Administrator for  Science to Donald L. Houston, Administrator.

IV(14) USDA, January 23,  1984 letter  from John  W. Marshall  to Dr. John  A.
       Moore,  D.V.M.

-------
                                        32
IV{15)  Berk,  B.  1974.  Fumigant residues of carbon tetrachloride, ethylene
       dichloride and ethylene dibromide in wheat,  flour, bran, middlings and
       bread.  J. Agri. Fd. Chem. 6:977-984.

IV(16)  Wit, S., A. Besemer, H. Das, W. Goedkoop, F. Loosjes, ad E. Meppeling.
       1969.   Results of an investigation on the regression of three fumigants
       (carbon tetrachloride,  ethylene dibronide and ethylene dichloride) in
       wheat during processing to bread.  Report No. 36/39 Tbx., National
       Institute of Public Health.  Bilthoven, Netherlands.

IV(17)  Litton Bionetics, Inc.   1978.  Determination of residues of ethylene
       dibromide in wheat and  wheat products:  Final Report, Project No.
       20927, May, 1978.  In:  Rebuttal Submission #22 & 22E (3000/25), of
       unpublished report, by Ferguson Fumigants, Inc.  June 1, 1978 and Sept.
       22, 1978.

IV(18)  Getz, M.E. and K. R. Hill. 1983.  Ethylene dibrcmide residues in wheat,
       flour, bread, and grapefruit - An analytical method evaluation and
       results.  A report to  the Administrator, ARS, transmitted  to R.
       Johnson, September 15, 1983. 10 pp.

IV(19)  Rains, D.M. and Holder, J.W.; J. Association of Official Analytical
       Chemists, Vol. 64, No. 64, Vol. 5, 1981).

-------
                                    33
V.  Dietary Patterns/Dietary Burdens





A.  Dietary Patterns







The food consumption estimates were prepared using The EPA



Tolerance Assessment System (Chaisson et al.,  1984) and dietary



estimates derived from the most recent nationwide survey of



individuals.  This survey of three day dietary records for



30,770 individuals was conducted by the U.S. Department



of Agriculture in 1977-78.  The USDA survey design is



summarized in section V-D-1.  The methods which were used to



adapt the USDA data to a form comparable to the residue data



described in chapter IV are summarized in section V-D-2.





Briefly, the data file used to estimate EDB dietary



burdens contains the quantity and form of each Raw Agricultural



Commodity (RAC) eaten per unit of body weight/day.  The



record is on an individual basis, and is accompanied by



demographic and soci©economic information for that individual



(age, sex, ethnic origin, region of the country, season



surveyed, pregnant or nursing, etc).  Therefore, subgroups



of the U.S. population can be analyzed for differences in



their dietary profiles, allowing the calculation of differences



in dietary burdens of EDB.  Subgroups analyzed for exposure



to EDB were the U.S. population (48 states, all seasons), 4



regions (northeast, north central, southern, and western), 4



seasons (spring, summer, fall and winter), ethnic  (hispanics,



non-hispanic whites, non-hispanic blacks, and-other), infants

-------
                                      34
 (nursing and non-nursing), children {1-6 and 7-12 yrs),



 teenagers  13-19  (male & female), 20+ (male and female), and



 females who were pregnant or nursing.





 Two analytical procedures are available for estimating exposure



 to EDB.  The first procedure uses values for mean consumption



 of products to estimate the average exposure for the U.S.



 population and for subgroups of the population.  These exposure



 estimates were then used to estimate risks.





 The second procedure generates a set of of exposure



 estimates - one for each respondent in the USDA survey -



 that is, the estimated exposure to EDB for that day for that



 individual.  Individual exposures are then analyzed by subgroup



 to estimate the mean (weighted) exposure and to estimate the



 percentage of the subgroup who exceeded predetermined



 daily intake levels.  This procedure generates a curve which



 describes the distribution of exposure over the population,



 and allows other regulatory options to be compared for their



 impact upon exposure.





This approach provides a mechanism to examine differences in



eating patterns among subgroups of the U.S. population.





Subpopulations were examined for differences in either



quantities or patterns of grain consumption.  Consumption of

-------
                                    35
 the three major grains is summarized in Table V-l for subgroups
 who consumed more grains than the national average or who
 had different patterns of consumption.  Infants and children
 under the age of 12 consume almost twice the national average
 grain consumption - when consumption is expressed per unit
 body weight.  Certain ethnic groups also consume more grains
 than the national average.  The contribution of each grain
 to the diet varies for different subgroups.  Grain consumed
 by the U.S. population is 74% wheat, 10% corn, 8% rice and  8%
 "other" grains.  Hispanics have different patterns, reporting
 56% of their grain consumed as wheat, 17% as rice and 21% as
 corn.  Non-nursing infants (less than 1 yr) have very different
 patterns with grain consumption distributed as 4% barley, 8%
 corn, 21% oats, 37% milled rice and 29% wheat.  Patterns for
 other subgroups are summarized in Table V-l.

 B.  Dietary Burdens
 Exposure is a function of the residue level in each grain
 at the time it is consumed and the quantity of that
 food consumed.  Dietary burden is the sum of the EDB consumed
 in each grain.  Using the second procedure, described above,
 it is possible to analyze each individual's dietary burden.
 For example, non-nursing infants consume more rice than
other grains.  Therefore, the fact that monitoring data has
 found lower levels of EDB in rice than in other grains will
have a greater impact on the dietary burden for non-nursing
infants than it will for the U.S. population.  This is
illustrated by two hypothetical examples: 10 ppb on all grains

-------
                                       36
 Table V-l.   Patterns of Grain Consumption
 SUBGROUP      WHEAT      RICE      CORN     OTHER       ALL GRAINS

                   g food/kg body weight/day{% of grain consumption)
US. population 1.4(73)    0.16(8)    0.19(10)   0.15(8)        1.9
(48 states)

Ethnic groups
Hispanics
Non-hispanic
whites
Non-hispanic
blacks
Other
1.35(56)

1.44(78)

1.26(65)
1.34(55)
0

0

0
0
.41(

.IK

.25(
.78(
17)

6)

13)
32)
0

0

0
0
.49

.14

.35
.18
(21)

(8)

(18)
(7)
0

0

0
0
.14(

.16(

,09(
.15(
6)

9)

5)
6)
2.4

1.9

2.0
2.5
Infants & Children
Non-nursing
infants
1-6 yr.
7-12 yr
Regions
Northeast
North Central
Southern
Western

1.03(29)
b/
b/

1.47(74)
1.45(76)
1.35(71)
1.36(72)

1



0
0
0
0

.32(
b/
b/

.20(
.IK
. 16(
.17(

37)



10)
6)
8)
9)
a/
0



0
0
0
0


.29(8)
b/
b/

.09
.14
.26
.23



(5)
(7)
(13)
(12)

0



0
0
0
0

.90(
b/
b/

.19(
.16(
.IK
.17(

25)



10)
9)
6)
9)

3.5
4.3
3.0

2.0
1.9
1.9
1.9
 a/ All milled rice.
 b/ Not presently available

-------
                                      37
except 1 ppb on rice is:

                                 DIETARY BURDEN
Residue levels       U.S. population        Non-nursing infants
  (ppb  EDB)               rag EDB/KG body weight/day

10 ppb, all grains
  0 ppb other foods      1.9 X 1Q-5              3.5 X KT*

10 ppb all grains except 1.8 X 10~5              1.3 X 10~5
  1 ppb rice, 0 ppb
  other foods
The second analysis procedure, described above, also allows

analysis of the distribution of exposure across individuals in

a subgroup, and it is then possible to answer the question,

"Is the average dietary burden due to a few individuals with a

large exposure or is it due to a large number of individuals

with a low exposure?" The results of monitoring data,  summarized

in Section IV of this document were used to assess the potential

dietary burden for the U.S.  population and for subgroups.

The mean dietary burden for the U.S. population from consumption

of EDB-contaminated grain is 0.55 x 10~5 mg EDB/KG body

weight/day.  Generally dietary burdens parallel grain consumption.

That is, subgroups, such as 1-6 year olds who had higher grain

consumption also had higher dietary burdens of EDB.  However,

the dietary burden is not directly proportional to grain

consumption since different residue levels were estimated for

corn, wheat, rice and other grains (Table IV-10).  Children

(1-6) and hispanics have mean exposures which are greater than

the U.S. population (Table V-2).


Fig. V-l shows the percent of individuals whose dietary

burden is currently greater than the levels of EDB shown on the

-------
                                            38
        V-l •   Effects  of sets of proposed action levels on the percent  of
         individuals  who  exceed each level of EDB:  U.S. POPULATION
c
0)
•a
EH
cfl
-P
^ 60
? c
  vH
CO T3
H OJ
Cd 0)
3 o
T3 X
•H a
 T3

 H
                    current residue levels
Circles numbers from Table VII-1
 ermissible  residue  options
                        X - mg EDB x IcrVKG body  weight/day

-------
                                      39
 Table V-2.
Dietary Burdens for the U.S.  population  and
selected subgroups
                     Mean Exposure             DISTRIBTUION
                    (mg EDB/kg B.W./day)   {Percent  of  Individuals
                                             in Group  exceeding
                                           0.00001  mg/KG  B.W./day)b/
                                            Current    Under
                                                      Recommended
        Current   Under
                 Recommended

U.S. Population
Hispanics
Non-hispanic
white
Non-hispanic
blacks
Other
Nursing infants
Non-nursing infants
Children(l-€ yr)
Children(7-12 yr)
Males(13-19 yr)
Females (13-19 yr)
Males<20+ yr)
Females (20+ yr)
Pregnant (13+ yr)
Nursing (13+ yr)
Spring
Summer
Fall
Winter
Northeast
Northcentral
Southern
Western

.55x10-5
.79x10-5

.52x10-5

.63x10-5
.57x10-5
.26x10-5
.67x10-5
1.28x10-5
.92x10-5
.63x10-5
.50x10-5
.42x10-5
.34x10-5
.40x10-5
.44x10-5
.53x10-5
.54x10-5
.57x10-5
.56x10-5
.51x10-5
.52x10-5
.59x10-5
.52x10-5
Action a/
.34x10-5
.43x10-5

.34x10-5

.36x10-5
.39x10-5
.19x10-5
.45x10-5
.77x10-5
.57x10-5
.40x10-5
.31x10-5
.28x10-5
.22x10-5
.25x10-5
.30x10-5
.34x10-5
.34x10-5
.36x10-5
.36x10-5
.35x10-5
.35x10-5
.35x10-5
.35x10-5

12
25

11

18
12
5
23
62
34
12
5
3
2
3
4
11
12
13
12
10
11
14
14
Action a/
3
3

3

3
5
1
8
21
6
1

_
-
1
3
3
3
4
3
3
3
3
3
a/ Recommended action levels of 900 ppb in whole grains, 150 ppb in
   milled products, and 30 ppb in ready-to-eat products.  Calculations
   assume 100% compliance with recommended levels.
b/ 0.00001 mg/kg B.W./day (approximately 2 times the mean exposure of
   the U.S. population) was arbitrarily selected as an index value
   for examining the proportion of individuals in each group with
   potentially high exposure.

-------
                                     40
X-axis.  Fig. V-l also shows the distribution of exposure
which could be expected to result from 6 possible
sets of permissible residue levels.  Based on the residue
data, currently, 15% of individuals in the U.S.  population
exceed 0.00001 mg/kg body weight/day.  Hispanics have higher
exposure with a mean of 0.0000080 mg/kg body weight/day and
29% of the hispanics surveyed exceeded 0.00001 mg/kg body
weight; 52% of the hispanics exceeded the U.S. population's
average dietary burden.  Children (1-6) have a mean dietary
burden of 0.000013 mg/kg body weight and 70% exceeded 0.00001
mg/kg body weight;  94% exceed the U.S. population's national
average dietary burden.  As average EDB consumption levels
rise, there is a rapid decline in the percent of individuals
with exposures above that EDB level because the general
population has relatively uniform exposures - i.e.  a large
number of individuals with similar consumption patterns.
However, it must be noted that each individual was surveyed
for only 3 days.

EPA has considered 11 possible sets of permissible residue
levels.  It is assumed that as a result of regulatory actions,
all commodities above these levels would be removed from the
food supply and that the average residue level for each type
of grain would decline (Table IV-10).

The proposed regulatory action, assuming complete compliance,
will lower mean exposure for each population subgroup and

-------
                                       41
the percent of individuals within each  subgroup  who  exceed



specified levels of EDB (See Figures  V-l,  V-2, and Table  V-2)



In all subgroups, the lower the permissible  residue  levels,



the lower the dietary burden.

-------
                                        42
  FigV-2 Effect of  sets of proposed  action  levels  on  the  percent  of
      individuals who  exceed  each  level  of  EDB:  CHILDREN (1-6  YRS)

-------
 C.    Summary  of  Procedures Used to Estimate Risk





 The  average levels of  residues in cooked and uncooked



 grain products was determined using all available residue



 data.  Estimates of average residue levels were calculated



 separately for corn, wheat and "other grains".  The category



 "other grains" includes rice, barley, oats, rye, buckwheat



 and  sorghum.  In the determination of average residue levels,



 all  samples with nondetectable levels of EDB (less than 1



 ppb)were conservatively treated as 0.75 ppb.  The mean residue



 level of each product  at the uncooked stage was reduced by a



 factor of 80% to reflect an assumed reduction in residues



 during cooking.  The average residue for both stages (cooked



 and  uncooked) were then combined to yield an overall average



 which uses all of the  available residue data from both cooked



 and  uncooked samples.  In combining this data, the means for



 cooked and uncooked grains were weighted by the number of



 samples in each data set.  Mean residues estimated for each



 set of maximum permissible residue levels are summarized in



 Table IV-10.  Average  residue levels at each level were



 determined by deleting from the data set those samples above



 the  level.  Samples were not gathered in proportion to their



diet.  For example, bread, rolls and biscuits represent 30%

-------
                                    44
and cakes about 10% of the grain consumed by adults (Table



V-3), yet sampling was not proportional.   Also,  the proportion



of each of these food eaten by different  subgroups varies.



For example, ready-to-eat cereals represent from 3-6% of the



grain consumed by individuals over the age of 15, but 60% of



the grain consumed by infants.  Residue has been detected in



only 2 (of 86) baby cereals and both positive samples were



near the limit of detection (1.5 ppb).





In spite of these observations, the monitoring data was



not weighted to account for the nonproportional sampling since



there was a wide variation in the residue levels reported within



product lines as well as across product types.





The estimated mean levels of EDB for each grain type at each



set of maximum permissible residue levels were then entered



into the dietary consumption models (described in Chapter V)



and an estimate of the daily dietary burden was determined



for the U.S.  population and  for each  subgroup.





The estimates of dietary burden were  then used in  the appropriate



equations  (Chapter VI) in order to predict  risk.   Dietary



burdens were assumed  to be constant over  the  2-3  year exposure



period for which  risk has been estimated.

-------
                                        45
   Table V-3.  Distribution of consumption among grain
products.


Age


Male & Female
Less
than 1 yr
1-2
3-5
6-8
Males
15-18,
23-24,
Females
15-18
23-34
Breads,
Rolls,
Biscuits




10
17
25
26

30
32

29
30
Other
Baked
Goods

Percent of


12
15
20
27

25
23

22
23
CEREALS
Cooked
Cereals
Pastas
Total Grain


11
19
21
20

11
13

16
16
, PASTAS
Ready
to Eat
Cereals
Consumed


60
9
9
9

6
3

5
4
MIXTURES
(mainly
grains}
!/



7
40
25
18

27
29

28
27
a/ Include dry infant cereals.
  Adapted from Table 1.3a, Grain Products,  Fats,  Oils;  (USDA,
  1980).

-------
                                46
The procedure used to generate risk  estimates  for  children



(1-6) and to the U.S. population is  summarized below:
MEAN EDB RESIDUE
uncooked
corn
wheat
"other"
80%
Reduction
)
cooked
corn
wheat
"other"

->
DIETARY
MODEL

Individual
Consumption
of each
grain


-?
DIETARY
BURDENS

Mean
Dietary
Burden by
subgroups
& Distri-
bution of
Exposures

-*
RISK

Cancer
risks
due to
dietary
exposure


-------
                                    47
D.  Technical Notes



A detailed description follows of the USDA Food Consumption



Survey and procedures used to convert that survey to equivalent



quantities of raw agricultural commodities.



          1. USDA Survey Design.



Briefly, the USDA 48-state sample survey design consisted of



three levels of stratification.  First, all housing units in



the United States (based on the 1970 Census) were grouped



according to the nine census divisions. Within each division



housing units were then grouped into three zones — central



cities, remaining metropolitan areas, and non-metropolitan



counties.  A third level of stratification was carried out



within each of the 27 Census division/zone combinations.



This was accomplished by forming groups containing approximately



600,000 households based on political/economic similarities



and geographical proximity.  Overall, the three levels of



stratification resulted in a total of 114 strata.  Each



stratum was divided into primary sampling units (PSU) and



one PSU was selected from each stratum with probability



proportional to size.  Each selected PSU was divided into



clusters of 100 or more housing units called "area segments".



A total of 2550 segments were selected with the number of



segments in a PSU being proportional to the size of the



stratum in which the PSU was located.  The selection probability



of an individual segment in a PSU was proportional to the



ratio of the number of units in the segment to the total

-------
                                   48
number of units in the PSU.   Finally,  housing  units were
selected from area segments  with the  sampling  ratio determined
by:  the total number of interviews desired;  the estimated
occupancy rate; the estimated completion rate; and the  increase
in the number of housing units from 1970 to 1977.

To provide seasonal information, four samples, independently
selected from the same set of PSUs, were implemented in
consecutive calendar quarters beginning in April 1977.   The
sampling ratio for each quarterly sample was determined so as
to yield approximately 3750 completed household interviews in
each quarter.

A subsample of household members was selected to complete
a 3-day record (24-hour recall and a  2-day diary) of individual
food items consumed.  For the Spring survey, every household
member was eligible to complete this  record.  In the remaining
three quarterly samples, all persons younger than 19 and one-
half of all persons 19 years and older were eligible.

          2.  EPA's Food Composition Data.
A computer search of the USDA data revealed that 3734 individual
food items — pizza, chocolate cake, etc. — had been
consumed on at least one occasion.  Bach food item was then
examined and a determination made as to the:
    (1) raw agricultural commodities (RACs) — milk, wheat,
        carrots, etc — which make up  the food  item,
    (2) percentage of each RAC, by weight, and
    (3) form of each RAC "as eaten" (cooked,  raw,  fried, etc)

-------
                                   49
    (4) water used in cooking is not included in the estimate
        of the consumption of each RAC, but is retained as a
        separate entry - that is 100 g of cooked rice, would
        be coded as 30 g rice and 70 g water.

Information on how the formulations were derived is given

in White, et al. (1983).  The food composition data was then

merged with the USDA food consumption files to estimate each

individual's RAC consumption which was then adjusted for that

person's self-reported body weight - that is, all consumption

is expressed as g RAC per kg body weight/day.

-------
                                    50
B. References

Chaisson, et al.  Tolerance Assessment I.  A new approach for
estimating public risk due to pesticide residues in food.  In
press, 1984.

White, S.B., et al.  The construction of a raw agricultural
commodity consumption data base.  Interim Report No. 1.   RTP,
April 27, 1983.

USDA, Science and education Admin.  Food and nutrient intakes
of individuals in 1 day in the U.S., Spring, 1977 (Preliminary
Report No. 2).

-------
                                        51
 VI.  Cancer Risks

 As noted in previous position documents (PD 2/3 and PD 4), the Agency has
 concluded that BOB is a potent animal oncogen and a likely human oncogen.

 At dietary levels of exposure that have been, estimated for grain-based
 products, cancer risk must be of primary concern.  Although potential
 reproductive hazards can result from exposure to EDB, the inhalation dose
 levels which have been associated with gonadal toxicity are much greater
 than the observed and estimated dietary levels.  Although EDB may pose a
 risk with respect to heritable mutations, it is not yet possible to
 quantify this risk.

 The evidence supporting the Agency conclusion that EDB is an animal oncogen
 results from studies conducted in two species and several routes of
 exposure.  The initial studies that triggered the RPAR were the NCI gavage
 studies (Hazelton, 1973).   Both the MCI rat and mouse bioassays showed
 significant  increases in several tumor types.  The most common tumor type
 observed was a gastric carcinoma.  By the end of the first year of the
 test,  90% of low dose male rats were diagnosed as having this malignant
 tumor.

Other  studies, conducted via the inhalation route of administration, also
found  EDB to be a potent animal oncogen.   The Midwest Research Institute
under  contract to NIOSH found significant increases in the incidence of
tuners of the  spleen, manniary gland,  adrenals and subcutaneous tissues at
an exposure  level of 20  ppm (Wbng,  1979).   NCI found EDB to be oncogenic to
rats or mice exposed to  dose  levels of 10  or  40  ppm via inhalation (NCI,

-------
                                        52
I960).  An  increased  incidence  of  hemangiosarconas (a cancer of the



circulatory system) was  found in both  rats  and mice.   In addition,  a highly



significant increase  in  tumors  of  the  respiratory system (nasal cavity



tumors  in rats and lung  tumors  in  mice) were observed.   Finally,  a  skin



painting study (Van Duuren,  1979)  found that repeated painting of EDB on



the skin of mice  induced not only  skin tumors but lung  tumors as well.








In summary,  EDB has shown to be oncogenic  in two species (rats and  mice)



in both sexes and via three  routes of  administration  (gavage, inhalation



and skin painting).







Quantitative risk estimates  in  general contain a great deal of uncertainty



because they must extrapolate from laboratory animals to humans and fron



the very high exposures  used in the  laboratory studies to the much  lower



human exposures.  The estimates of risk from dietary  exposure to EDB



require yet another extrapolation—from partial  to full lifetime



risk—because the exposed animals  in the laboratory study did not live for



more than about half  their normal  lifespan. This additional extrapolation



step introduces yet more uncertainty into  the process.   In addition, the



only oral studies available  administered EDB by  gavage rather than  through



the diet.   The use of gavage data  required  another step to correct  for



gavage vs dietary routes, utilizing  data  fron a  different chemical.







Much care and effort  has been put  into the  quantitative extrapolation



procedure to ensure that all the assumptions  and extrapolations are handled



in the most  appropriate  way. Nonetheless,  there is unavoidable uncertainty



remaining in the estimates and  they  are appropriately taken as only rough



indications  of potential risk.

-------
                                          53
 The  cancer model used in the PD 4 to extrapolate risk from the NCI



 bioassays,  has been developed by the EPA Carcinogen Assessment Group



 (GAG).   The risk model used in the PD 4 to estimate risks fron dietary



 exposure incorporated a tine parameter which allowed the differential



 risks  for less than lifetime exposure for different age groups to be



 estimated fron the  data in the oral (gavage) study.  This was possible



 because  of the extraordinary rate of unscheduled deaths with tumor which



 occurred in the study.  CAG has concluded that "the model fits the 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 1013)  x d x fy,1/3 [(t-s)7'6-(t-f)7-6]



                 W = 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 fran age s to f



                 t = age at end of observation period (years).



   For a lifetime exposure,  V^ = 60 kg., s = 0, f = t = 70 yrs. of



   age,  and  the above equation has the form p(d) = 1 -e41d.





Earlier,  before the PD  4 model was developed,  the Agency used a simpler model



which did not utilize this time-to-death-with tumor information.  Instead,



this model made the simplistic assumption that risks from less-than-lifetline

-------
                                          54
exposures are directly proportional to the length of the exposure, regardless
of the aye at which exposure occurs.  This model takes the form:
             P(d) = i-e-18.71d
   where:    d = average ing/kg/day exposure over a 70-year lifespan

Both of these models were used to generate the range of risks in Table VI-1.

-------
                                            55
     Table VE-1.  Career Risks of Alternative Permissible Reside Options
                          fa: HB  in Grain Products
Permissible
Residue
E"
EBa/
Exposure
(EstinetBd A±ual
Intake)
HB in
Grain Diet
Mg/Kg/tay
Child. Mult
	 Canoer Risk c/
(2 to 3-Years Bqpbsure) d/
Child
(lyr.)
ASults
(30 yr.)
1
2
3
4
5
6
7
8
9
10
lit/
Current
Bqposure
50/250/
5,000
30/150/
3,000
30/150A003
30/150/900
20/100/2,000
10/50/1,000
5/25/500
1/5/100
<1/1/LOO
SlldlTQ 9ucQj6
*
2.9
2.1
1.8
1.8
1.8
1.6
1.1
.8
.5
«
1.2
ftwerage
a/ RarmissibLe residue options
1.3x106 iOH*-icr* ^(pv-icn
6x10-6
9x10-6 IQ-S-ICT4 MT6
4x10-6
8x10-6 10^-10^ 10T^
3x10-6
8x10-6 10-^rlOr^ 10"^
3x10-6
8x10-6 lQr6-10r4 10"^
3x10-6
Txio-6 lor6rior^ icrT-icr^
3x10-6
5x10-6 lO^slQ-^ 10"7-10~^
2x10-6
3x10-6 lQ"*MLQr5 10~7-icr^
2x10-6
2x10-6 KT^lCrS lD-7
9x10-7
<2xlO-6 aO"5 <10~7
0x10-7
5x10-6 ]0~^rlOT^ 10~7-1CI~6
2x10-6
lighBri are Br ready-tb^eat corisunEr products, milled
c/
d/
fton the finished pxxJuct 'to the inoooted prriict and a 20-fbld increase fran
unctxjked cmtrt to grain (with tha esoepticns of .#4, #5, vhidi assure a 6-7-fiald
increase fcon inoooted pxxiuct to grain) .
JermissitQe residue optioi 11 assures 6 norths enftxosnent at 20/100/2,000 followed
ty peonanent  enfeEoamat of 10/50/1 f 000 ppb.
f&vp in canoer risk using dietary career models presented in the H> 2/3 and ED 4.
A constant apoajre ves assured 'fir calculation  purposes.  It is predicted that
exposure will actually decline after the suspension is in effect.

-------
                                       56
References

 1.  Hazelton Laboratories.   1978.   Technical  Background  Information Report
     on Carcinogenesis Bioassay of  1,  2-Dibronoethane  (EDB).  US-DHEW, NIM,
     NCI.  11-14-78.

 2.  National Cancer Institute.  1978.   Bioassay of  1,  2-Dibroraoethane for
     Possible Carcinogenicity.   TR-86  (CAS No. 106-93-4).  Carcinogenesis
     Testing Program.  DHEW Publication No. 78:1336

 3.  Van Duuren, B.L. et al.  1979. Carcinogenicity of Halogenated
     Olefinic and Aliphatic  Hydrocarbons in Mice.  J.  Nat. Cancer Institute
     63(6):1433-39.

-------
                                             57

              VII.  Toxicity of Likely Chemical Alternatives

A.  Aluminum Phosphide

Aluminum phosphide is formulated as tablets or pellets which liberate
highly toxic phosphine gas when the aluminum phosphide reacts with moisture
in the commodity or surrounding air.  Because of the highly toxic nature of
the phosphine which is liberated, formulations of aluminum phosphide have
also been considered as highly toxic materials  (EPA, 1981).

A two-year rat feeding study examined the effects of feeding diets
fumigated with phosphine gas to rats (Hackenberg, 1972).  Analytically
determined levels of phosphine in the diet ranged from 0.2 ppm to 7.5 ppm
and averaged approximately 1 ppm.  Ho differences were found between
control animals and those animals receiving treated diet with respect to
growth, food consumption, survival, morbidity,  hematology, blood chemistry,
gross and microscopic pathology.

B.  Ethylene Bichloride

Ethylene dichloride (1,2-dichloroethane) has a  relatively low toxicity  to
mammals on an acute basis (LC5Q = 12,000 ppm. or 48 mg/L for 0.53 hour
of exposure).  The primary effects of human exposure to ethylene dichloride
were reported to be central nervous system  (CNS) depression and
gastrointestinal upset.  Liver, kidney and adrenal injuries occurred  in a
dose-related fashion.

-------
                                           58
This chemical is rapidly metabolized and eliminated  in mammals.  The
metabolic pathway of EEC is not adequately delineated, but very  likely
occurs via a dechlorination reaction.  The metabolism of  EDC may be
catalyzed by the glutathione system in the soluble fraction of the liver.
EDC metabolities include monochloroethanol and monochloroacetic  acid which
are more toxic than EDC.

The chemical was shown  to be mutagenic in  the reverse mutation assay  in
Salmonella typhimurium  TA1535 and TA100 with and  without  metabolic
activiation  (McCann, 1975).  Metabolities  of EDC  are also known  to be
mutagenic (McCann, 1975).

There  is evidence that  exposure  to ethylene dichloride  poses  a carcinogenic
risk to man.  An NCI study showed statistically significant  increase of
tumors in both rats and mice after  long-term oral intubation (Dhew,  1974).
A statistically significant positive association  between  the  dosage  and the
incidence of squamous cell carinomas of  the  forestomach and
hemangiosarccnias of the circulatory system occurred  in male rats but not in
females.  There was also a significantly increased incidence of
adenocarcinonas of the  mammary glands  in female rats.   The incidence of
mammary adenocarcinoma  in female mice  were statistically significant.  The
EPA Carcinogen Assessment Group  concluded that  EDC was 50 times less potent
as an  oncogen than EDB  (Albert,  1978).

C.  Carbon Tetrachloride

CC14 was  found to  be  cncogenic  (liver tumors)  in rats, mice, and hamsters
after  oral administration.  Other chronic effects observed in laboratory
animals  included cirrhosis of  the liver after oral, subcutaneous, or
inhalation exposure of  rabbits,  rats,  mice,  and hamsters; severe liver

-------
                                        59
 damage observed in humans after chronic exposure; kidney damage in humans
 and rats (nephrosis and degenerative changes).

 Fetotoxic effects have been observed by Schwetz,  et al (1974)  in rats
 exposed bo CC14 vapor.  Bhattacharyya (1963)  also reported fetal as well
 as  neonatal effects in rats exposed via inhalation.  Testiticular atrophy
 and impaired spermatogenesis have also been observed in male rats which
 were given intraperitoneal doses of CC14 (Chatterjee, 1966:  and Kalla and
 Bansal,  1975).

 No  genotoxic activity has been reported in bacteria (McCann et al, 1975),
 yeast (Callen et al,  1980),  or rat liver cells  ir\ vitro (Dean and
 Hudson-Walker,  1979).   In vitro experiments indicate (£14 may also bind
 to  other macronolecules which could be the reason many microbial or jri
 vitro assays vgere negative (Rocchi, 1973,  Uenlejce, 1976).

 D.   Methyl Bromide

Methyl bromide  has not been well studied toxicologically.  It is currently
being  tested  for oncogenicity by the Dutch government and the National
Toxicology Program.  Three in vitro mutagenicity  studies indicated a
mutagenic  potential for methyl bromide with mutagenic effects shown in a
dose dependent manner.  Test systems used  were  Salmonella, the yeast
Saccharomyes cerevisiae and  Chinese hamster ovary cells.   A 3-generation
reproduction study via  the inhalation route is  being sponsored by several

-------
                                         60
manufacturers of methyl bromide.  Teratogenicity  studies of methyl  bromide
(in the rat and mouse) have been submitted  to the Agency.  Although they
have not yet been evaluated, the authors of the study  report  no  teratogenic
effect at the highest dose level administered.

E.  Carbon Disulfide

Carbon disulfide is a volatile and extremely flamable  liquid.

Chronic exposure to carbon disulfide at airborne  concentrations  greater
than 30 mg/m^ has been associated with diseases of  the cardiovascular  and
nervous system.  Acute exposure to carbon disulfide at airborne
concentrations greater than 60 mg/m^ has been associated with
physiological and behavioral abnormalities.  The  chemical was also  reported
to cause a marked decrease in visual acuity as a  result of an effect on the
optic nerve (Potts, 1980).

In experimental animals, subchronic exposure to concentrations greater than
50 ppn caused brain weight depression accompanied by histopatholgoical
changes in the nervous system in rats.  In  addition, compound-related  renal
lesions and pigmentation of the spleen were also  reported  in  mice.

-------
                                          61


F.  REFERENCES

 1.  Albert, R.E.  The Carcinogen Assessment Group's Preliminary Report on
     Ethylene Dichloride.  April 7, 1978.

 2.  Bhattacharyya, K.  Fetal and Neonatal Responses to Hepatotoxic Agents.
     J. Pathol. Bact. 90:151-161.

 3.  Chatterjee, A.  Testicular Degeneration in Rats by Carbon Tetrachloride
     Intoxication.  Experimentation.  22:395-396, 1966.

 4.  Dean, B. J., and G. Hudson-Walker.  1979.  An in virto Chromosome assay
     using cultured rat-liver cells. Mutat. Res.  64:329.

 5.  Department of Health, Education and Welfare.  Bioassay of 1,
     2-Dichloroethane for Possible Carcinogenicity.  Public Health Service
     National Institute of Health, National Cancer Institute DHEW
     Publication No. 78-1305 (1974).

 6.  Environmental Protection Agency.  1981.  Pesticide Registration
     Standard.  Aluminum Phosphide.

 7.  Hazelton Laboratories.  1978.  Technical Background Information Report
     on Carcinogenesis Bioassay of 1, 2-Dibromoethane (EDB).  US-DHEW, NIM,
     NCI.  11-14-78.

 8.  Hackenberg, U.  (1972).  Chronic Ingestion by Rats of Standard Diet
     Treated with Aluminum Phosphide.  Tox. App. Pharm. 23(1):  147-158.

 9.  Kalla, N. R., and M. P. Bansal.  1975.  Effect of carbon tetrachloride
     on gonadal physiology in male rats.  Acta Anta. 91:380-385.

10.  MeCann, J., E. Choi., E. Yamasaki, and B. N. Anes.  1975.  Detection of
     carcinogens as mutagens in the Salmonella/microsome test: Assay of 300
     chemicals.  Proc. Natl. Acad. Sci. 72:5135-5139.

11.  McCann, J., Siimion, V., Stretiwieser, D., and Ames, B.N.  "Mutagenicity
     of Chloroacetaldeyde.  A Possible Metabolic Product of 1,
     2-Dichloroethane (Ethylene Dichloride), Chloroethanol, Vinyl Chloride
     and Cyclophosphomide" Proceedings of the National Academy of Science,
     U.S.A. 72: 3190-3193, (1975).

12.  National Cancer Institute.  1978.  Bioassay of 1, 2-Dibrcmoethane for
     Possible Carcinogenicity.  TR-86 (CAS No. 106-93-4).  Carcinogenesis
     Testing Program.  DHEW Publication No. 78:1336

13.  Potts, A.M., Gonasun, L.M. 1980. Toxic Responses of the Eye.  In
     Casarett and Doull's Toxicology, Second Edition, Macmillan Publishing
     Co.,  Inc., New York,  p. 301.

-------
                                         62

14.  Rocchi, P., G. Pordi, S. Grilli, and A. M. Ferreri.  1973.  In vitro
     and in vivo binding of 0014 with nucleic acids and proteins in rat
     and mouse liver.  Int. J. Can. 11:419-425.

15.  Rosenkran, S., Carr, H.S., and Rosenkranz, H.S.  2-Haloethanols:
     Mutagenicity and Reactivity with DNA.  Mutation Research, 26 367-370
     (1974).

16.  Schwetz, B. A., B. K. J. Leong, and P. J. Gehring.  1974.  Embryo- and
     fetotoxicity of inhaled carbon tetrachloride, 1, 1-dichloroetnane, and
     methyl ethyl ketone in rats.  Tbxicol.  Appl. Pharmacol. 28:452-464.

17.  Uenleke, H., H. Greim, M. Kramer, and T. Werner.  1976.  Covalent
     binding of haloalkanes to liver constituents but absence of
     mutagenicity on bacteria in a metabolizing test system.  Mutat. Res.
     38:114.

18.  Van Duuren, B.L. et al.  1979.  Carcinogenicity of Halcgenated Olefinic
     and Aliphatic Hydrocarbons in Mice.  J. Nat. Cancer Institute
     63(6):1433-39.

19.  Wong, L.C.K.  1979.  Study of Carcinogenicity and Toxicity of Inhaled
     1, 2-Dibronoethane in Rats Treated with Disulfiram.  Midwest Research
     Institute Final Report, NIOSH Contract No. 210-76-0131.  2-8-79.

-------
                                         63

 VIII.  Economic Impacts

 A.  Introduction

 The purpose of this section is to provide information on economic
 impacts and implications of implementation of national maximum
 permissible residue levels for EDB in grain,  flour and related finished
 or consumer products and economic impacts of  emergency suspension of
 grain fumigation and spot-treatment uses of EDB.   Specifically, this
 section focuses ugpn alternative sets of maximum  residue levels for
 grain and related products which may contain  EDB  residues from either
 treatment of grain or spotrfumigation of milling  machinery.   Economic
 impacts of emergency suspension are evaluated for 6 months and for the
 remainder of the cancellation proceedings already in  process.  The
 cancellation hearings are expected to require about two years for
 completion.

 This  report  is intended  to meet the requirements  for  Regulatory Impact
 Analysis  as  established  by Executive Order 12291, the Regulatory
 Flexibility  Act and Sectiqn 25 of FIFRA.   It  also serves as  an input for
 preparing any analysis required under the Paperwork Reduction Act of
 1980.

 Executive Order 12291  require^ that Adequate  information concerning the
 need  for,  and consequences  of,  a proposed action  be presented.  The
Order requires that j  finding  of potential benefits to society will
outweigh  the  potential cost for any reccranended action.  Executive Order
12291 also recognizes  that  legal constraints  may  play a role in

-------
                                          64
selecting  among  alternative  approaches  to achieving  regulatory goals.
The Regulatory Flexibility Act  requires that  regulations  take  special
note of  the  impact of  proposed  regulations on small  entities.   The
Regulatory Flexibility Act requirements can and  should  be combined with
Executive  Order  12291  requirements.   FIFRA, Section  25, requires that
the Administrator of EPA consider such  factors as  the effects  of
regulation on production and prices of  agricultural  ooranodities, retail
food prices and  otherwise  on the agricultural economy when issuing
regulations affecting  pesticides.  These three mandates for
consideration of economic  aspects of  pesticide regulatory matters are
applicable to the revocation of the present exemption and tolerances
governing  EDB under the  Federal Food, Drug and Cosmetic Act, but not to
the emergency suspension,  which is not  rulemaking.

As noted above,  this analysis focuses on alternative options for
issuance of maximum permissible levels  for EDB in  food/feed grain
products.  The specific  options range frcm less  than one  part  per
billion (ppb) to 100 ppb in  final consumer products  (Table VTII-1).  For
each option at the final consumer product level  (ready-to-eat  breads,
cereals, etc.),  there  are  corresponding levels identified for
processed/uncooked foods (e.g., milled  flour) and  for raw grain.  The
alternatives were selected at the final consumer product  level to
reflect a  range of levels  which have  differing exposure/risk

-------
                                         65
 implications for human health.  Given those values, the other
 permissible residue levels for processed/uncooked foods and raw grain
 were assumed to have fixed relationships with the ready-to-eat products.
 Obviously, actual levels of EDB residues at the three stages of the
 marketing system would vary considerably.  However, those contained in
 Table VIII-1 are internally consistent within options and are thought to
 be reasonable approximations.  A separate option for raw grain of 900
 ppb was also considered but is not shown in-Table VIII-1.

 This analysis is not highly quantitative due to the complex nature of
 the problem and also due to the short time available for preparing this
 analysis.  Quantitative estimates of usage and economic impacts are
 included whenever possible.  However, in many cases only qualitative
 estimates are presented'.  The quantitative estimates- should best be
 considered as illustrative of. possible economic effects rather than
 values with any known probability of occurrence.

 B.   Economic Impacts of Setting Maximum Permissible EDB Levels for Raw
     Grain

        1.   Usage of  EDB and Other Fumigants

 EDB  is  used  as a liquid grain fumigant in mixtures whose major
 constituents are carbon tetrachloride/carbon disulfide and carbon
 tetrachloride/ethyl-ene  dichloride.   These funigants are used as remedial
 treatments to control  insects in stored grain.  EDB is also used in
combination  with  chloroform and carbon disulfide, but the overall volume
of this product  is small  in comparison to the carbon tetrachloride-based

-------
                                 66
Table VIII-1.   Maximum Permissible Levels of EDB in Grain


Option


1
2
3
4
5
6
7
8
9
10


Ready- to-Eat


100
50
30
20
10
5
2
1
C
20AO
and Related Products

Prepared/Uncooked


500
250
150
100
50
25
10
5
1
100/50


Raw Grain


10,000
5,000
3,000
2,000
1,000
500
200
100
100
2,000/1,000

-------
                                        67



 products.  An estimated 425,000 gallons of formulated products



 containing 170,400 pounds of EDB active ingredient were used for insect



 control in grain storage facilities in 1983 (Table VIII-2).  At typical



 application rates, approximately 202 million bushels of grain stored in



 1983 were treated with fumigants containing EDB (Table VIII-2).







 The majority of EDB usage in the U.S. is to control stored product



 insects of wheat.  Using an application rate of 2 gal. per 1,000 bu., an



 estimated 159.4 million bushels of wheat were treated in 1983.  This



 quantity of wheat treated with EDB represents about 5.4 percent (Table



 VIII-3)  of the wheat in storage on October 1,  1983 (USDA, Crop Reporting



 Board,  SRS.,  1983. pp. 3).  In 1983, approximately 14 million bushels of



 corn were treated with 42,500 gallons of EDB-containing liquid grain



 fumigants assuming an application rate of 3 gallons per 1,000 bushels.



 This amount of corn treated with EDB-containing liquid grain fumigants



 represents about 0.5 percent of the corn in storage on October 1, 1983.



 The usage  on  oats,  barley,  rye and rice comprised approximately 10



 percent  of the EDB usage in 1983.  Crop-specific data for these



 canned ities are not available, but overall,  approximately 21 million



 bushels  were  estimated to have been treated in 1983.  The remaining



 usage of EDB  was to treat approximately 7 million bushels of grain



 sorghum, which would equal  1.8 percent of that commodity in storage on



 October  1,  1983.







 While actual  usage  statistics are not available to indicate the precise



 locations of  EDB usage,  it  is believed that  approximately 65 percent of



 the annual EDB  usage is  on  farms and the remaining 35 percent used in



 all off-farm  positions.   According  to Douglas  Chemical Co. data,



Minnesota, North  Dakota,  Iowa,  South Dakota, Nebraska, Oregon and



 Indiana were states which received  the bulk (60.8 percent) of Douglas1



1983 distribution of EDB containing products (Table VIII-4).

-------
                                              68
                     Table VIII-2.  Estimated Usage of EPS-Containing
Liquid Grain Fumigants by Crop, 1983

Stored Commodity

Wheat
Corn
Oats, Barley, Rye, Rice
Grain Sorghum
Total

Percent of
EDB
Usage

75
10
10
5
100.0

Pounds
A.I.
Applied

127,800
17,040
17,040
8,520
170,400

Gallons
Applied

318,750
42,500
42,500
21,250
425,000

Application
Rate
(gal/1,000 bu)
2
3
2
3
N/A

Bushels
Treated
(000)
159,375
14,167
21,250
7,083
201,875
Source:  EPA Estimates;  Douglas Chemical Co.,  1984.;  Harein and Sumner, 1982.

-------
                                           69
  Table VIII-3.  Estimated BOB Treated Grain as a Percentae of Grain Stocks
On-Farms and All Stock Positions, Various Dates, 1983

Commodity Bu. Treated

Wheat 159,375,000



Corn 14,167,000



Sorghum 7,023,000




On-Farm

.Jan 1
Apr 1
Jun 1
Opt 1
Jan 1
Apr 1
Jun 1
Oct 1
Jan 1
Apr 1
Jun 1
Oct 1
Treated
Storage

13.7
18.0
22.9
12.9
6.23
0.3
.0,4
0,9
2.6
4.7
7>.4
12.0
as Percentage of
All Positions
'Percent 	
6.3
8.5
10.3
5.4
0.2
0.2
0.3
0.5
0.9
1.1
1.3
1.8
Sources:  USDA.  Grain stocks,  .October 1983; and Table VI11-2.

-------
                                             70
             Table VIII-4.  Distribution of Douglas Chemical Co.
Liquid Grain
Fumigants Containing
BOB by State, 1983

State
MN
ND
IA
SD
NE
OR
IN
IL
MT
AR
OH
MO
CO
PA
VA
GA
MI
NY
WI
NC
AL
KS
WY
TX
Subtotal
Misc. Dealer Sales
Total
Gallons
55,602
27,240
27,098
25,660
24,235
21,996
20,555
15,164
14,859
12,219
11,535
8,141
5,610
5,082
3,635
3,414
2,755
2,312
2,247
1,938
965
335
251
106
292,954
20,046
313,000
Percent of Total
17.8
8.7
8.2
7.7
7.0
6.6
4.8
4.7
3.9
3.7
2.6
1.8
1.6
1.2
1.1
0.9
0.7
0.7
0.6
0.3
0.1
0.1
0.1
<.l
93.6
6.4
100.0
NOTES:  Total Ibs. of EDB = 78,953 (Douglas products only).
        Lbs. of EDB per gal. = 0.252 (Douglas products only).
        Estimated total Ibs. EDB on all formulations = 170,383.
Source:  W.P. Conrath, Vice President, Douglas Chemical Company.  January 18,
         1984.

-------
                                          71




 According to USDA, the most commonly used fumigation method  in small



 farm-type bins is the application of liquid fumigants in a coarse spray



 over the grain surface (USDA, 1979).  This method relies on  gravity to



 distribute the fumigant down through the grain mass as the



 heavier-than-air vapors change from the liquid form.  Pre-application



 procedures include leveling the grain surface and sealing the juncture



 between the sidewall and roof with masking tape.  Fumigants  are usually



 applied using a bucket pump or garden sprayer adapted to produce a



 coarse spray.  Fumigants are normally applied from outside the bin.



 After treatment,  the access door and fillhole lid are sealed with



 masking tape or plastic sheeting to prevent fumigant vapors  fron



 escaping to the outside.







        2.   Profile







 Grain production  is one of the major agricultural sectors in the U.S.



 In 1983, a total  of 8.281  billion bushels of major grain crops were



 produced.   The total supply (production,  beginning inventory and



 imports) for  the  1983/84 crop year was 14.006 billion bushels (Table



 VIII-5  ).   Several hundred thousand farms are engaged in production of



 major grain crops.   Nearly 900,000 farms  produce corn and more than



 380,000 produce wheat.  Other crops involve fewer farms.







 There were a  total of more  than  1 million grain farms in the U.S.  in



 1978; 589,480  had sales exceeding $20,000 (U.S.  Dept. of Commerce,



 1981).  While  actual data are  not available,  the average quantity of



 commodities stored on farms in bulk on January 1, 1980-1982 could range



between 7,576  and 13,497 bushels  per farm (Table VIII-6).  The estimated



average storage capacity per  farm could range between 10,956 and 19,518



bushels.

-------
                                    72
Table VIII-5.  Economic Profile  Information on U.S. Grain Production


Grain

Wheat
Corn
Oats
Barley
Rye
Rice
Sorghum
Total
Sources:




Total Production
1983

2,425
4,121
473
532
28
220
482
8,281
Grain Stocks, USDA, Jan. 23
Feed Outlook and Situation
U.S. Census of Agriculture,


Total Supply
1983/84

3,970
7,262
715
765
35
378
881
14,006
, 1984
Report, USDA, Nov., 1983
1978, Census Bureau


Stocks
1A/84

2,325
4,928
378
378
20
123
651
8,803




No. Farms
(Harvesting)
1978
tnnn\
\\f
-------
                                             73

       Table VIII-6.  Estimated Quantity of Commodity Stored on Farms
January 1,

Stock Position
Date
January 1
April 1
June 1
October 1
April 1, June 1,

Quantity
Stored
7,956,000
5,242,000
3,575,000
3,071,000
and October 1, 1980-8:

Estimated. I
Grain Stored
"ATI "Farms b/ ~
	 *• -Busfiels^ =•- ••
7,576
4,992
3,404
2,924
I Average a/

\ye.rage
Pec Farm
"Farms "with
sales >$ 2 0,000 c/
13,497
8,893
6,065
5,210
a/  Includes corn, sorghum, oats, barley, wheat, rye, flax seed, soybeans,
    and rice.

b/  Quantity stored divided by the 1,050,178 grain farms in the U.S. (1978
    Census of Agriculture).  Assumes all farms have equal storage capacity.

c/  Quantity stored divided by the 589,480 grain farms in fche U,S. with
    sales greater than $20,000.  Assumes .all grain .storage capacity is
    concentrated on larger farms.

-------
                                        74
Circular  steel  bins  are  the most  caramon structures used  for on-farm
storage.  While precise  data  are  not  available,  it is estimated that
steel  bins  comprise  90-95  percent of  long-term on-farm grain storage.
There  are still seme structures built of  wood  which are  generally used
for  storage of  bumper crops.   Newer structures (i.e., circular steel
bins)  usually have aeration systems while older structures (wooden
construction) are often  not so equipped.   Most modern structures can be
treated with any of  the  common grain  fumigants.   The exception would be
crib-type structures.

Off-farm  storage includes  all forms of grain containers  used to hold,
handle, or  transport grain frcm the time  the grain leaves  the fam until
it is  either sold to processors or is exported.   This definition
includes, but is not necessarily  limited  to, country elevators, terminal
elevators,  port facilities, railroad  cars, barges, trucks, and so forth.

There  are over  15,000 off-farm grain  handling  and storage  facilities in
the U.S.  with nearly 7.5 billion  bushels  of capacity. Commercial
facilities  in the Corn Belt,  Northern and Southern Plains  comprise about
two-thirds  of this capacity.   The Lake, Delta  and Pacific  States
ccroprise  an additional 20  percent of  capacity.   The remaining capacity
is nearly uniformly  distributed throughout the remaining production
regions.  January 1  grain  storage stocks  (1979-81) averaged 4.3 billion
bushels or  57 percent of capacity.  Corn  and wheat doninate grain
storage with 43  and  26 percent of total volume,  respectively.  Soybeans
account for an  additional  19  percent  of total  storage volume.

-------
                                         75





        3.  Quantities Exceeding Maximum Permissible Residue Levels








 A key aspect of evaluating economic impacts of maximum permissible



 residue levels is estimation of the quantities of grain that would



 exceed such specified levels.  In this case, the quantities have been



 estimated, as a percentage of stocks as of February 1, 1984 that would be



 used for human food consumption or export.  The procedure used was as



 follows:







            a.  Adjust 1/1/84 stocks for pro-rata share which would not



 be  used for U.S.  human consumption or export,  i.e. feed/seed uses.  A



 percentage was estimated to be used for feed/seed and was subtracted



 fron the total 1/1/84 stocks, based upon projected 1982/83 usage



 (1983/84 not available).







            b.  Stocks for food/export use as of 1/1/84 were reduced by



 8.3 percent to approximate stocks as of 2/1/84, due to assumed dispersal



 during January, 1984.







            c.  Grain stocks estimated to exceed maximum permissible



 levels of  EDB were  based upon a statistical analysis of grain residues



 submitted  by GMA/states/USDA.   This statistical analyses was performed



by EPA staff (OPPE).







            d.  Qiantities exceeding maximum permissible levels were



computed by multiplying  the percentages of grain samples exceeding



permissible residue  levels by food/export stocks as of 2/1/84.

-------
                                          76
 The results of the estimation process are presented in Tables VIII-7 and
 VIII-8 for each maximum permissible residue level.  The grain stocks
 estimated to exceed the maximum permissible residue levels ranged frcm a
 low of 0.2 percent exceeding 5,000 ppb (9.5 million bushels) to a high of
 6.3 percent at 100 ppb (250 million bushels).

        4.  Impact Analysis

            a.  Direct Impacts

 One of the potential economic impacts from enforcing maximum permisssible
 residue levels is that grain could be withdrawn fron the market.  Such
 withdrawals could be made of grain from storage at the farm level or from
 storage in ccmnercial channels following marketing frcm the farm.  In this
 analysis,  estimates are made of the aggregate value of bushels that may be
 lost from such eventualities.   The overall aggregate value of all types of
 grain  exceeding permissible residue levels is shown in Table VIII-9.
 Also,  the  approximate dollar values of typical sized lots of grain that
might  be  involved are summarized  in Table VIII-10.  A typical lot of grain
 at  the  farm level could range  from 6,000 to 13,000 bushels compared with
 50,000  to  500,000 bushels or more in commercial channels.  (Some shipments
of grain are  in the millions of bushels.)   The value of a 10,000 bushel
lot of grain  at the  farm level  ranges frcm about $17,000 to nearly $40,000
 (Table VIII-10).   The values for  comercial  lots of grain would range from
about $427,000  for oats to nearly a million  dollars for rice.

-------
                                             77
               Table VIII-7.  Estimated Quantities, of Grain. Stocks- Available


Grain

Wheat
Corn
Oats
Barley
Rye
Rice
Sorghum.
Total


Stocks:

mi J. •- oil'*
2*325-
4,928;
3:7.8^
3,78;
20?
123;
652
8',.803,
for Food, or Export

Peizsent. for-
Eood^Export
01983/8*):
rv*l-
»
3:7?
8
43=
15
9»
30;
50i
,. February 1,.

Stocks:
15/1/841 f or:
Eood/Export:

W
i^a
»
163!
3'
L20)
MS
41.3&7?
1984:

Stocks^
Used!
Janx 84

169
152
3
14
-
10
16
364


Stocks 2/1/84
for Food/
Export

1,855
1,671
27
149
3
110
179
3,993
Source:  See* Table VIII-5

-------
                                          78
Table VIII-8.  Quantities of Grain Estimated to Exceed Maximum Permissible Residue
Levels for EDB, U.S., February 1, 1984










Permissible Residue
Level
No.

1
2
3
4
5
6
7
8
9
10
11

1
2
3
4
5
6
7
8
9
10
11 2
Base

PPB

None
5,000
3,000
2,000
1,000
900
500
200
100
100
2,000/1,000

None
5,000
3,000
2,000
1,000
900
500
200
100
100
,000/1,000
( food/export
stocks 2A/84)
Wheat

Corn

Oats

Barley

Rye

Rice

Sorghum

Total


_
0
0
0
0.5
0.5
1.4
NA
9.2
9.2
0/.5


0
0
0
0
9
9
26
NA
171
171
0/9
1,855

_
0.3
0.3
0.3
0.3
0.7
1.4
NA
3.8
3.8
.3/.3


0
5
5
5
5
12
23
NA
63
63
3/7
1.671

_
0.9
0.9
0.9
1.8
1.8
1.8
NA
3.6
3.6
.9/1.8


0
0.2
0.2
0.2
0.5
0.5
0.5
NA
1.0
1.0
0.1/0.6
27

_
0.9
0.9
0.9
1.8
1.8
1.8
NA
3.6
3.6
.9/1.8
- -Mil.
0
1.3
1.3
1.3
2.6
2.6
2.6
NA
5.1
5.1
0.7/3.2
149

_
0.9
0.9
0.9
1.8
1.8
1.8
NA
3.6
3.6
.9/1.8
bushels-
0
-
-
-
-
-
-
NA
0.1
0.1
-
3

_
0.9
0.9
0.9
1.8
1.8
1.8
NA
3.6
3.6
.9/1.8


0
1
1
1
2
2
2
NA
4
4
0.5/2.5
110

_
0.9
0.9
0.9
1.8
1.8
1.8
NA
3.6
3.6
.9A-8


0
2
2
2
3
3
3
NA
6
6
1/4
179

_
0.2
0.2
0.2
0.6
0.7
1.4
NA
6.3
6.3
,2/.6


0
9.5
9.5
9.5
22.1
29.1
57.1
NA
250.0
250.0
5.3/17.3
3,993


-------
                                                79
              Table VI11-9.  Value of Grain Exceeding EPS Maximum Permissible
Residue Levels,





February 1, 1984





Action Level
No.

1
2
3
4
5
6
7
8
9
10
11
PPB

None
5,000
3,000
2,000
1,000
900
500
200
100
100
2,000/1,000
Price/bushel
Wheat

$ 0
0
0
0
32
32
92
NA
602
602
32
$3.52
Corn

§ 0
16
16
16
16
40
76
NA
208
208
16/24
3.30
Oats

$ 0
0.3
0.3
0.3
0.9
0.9
0.9
NA
1.7
1.7
0.2/1.1
1.71
Barley

Rye

$ 0 $ 0
3
3
3
7
7
7
NA
13
13
2/9
2.57
-
-
-
-
-
-
NA
0.3
0.3
0/0
2.75
Rice

$ 0
4
4
4
8
8
8
NA
16
16
2/10
3.99
Sorghum

$ 0
6
6
6
9
9
9
NA
17
17
4/12
2.85
Total

$ 0
26.3
26,3
26.3
72.9
96.9
192.9
NA
858.0
858.0
13/86
-
Source:  Ag. Prices,  USDA,  Dec.,  1983.

-------
                                           80
      Table VIII-10.  Prices and Market Value of Grain,  at Farm and
Non-Farm Levels,
Estimated Average
Lot Sizes, November, 1983 Prices

Item
Lot size
Wheat
Corn
Barley
Oats
Rye
Rice
Grain Soryhura
Price/bu.
11/83 a/
1,000 bu
§3.52
3.30
2.57
1.71
2.75
3.99
2.85
Value
Farm
10
(6-13)
$35,200
33,000
25,700
17,100
27,500
39,900
28,500
of Grain/Lot
Non-Farm
250
(50-500+)
$880,000
825,000
642,500
427,500
687,500
997,500
712,500
a/  Prices received by farmers, Nov. 1983,  except for rye,  which is
    preliminary estimate.

Source:  USDA:  Ag. Prices, Dec. 1983.

-------
                                        81
 These aggregate dollar values'and values for a typical lot are obviously
 very significant if they were to reflect the value of losses to operators
 at either level in the market as described earlier in this section.

 The fanner or other owner of grain with EDB at higher than permissible
 levels realistically would face a number of options short of losing the.
 entire value of the grain.  Such options would include delay of marketing,
 which would tend to reduce EDB residues due to the natural dissipation  of
 the chemical; turning or moving the grain; and aerating the grain.  All of
 these options have attendant costs of one type or another.  Also, the
 owner of the grain may need to sample amd test for residues which would
 generate additional cost.  These types of costs are addressed below in
 this section of the report.

 Turning the grain  is one of the methods that farmers and grain merchants
 may employ to help speed residue dissipation.  This turning cost is
 generally estimated to be between one-fourth and one-half of a cent per
 bushel  with seme estimates going as high as one cent per bushel.  The
 major obstacle to  overcome when turning grain is breakage.  It has been
 estimated that grain is reduced one grade due to breakage with each
 turning.   For this reason, turning costs for corn would be prohibitive.

 As  noted  above,  the owner of  grain may wish to have his product analyzed
 chemically for residues of EDB.   Agency staff has recently contacted a
 number of  vendors  and  found that the cost of analyzing grain or related
 food products for  EDB  is approximately $100 to $110 per sample at
commercial laboratories.   If  a large number of samples (i.e., hundreds)

-------
                                        82
 are  to  be  analyzed,  lower unit cost may be obtained.   There are additional
 costs to be  incurred in obtaining  and  transporting  the sample to the
 laboratory in  a manner in which allows for accurate laboratory measurement
 of actual  residues.   In instances  where statistically valid estimates must
 be taken of  very large lots of grain or a number of lots from a particular
 geographic area or source,  a survey design and  randan sample must be
 developed, all of which is an added expense.

 Agency  staff estimated that approximately 150 to 300  commercial residue
 chemistry  laboratories are available for residue analysis of foodstuffs in
 the  U.S. at  the present time.   The capacity of  these  laboratories for EDB
 testing  is not known.   If nothing  else,  there would be a relatively large
 physical transportation problem in getting all  of the samples that could
 be taken nationwide  to the relatively  few laboratories nationwide.  The
 extent  to which the  testing sites  might  be geographically dispersed has
 not  been evaluated.

The  amount of  grain  required to do chemical residue analysis of grain is
 likely to be relatively small  such as  possibly  a quart i.e.,  up to 3 Ibs.
of grain.  The market  value of corn is only about 6 or 7 cents a pound and
rice.  The most expensive grain, about 9 or 10  cents  per pound.

-------
                                         83
        b.  Indirect Impacts

Implementing EDB residue guidance at the raw grain  level  in  the  industry
could have indirect effects on other sectors in  the economy.   Products  for
which grain is a raw product could be affected as well  as substitutes for
such products at the intermediate and consumer levels.  Also,  the
livestock sector could be affected if there were significant  disruptions
in the supply of feedgrains.  In view of the relatively small maximum
percentages of the grain sector that would be likely  to be affected,  the
potential for such indirect economic effects is  quite limited.  To the
extent they do occur, they would be temporary.

-------
                                           84
C.  Economic Intact of Setting Maximum Permissible EDB,Levels  for Uncooked
    Grain Products

              1.  Usage, of EDB and Other Fumigants.

This section analyzes the potential economic effects of a.range of proposed
maximum permissible EDB,.residue  levels for grain-based products which require
further cooking.  Products included as processed/uncooked are  typified by
wheat flour, corn-based products, blended and  prepared flour mixes (e.g., cake
mixes, muffin mixes, etc.) and milled rice.  Other product  sectors,  such  as
wet corn milling  (producing corn starch and corn  syrup) and brewing, were not
included.  Only those industries in the GMA/states/USDA data bases which
produce processed/uncooked grain-based products with known  residues  are
considered in this analysis.

BOB finds its way into the human food chain in grain in a variety of ways.  As
discussed in the previous section, the first route of entry is the direct
application to grain stored in bulk.  Treated  grain goes  into  the market,
becomes mixed with untreated and other treated grain and "disappears" as
exports and feed  and food components of domestic  consumption.  The grain
containing EDB residues which "disappears" into the domestic diet becomes the
initial source of the dietary burden.  The second major route  of entry is
fumigation within the industries which utilize raw grain  in milling  and
processing.

-------
                                           85
 EDB can be used as a spot treatment of cereal processing and flour milling



 machinery to rid the internal portions of the machines (which are virtually



 inaccessable)  of insect infestations.   The major use of the so called "spot"



 materials containing EDB are wheat and corn-milling facilities.  EDB was also



 used to a very minor extent in a variety of other cereal grain processing



 industries (e.g.,  brewing).  Virtually every firm utilizing U.S. grain as a



 feed stock or a basic input item could potentially produce products with



 detectable EDB residues.







 EDB in  combination with methyl bromide was the most conmonly used fumigant



 mixture to spot-treat insect infestations in flour milling machinery.



 Approximately 300,600 pounds of EDB active ingredient were used annually by



 the flour milling  industry (Table VIII-11).  While actual data are not



 available for 1983,  the overall usage  of EDB for spot-treating flour milling



 equipment was  probably less than typical usage in most recent years.  The



major companies in the milling industry apparently discontinued usage of EDB



 in  their mills in  August  and September, 1983.  Since that time, the usage of



 EDB in  the milling industry appears to be very limited.  Moreover, the largest



distributor of spot-funigant materials (with about 60% of the market) is not a



 party to  the  on-going cancellation hearings and registrations for these



products  are cancelled.   Other suppliers are parties to the cancellation



hearings,  but  are  no longer producing  EDB-containing products for usage in



spot-treating milling machinery.

-------
                                           86
 Table VIII-11.  Estimated Annual U.S. Usage of EDB as a Spot Fumigant, 1978-1983



                                     "Annual U.S. Usage of EDB
 Source                                Date
Ferguson Fumigants, Inc.
Research Products Go.
Millers' National Federation
Millers' National Federation
Millers' National Federation
Environmental Protection Agency
U.S. Department of Agriculture
Ferguson Fumigants, Inc.
Douglas Chemical Co.
Centaur Associates
Centaur Associates
1978
1978
1978
1979
1980
1980
1981
1981
1982
1983
1983
290 ,000
600 ,000
323 ,087
323 ,235
285,759
465,000
162 ,000
230,518
164 ,000
300,000
300 ,591
Source:  Adapted from Centaur Associates, 1983.

-------
                                           87
 Centaur (1983) estimated that 79 percent of flour mills in the U.S. used EDB
 as a spot fimigant.  Centaur estimated that 231 to 243 flour mills used EDB
 spot fumigants an average of 10.0 to 10.69 times per year.  Respondents to
 Centaur case studies indicated average usage of EDB-containing fumigants
 approximated 10.5 gallons per mill fumigation.  Obviously, larger mills use
 more and smaller mills use less. The range in reported usage was 2 to 45
 gallons per treatment.

           2.  industry Profile

 The major  raw grain materials for the milling and processing industry include
 wheat,  corn, oats, barley and rice.   Of  these raw materials, wheat is by far
 predominant in terms of U.S. consumer importance, and is the most frequently
 treated with EDB.

 The flour  milling  industry is widely dispersed throughout  the U.S.   Wheat
 flour mills  tend  to be geographically dispersed due  to the bulky nature of the
 input.   Transportation costs encourage producers to  locate wheat flour mills
over a  wide  geographic range (Table  VIII-12).   The Corn Belt, Northern Plains
and Northeast states comprise over half  of total average daily milling
capacity.  Ownership in the  wheat flour milling industry is concentrated in
relatively  few firms.   The  leading 5 firms comprise  about  72  percent  of the
wheat flour milling  capacity (Table  VIII-13).

The production of  flour  and  other grain-mill  products from raw grains involves
an intricate process of many grindings and  sittings.   While the  details of the
entire process will  not  be given  here, in  general, about 72 percent of the

-------
Table VIII-12.   Daily Milling Capacity of vheat Flour Mills by Region

Region

Northeast
Southeast
Appalachian
Lake States
Corn Belt
Northern Plains
Delta States
Southern Plains
Mountain
Pacific
Puerto Rico
Total
Source: Milling

Nunber of
Mills

45
12
50
20
37
26
2
10
19
18
2
241
and Baking News.
in the U.S., 1983

Active Daily
Capacity
(cwt)
146,962
37,390
96 ,691
116,350
254,673
156,170
11 ,300
69,440
68,450
103 ,025
9,000
1,069,451

Average
Daily Capacity
Per Mill
(cwt)
3,266
3,116
1,934
5,818
6,883
6,007
5,650
6,944
3,603
5,724
4,500
4,438

Pet. of
Total
Capacity

13.7
3.5
9.0
10.9
23.8
14.6
1.1
6.5
6.4
9.6
0.8
100.0
1983 Milling Directory Buyers' Guide. Kansas
   City,  MO.   Sosland  Publishing Company, 1982, 120pp. As cited in Centaur
   Associates, October 13, 1983.

-------
                                      89

       Table VIII-13.  Top Ten Wheat-Durun-Rye Milling
Conpanies and Their Daily Capacities

Wheat
Company

Con Agra, Inc.
ATM Hilling Co.
Cargill, Inc.
The Pillsbury Co.
International Multifoods
General Mills, Inc.
Dixie-Portland Flour Mills
Nabisco Brands, Inc.
General Food Processors, Inc.
Bay State Milling Co.
Total
Total U.S. 1
Flour Capacities
(cwt)
186,300
135 ,700
130,500
114 ,900
60,000
55,000
51,000
43,000
31 ,300
30 ,000
837 ,000
,069,451
Source: Milling and Baking News. 1983 Milling
No. of
Wheat Flour
Mills

25
17
13
8
9
8
5
5
4
5
99
241
Directory Buyers1 Guide.
Kansas City, MO.   Sosland Publishing Company, 1982, 120 pp.   As
cited in Centaur  Associates, October 13, 1983.

-------
                                         90
Wheat  kernal  is extracted as flour  with the  remainder as raillfeed.   The



majority of millfeeds  are used  in a wide variety  of  animal  feeds.   The value



of  flour and  other mill  products approximated $5.7 billion  in 1983  (Table



VIII-14).







Flour  can be  treated  in  a number of different ways  to improve some  particular



property.  Such additives as maturing  agents,  bleaching agents and  self-rising



ingredients are blended  into the flour at the mill.   Other  additives such as



vitamins and  iron are  added  at  the  bakery.   Blended  and prepared flour



products were valued at  about $1.5  billion in 1983  (Table VIII-14}.  Two other



industry components of this  sector  are rice  milling  and wet corn milling.  The



value  of shipments from  rice milling and wet-corn milling industries were $1.5



and $4.6 billion, respectively  in 1983.







          3.  Application Practices and Rates







The primary registered uses  of  EDB  as  a spot fumigant are in cereal processing



equipment, flour elevators,  empty flour bins,  flour-milling equipment, grain-



mill equipment and grain-milling machinery.   EDB  is  particularly effective in



flour mills against most caramon stored product insect pests that infest



milling machinery and  equipment.  There are  many  ledges and obstructions in



milling machinery and  equipment that form "dead spots" where product can



accumulate.   These areas are particularly susceptible to infestation with



insects and proper "spot" treatment of these critical locations prevents



insect growth and reproduction.

-------
                                               91
         Table VIII-14.  Value of Shipments for Selected Industries and Years
 Industry                  1972     1977    1979      1980      1981      1982      1983  a/


 Flour and other
  grain mill
  products                2,380    3,683    4,218     4,835     5,103     5,494     5,685

 Rice milling               681    1,263    1,397     1,818     1,884     2,040     2,111

 Blended and prepared
  flour                    705    1,012    1,172     1,289     1,369     1,460     1,511

 Wet corn milling           832    2,015    2,442     3,239     3,378     4,479     4,634
  Total                  4,598   7,973   9,229   11,181   11,734   13,473    13,941

~aj  Estimated.                                                                    ~

Source:  U.S. Department of GCranerce:  1983 U.S. Industrial Outlook.  Washington,
         D.C.  1983.  pg. 37-2.

-------
                                           92
 In typical spot fumigation,  a -70% EDB and 30% methyl .bromide formulation



 is applied by mill employees or by ccnmercial applicators.  The actual



 application of the EDB product is to only certain parts or "spots" in the



 milliny machinery.  Fumigators move from the bottom of the mill to the top of



 the mill  fumigating individual pieces of equipment as they go.







 Generally when a mill is fumigated, it is shut down.  The milling equipment is



 allowed  to run dry to remove as much grain and flour from the milling machinery



 as possible.  Doors and windows are closed and locked except for the exit door



 for the  fumigators.  Typically, the fumigatoral/ enter the mill wearing



 appropriate full-face respirators applying the fumigant to the predetermined



 spots, working floor-by-floor.







 In most cases the applicator uses an application gun, which can be adjusted to



 allow predetermined amounts  of fumigant to be applied.  Usually 1 to 4 ounces



 of EDB-containing fumigant is used per "spot."  As each spot is fumigated, the



 applicator will close up the equipment so that the EDB will volatilize and



 remain within the equipment  for as long as possible.







 When the  application is complete, a process taking no more than a few hours,



 the mill  is locked and posted with warning signs.   The "hold" time within the



 mill is usually 24 hours,  the time needed to kill the insects.  Centaur (1983)



 in case studies indicated that a hold time of 41 hours is cannon.  In several



 cases, Centaur reported that the applicators complete the fumigation by



 Saturday  noon and the mill remains closed and locked until Monday when it is







T/  Fumigation is always done by a minimum of two-person teams.

-------
                                          93
 aired out.   Some time prior  to the resumption of the milling operations, the
 mill  is  aerated  as much  as possible by opening windows  and  doors  and  turning on
 fans  and milling equipment blowers.  Start-up of mill operations  occurs
 anywhere from zero to several  hours after  reopening. In Centaur  case studies,
 the average  time for  reopening to start-up was 3.89  hours,  but the  range in
 time  varied  from zero to 15  hours.

          4.   Projected  Quantities Exceeding  Maximum Permissible  Residue
               Levels

 The eventual  economic impacts  of  choosing  a maximum  permissible residue level
 for EEB  in flour  will depend on the actual profile of residues existing in
 flour  and grain mill  products. A variety  of  data on EDB residues have been
 received from GMA, several states,  and USEft.   It is  not known how
 representative these  data may  be  of actual residues  across  the nation but for
 now,  these data will  be  used in analyzing  the economic  impacts of the proposed
 permissible residue levels.  Quantities of stocks at mills, bakeries  and retail
 outlets with  violative residues could not  be  estimated  with available data.
 •Hie value of  stocks potentially violative  of  each given level were  derived
 using  the following methodology and assumptions.

          a.   All residue data from GMA/states/USDA  were assembled  into one
data base by  statisticians in  the ^ency's Office of Policy, Planning and
 Evaluation.   The percentage of samples of  uncooked grain-based products
exceeding each level  were derived  fron OPPE printouts.

-------
                                           94
               b.   The individual inventory values for  wheat, corn, rice and



 other  uncooked grain-based product stocks were not immediately available.   The



 percentages of wheat, corn, and other grain-based products which exceed each



 permissible residue level, were combined to form a single percentage.   This was



 accomplished by multiplying each percentage that exceeds the level by the  food



 factor coefficients utilized in the Tolerance Assessment System model.  The TAS



 model  is the model utilized in determining dietary intake and risks from these



 intakes elsewhere  in the document.







               c.   About 2/3 of the flour, milled products and blended and



 prepared mixes are baked within one month of milling.   The remaining 1/3 of



 these  products are packaged in consumer-sized units and may remain on retail



 shelves for an average of  4 months (but usually no longer than 8 months).



 Milled-rice products were  assumed to remain in inventory for 6 months.







               5.   Economic Impacts







               a.   Aggregate







Using  the asssumptions and data previously discussed,  the value of stocks



potentially violative were calculated.   Table VIII-15  provides the percentage



of uncooked products projected to exceed permissible residue levels being



considered.   For corn and  other  grain-based products,  setting permissible



residue  levels at  250  ppb  or lower could threaten a considerable portion of



existing stocks.   Only a small  portion  of wheat products (the major component)



would be affected  until more restrictive action levels of 50 and 25 ppb are



enforced.   Because wheat products are dominant, values of products exceeding

-------
                                                 95
        Table VIII-15.   Estimated Percentage of Samples and Value of Uncooked Products


Maximum
Permissible
Option Residue
Level
(PPB)
1 No Action
2 250
3 150
4 100
5 50
6 25
7 10
8 5
9 1
10 50/100
Above Maximum Permissible

Percent of Samples Above
Maximum Permissible
Residue Level
Wheat Corn Other
	 Percent — -
N/A N/A N/A
0.9 5.6 0
1.6 7.3 0
2.4 7.9 2.2
6.3 11.9 2.2
12.5 15.2 2.2
22.3 23.8 4.3
33.4 36.3 6.5
N/A N/A N/A
1.6 7.3 0
Residue Level

Weighted
Average a/

N/A
1.2
1.9
2.9
6.2
11.0
19.4
29.1
N/A
2.9


Value of Products b/
Exceeding
Maximum
Permissible
Residue Level
	 $ million 	
N/A
22.4
35.5
54.2
115.9
205.7
362.8
544.2
N/A
54.2
a/  Samples were weighted using the following food factors:   wheat = .73;  corn = .1;  and
    other = .17
b/  Based on inventory value of $1.87 billion.

-------
                                             96





 permissible residue levels are relatively nominal at levels 100, 150 and 250



 ppb.  At 50 ppb, a significant percentage of wheat products were found to



 exceed this residue level (6.3%).  The value of products exceeding 50 ppb



 approximates $120 million and would be expected to generate siynficant



 economic disruptions.  At the 5 ppb permissible residue level, severe



 economic disruptions are likely for the short term (e.g. 6 months) as the



 value of stocks exceeding this level is about $540 million.







           b.  Typical Firm







 Although widely dispersed (Table VIII-12), the wheat flour milling industry



 is dominated by a small number of large companies (Table VIII-13).  The top



 four companies have more than half the daily milling capacity of the



 industry.   For options 1 through 4 (Table VIII-15), producers of corn based



 products  face a much higher  probability of having products in violation



 than wheat  cr  other  product manufacturers.  The probabilities (measured in



 percent)  for  corn and wheat  products exceeding maximum permissible residue



 levels began  to converge at  the more restrictive levels (50, 25  and 5 ppb}.



 The  probability at these levels  of having products exceeding permissible



 levels is about equal  for corn and wheat  products.







 While large firms dominate, the  impact of products exceeding any



permissible residue level may  not  accrue  in  proportion  to production



capacity.   As  compared  to anall  firms,  large firms may be able to  mere



closely control  the source of  their grain input  and be  in a relatively



stronger position to produce below-permissible residue-level products.



Large firms would also be  able to  absorb  product recall better.    At this

-------
                                         97
point, the distribution of the impacts on large and small firms cannot be
determined due to the anonymity of the sample origins.  Under all but the
most restrictive residue levels, few firms would be expected to be
seriously affected.  Some additional spot sampling of incoming grain for
EDB residues by millers can be anticipated at the cost of about $100 per
sample.  Compared with other overall operational costs, costs at these
levels would be considered only incidental.

-------
                                            98

 D.  Ready-to-Eat Grain Products

           1.  Profile

 Grain for human consumption reaches the consumer in a variety of forms.
 Major classes of grain-based food products include cereal breakfast foods,
 bread and cake products,  and cookies and cracker-type products.  Table
 VIII-16  provides recent data on the value of  shipments for these groups of
 food products.  After accounting for inflation,  these figures still reflect
 slight real  gains in  shipments  in recent years.   Table VIII-17 provides a
 profile  of these same three industry segments.

           2.   Quantities  Exceeding Allowable  Levels

 Estimation of economic impacts  is dependent upon data submitted to the Agency
on EDB residues  found  in  ready-to-eat products.   Table VIII-18 provides the
percentages of products estimated to exceed the  alternative residue levels
being  considered.  For  those ready-to-eat  products for which data on residues
are available, setting  permissible residue levels as  high as 50 ppb would
 impact on  only a  small  percentage of  products in stock.   Proceeding to lower
and lower  permissible residue levels down  to  1 ppb would  place significantly
higher and higher portions of ready-to-eat product stocks in situations where
regulatory action might be taken.

-------
                                             99
          Table VTII-16.  Value of Shipments for Selected Final
                           Grain-Based Products
Products
                                              Value of  Shipments
 1980
 1981
1982
Cereal Breakfast Foods

Bread, Cakes & Related Products

Cookies and Crackers


  Total
 3,476

11,703

 4,036


19,215
-$ Millions	

 3,950       4,298

12 ,345      13,255

 4,450       4,700
20,745
22,253
Source:  U.S. Department of Conmerce,  1983.

-------
                                         100
  Table VI11-17.  Industry Data for Segments of ReadyTo-Eat Grain-Based
Products Industry


No.
NO.


of establishments
of establishments

Cereal
Breakfast
Foods
48
13

Breads, Cakes
and Related
Products
3,062
1,945

Cookies
and
Crackers
324
152
 with less than 20
 employees

4 largest firm market
 share                   89%

Employment in
 thousands              15.4
  33%
178.0
 59%
45.4
Source:  U.S.  Department of Commerce, 1983.

-------
                                         101
  Table VIII-18.  Percent and Value of Existing Stocks of Ready-To-Eat
Grain- Based
Products Exceeding
Specified Maximum Residue
Levels for EDB

Option
1
2
3
4
5
6
7
3
y
10
Maximum
Permissible
Level
(ppb)
No action
50
30
20
10
5
2
1
<1
20/10 £/
Stocks Exceeding
Percent
N/A
0.1
0.9
1.5
6.1
8.7
N/A
19.2
N/A
1.5
Level
Value
($ mil.)
N/A
1.0
8.8
14.7
59.8
85.3
N/A
188.2
N/A
14.7
a/  Assumes 6 months enforcement at 20 ppb followed by permanent
    enforcement  at  10 ppb.

-------
 The immediate initial effect of  imposing a maximum permissible  residue  level
 would depend on the volume and value of stock  in  inventory having EDB
 residues from prior use of EDB as either a grain  fumigant or a  spot-fumigant
 for milling equipment.  Given the coincident imposition of an emergency
 suspension of EDB use under FIFRA, the contamination of food products would
 begin to decline from levels existing prior to the emergency suspension.
 Hence, the volume and value of contaminated products having residues above a
 given level would be limited to inventories of products at the  time of  the
 emergency suspension and setting of maximum permissible residue levels.
 Ready-to-eat baked goods have relatively short shelf lives and  therefore
 inventories must reflect this situation.

 For bread and cake products,  this analysis assumes that inventories would be
 one week's sales.  For other ready-to-eat products, an inventory of four
 weeks1  sales is  assumed.  Using these factors and the sales fron Table
 VII1-16, the value of inventories of ready-to-eat products are  estimated to
 be  about $980 million in 1983 (assumes an inflation rate of 3.47% for
 grain-based food products in 1983 over 1982).  Using the percentages of
 products exceeding the proposed permissible residue levels, the estimated
 value  of inventories potentially exceeding those levels can be  calculated.
 The results are  shown in Table  VIII-18.  Clearly there could be major
 economic dislocations depending on the option chosen.  In this  analysis, the
 impact  would  range from less  than $1  million  to as much as nearly $190
million in  value of ready-to-eat products that could be seized  for exceeding
permissible levels for EDB residues.

-------
                                          103

          3.  Economic Impacts

          a.  Aggregate

The wide range of  impacts associated with  the  alternative  permissible  residue
levels need to be  considered  in  the overall context of  the ready-to-eat
grain-based product market.   At  the least  strict  or highest permissible
residue levels, the economic  impacts would be  relatively small,  no more  than
$10 million in an  annual market  of over $20 billion.  At the more strict or
lower levels, the  impacts would  be significant at nearly $190 million, but
the value of commodities affected would still  only account for  less than 1
percent of annual  sales of the affected product classes.   The aggregate
impacts would likely  result  in short-term  economic dislocations but would not
cause long-term industry disruptions.

          b.  Typical Firm

Referring back to  Table VTII-17  which  provides information on sane aspects of
the structure of the  segments of the ready-to-eat grain-based industry,  it
can be concluded that cereal  breakfast foods  is dominated  by a small number
of large firms.  On the other hand, the bread  and cake  industry is much more
atomistic with a much larger  number of snail  establishments.  The cookies and
crackers segment fall in between the other two segments.

For several options considered,  an  individual  firm would face a low
probability of having products in violation.   For those firms that do have
such products, the impacts would generally be limited to the amount of
product in inventory. Both  large and  small firms would be potentially

-------
                                        104
affected as there  is nothing  in  these possible  actions  that  is  known to favor



one group over the other.   Individual firms which  happen  to  be  small,



however, may be affected.   The ability of  such  firms  to absorb  a recall of



its products must be questioned.  At this  point,  it cannot be determined how



many small firms might be  so  adversely  impacted as to cause  business failure.



It can be concluded that under most optional permissible  residue levels being



 considered, that  few firms would be expected to  be  impacted directly and in



a highly adverse fashion.








Individual firms oould also be affected  if they felt  it necessary to initiate



tests for EDB residues in  the raw materials such  as flour which they would



purchase in the future.  Most major firms  already use quality control



programs with respect to materials used  to produce their  outputs.  The



addition of bests  for EDB  residues would cost an  estimated $100 per sample.



For firms purchasing inputs in large volumes, the incremental costs would be



relatively minor.  The cost of residue  analysis would be  much more



significant for firms that purchased materials  in smaller lots  but feel



compelled to test  for such residues.







E.  Value of All Grain and Related Products Exceeding Maximum Permissible



    Levels







A summary of the value of  all grain and  related products  that  are estimated



to exceed maximum permissible residue levels  is presented in Table VIII-19.



The values range from less than  $50 million to  near $1.6  billion with the



most restrictive option.

-------
                                        105
Table VI11-19.  Value of Grain Products Exceeding EDB Maximum Permissible Levels,


Maximum Permissible
Residue Level
No.
1
2
3
4
5
6
7
8
9'
10
11
ppb
None
50/250/5 ,000
30/150/3,000
20 A 00/2 ,000
10/50A ,000
5/25/500
2/10/200
1/5/100
<1A/100
2 ,OOOA ,000
30A 50/900
by Type of Product

Raw
Grain

$ 0
26
26
26
73
193
N/A
858
858
13/86
97
, February

Prepared/
Uncooked

$ N/A
22
36
54
116
206
363
544
N/A
54
36
1 , 1984

Ready- to- Eat

§ N/A
1
9
15
60
85
N/A
188
N/A
15
9

Total

N/A
49
71
95
249
484
N/A
1,590
N/A
82 A 55
142

-------
                                          1C6
F.  Summary of Agricultural and Food Industry impacts

FIFRA, Section 25, requires that the Administrator of EPA consider the
effects of a regulation upon production and prices of agricultural
commodities, retail food prices and otherwise on the agricultural economy.
In view of the projected aggregate effects from implementing the permissible
residue levels being evaluated in the options, there would probably be
notable temporary impacts with respect to the affected agricultural
commodities if certain of these levels were adopted as enforceable limits.
Most growers, grain dealers, milling firms, bakers, food processing
companies, and grocery related firms would be affected to seme degree.
Individuals and firms at each point in the chain from grain production to
consumer products may have to absorb losses totalling fron less than $50
million to as much as $1.6 billion depending on the residue levels chosen.
There would be the possibility of isolated and temporary dissruptions in
supplies of raw materials and final consumer commodities.  The temporary
nature of the disruption, in some instances a matter of a few days, would
support the conclusion that overall supply and prices of agricultural
commodities would not be severely affected.  In general, the dollar-value
impacts decline as one moves from the grain storage, through the grain
processing sector and finally into the retail sector.  In no sector would
there be an expectation of any significant permanent structural or behavioral
changes.

-------
                                            107
 G.   Summary of Small Business Impacts







 The Regulatory Flexibility Act requires that agencies issuing regulations



 take special note of the impacts of proposed regulations on small entities.



 This analysis has focused upon the  owners of grain  which could be



 contaminated by EDB residues, grain milling companies, and processors of



 consumer  readyto-eat products.  The principal groups affected would be



 farmers,  grain storage companies, millers, and bakeries who might have raw



 materials resulting in EDB residues above permissible levels as finished



 products.  In view of the projected overall aggregate impacts of implementing



 the  permissible residue levels, we  conclude that, at  certain levels, there is



 a likelihood of an impact on  the significant number of small businesses and



 farms, many of  which would be small entities.  The  number and profile of



 those small entities could not be developed in detail with the data



 available but is suggested in the foregoing analysis.







There is  nothing in the proposals which would suggest that impacts would be



more likely to  occur  on small businesses than on larger ones.  However, it



 should be stated that in instances  where residues would exceed permissible



residue levels  and  products would be taken from the market,  there would be



significant economic  impacts  on those firms. Small firms would be less



 financially capable of  absorbing  losses resulting from seizures of their



products.   Countering arguments can be made that all  adverse impacts from



this proposal should  be temporary and henoe firms that can survive the



initial impacts  would likely  remain viable businesses.

-------
                                      1C8
 H.   Economic Impacts of Emergency Suspension

      1.   Introduction

 The Agency has evaluated the economic effects of an emergency suspension
 of  grain  and flour milling  uses of EDB.   Economic impacts were separately
 evaluated separately for a  six-month period allocated for a suspension
 proceeding and for two years which may be required to conduct the
 cancellation hearings.  The consideration of economic impacts stemming
 from a suspension  is limited to a two-year period because the maximum
 projected length of  a cancellation proceeding would be about two years.  A
 suspension order remains in effect only during a cancellation proceeding.
 Thus, only the impact which would arise during this period would be at
 issue in  a suspension.  Any impacts which would be caused by a suspension,
 but which would be felt after this period, are also considered.

 Economic  impacts of  suspending  grain storage and milling/cereal processing
 usage of  EDB during  the six-month period  were estimated by assuming all
 registered  alternatives are available.  The analysis generally provides
 qualitative estimates of impacts since data are not available to support
 precise quantitative  estimates.

The Agency's analysis indicates that a six-month and two-year suspension
of grain and milling  uses of EDB would not significantly affect U.S.
production  or  prices  of any commodities or services in relevant sectors.
Economic  impacts of the suspension  would  be minor in most cases, even at
 the  local/regional level.

-------
                                       I U>
     2.  Grain Fumigants








          a.  Usage Profile








Details of the profile of the grain production industry have been



presented in Section VIII-B of this report and will not be repeated here.








EDB is used as a liquid grain fumigant  in mixtures whose major



constitutients are carbon tetrachloride/carbon disulfide (CT/CD) and



carbon tetrachloride/ethylene dichloride (CT/EDC).  EDB is also used  in



combination with chloroform and CD, but the overall volume of this product



is small in comparison to the CT-based products.  An estimated 425,000



gallons of formulated products containing 170,400 pounds of EDB active



ingredient were used for insect control in grain storage facilities in



1983.  At typical application rates, approximately 202 million bushels of



grain stored in 1983 were treated with  fumigants containing EDB (See Table



VTII-2).  During the next 6 months, approximately 25%, or 51 million



bushels of grain would normally be treated with liquid grain fumigants



containing EDB.  This is the approximate quantity that will need treatment



with other materials during the next six months if EDB is suspended.








The most likely alternatives to EDB-containing liquid grain fumigant



mixtures are the other mixtures of carbon tetrachloride/carbon disulfide



and carbon tetrachloride/ethylene dichloride at on-farm storage facilties.



At off-farm locations, either the CT/CD or the phosphine producing



products would be the major replacements.  Both carbon tetrachloride-



containing products and the phosphine-producing products are currently  in



widespread use and available in sufficient supplies to meet grain  storage



treatment needs during either the six-month or two-year periods.

-------
                                        110
 Treatment costs with alternatives vary depending on many factors.



 However, the available alternatives can be used at roughly the same (Table



 VIII-20 and Table VII1-21)  or less costs than the EDB-containing materials



 without any significant loss in insect control.  Moreover, the usage of



 EDB-containing liquid grain fumigants during the first six months of the



 year  comprises only about 25% of annual usage.  Thus,  the demand for EOB



 during the six-month anergency suspension period can be met by



 alternatives.








           b.   Impact Analysis








 If EEB-containing  liquid grain fumigants are not available during a



 six-month suspension period,  treatment costs could decline slightly where



 users  shift  from EDB to CT/CD containing products.  Aggregate costs



 (assuming all  former BOB users shift  to CI/CD products)  for grain



 fumigation could decline by $88,000 to $388,000.  As for users who choose



 to utilize the CT/EDC and phosphine products instead of EDB, the cost



 increases experienced by CT/EDC users could  be approximately offset by



 cost decreases by  phosphine users (Table VIII-22).







 During a  two-year  period, treatment costs would decline by approximately



 $348,000  to $1.5 million if former users of  ED&-containing liquid grain



 fumigants  shift  to the variety of available  CT/CD products.  Increased



 costs that accrue  to  users  who  choose  to use the CT/EDC  product will  be



offset by  those  users who can utilize  the less costly  phosphine producing



products.

-------
                                             Ill
           Table VIII-20.  Prices for Selected Grain Fumigants a/, 1983-84
 Product
%EDB
  Unit
Price Per Unit
 CT/CD/ECB/Sp2
 Chlorofarm/EDB
 CT/CD/EDB/SO^
 CT/EDB/EDC/S02
 CT/EDC/EDB
 CT/CD (80-20)
 CT/EDC (30-70)
CT/CD (85-15)
CT/CD/Sp2/Pentane
 1.2
 5.0
 5.0
 5.0
 7.2
-0-
-0-
-0-
 1 gal
53 gal
 1 gal
 5 gal

53 gal
 5 gal
 1 gal

53 gal
 5 gal

54 gal
54 gal
 5 gal

53 gal
 5 gal

53 gal
 5 gal
      ($)

 7.95-9.75/gal
 6.45/gal
 6.45/gal
 7.52/gal

 5.20-6.30/gal
 6.16-6.95/gal
 6.65/gal

 5.85/gal
 6.50/gal

 5.30/gal
 5.15/gal
 7.65/gal

 5.30/gal
 5.95/gal

 5.30/gal
 5.95/gal
Phosphine -0-
Flask
40/Flask
a/ EDB = ethyl ene dibrcraide
    CT  = carbon tetrachloride
    CD  = carbon disulfide
    EDC = ethylene dichloride
    S02 = sulfur dioxide
    Phosphine = aluminum phosphide

Sources:  J.M. Schultz Seed Company (November  14, 1983), Dieterich, IL.
          Pueblo Chemical & Supply Co.  (December 5, 1983), Hannibal, MO.
          ADI Distributors Inc.  (1983).   Carmel, IN.
          industrial Fumigant Co.  (January 31, 1983).  Olathe, KS.

-------
                                              112
 Table VI11-21.  Comparative Treatment Costs for  EDB  and Alternative Liquid Grain
                                Fumigants,  1983-84


Product

CT/CD/EDB/S02
CT/CD/EDB/S02
CT/EDB/EDC/S02
CT/EDC/EDB
CT/CD/ (80-20)
CT/EDC
CT/CD (85-15)
CT/CD/S02/pentane
Phosphine


Unit

5 gal.
1 gal.
5 gal.
5 gal.
54 gal.
5 gal.
5 gal.
5 gal.
1,000 grams

Price Per
Gallon
(S)
6.45
7.95-9.75
6.16-6.95
6.50
5.30
7.65
5.95
5.95
$40/1,000 grams

Application
Rate a/
(gal ./I, 000 bu.)
2
2
2
2
2
3
2
2
150 grams
Material
Cost Per
1,000 Bu.
(S)
12.90
15.90-19
12.32-13
13.00
10.60
23.85
11.90
11.90
6.00





.50
.90






Note:  Application costs were based on treatment  rates  for wheat  stored  in metal
       bins.  Treatment costs for shelled corn and crops stored  in  wooden bins  are
       higher for all liquid fumigants.

a/  Harein and Sumner, 1982.

-------
           Table VIII-22.  Estimated Treatment Costs Using EDB Alternatives for 6-Month
and 2-Year Periods

Product

EDB-oontaining
CT/CD-containing
CT/EDC-conta in ing
Phosphine

Range in
Treatment Cost
per 1000 bu.
(?)
12.31-19.50
10.60-11.90
23.85
8.00 £/

6-Month
Treatment
Costs a/
(? 000)
628-995
541-607
1,216
408

2-Year
Treatment
Costs b/
$ (000)
4,977-7,878
4,282-4,808
.9,635
3,232


Change in Treatment Costs
with Alternatives
6 months
($ 000)
—
-88 to -388
+221 to +588
-220 to -587
2 years
($ 000)
—
-348 to -1,535
+1,757 to +4,658
-1,745 to -4,646
a/  Assumes treatment of 51 million bushels of grain during the next 6 months.
b/  Assumes treatment of 202,000 bushels annually.
c/  Includes additional labor and materials charge of $2/1,000 bu.

-------
                                        114
 During either the six-month or the two-year time frames, users should not



 experience any significant changes in the ability to control insects with



 the CT/CD alternatives.







 Overall for  either time  frame, the comparative costs of using other



 registered and readily available materials, suggests that users should not



 experience any sSrious economic impacts if EDB products are not



 available.







     3.   Spot Treatment







 EDB  in  combination with methyl bromide has been the  most commonly used



 funigant mixture  to spot-treat for  insect infestations  in flour milling



machinery. There has  been a significant decline in  the use of the EDB/



methyl bromide mixture as  concern  has been raised with  respect to the



 safety of  these products.







           a.  Usage/Profile







Details of the usage of EDB and  the profile of the grain milling industry



have been  presented in Section VIII.C.2  of this report.   Major findings



are that up to 79% of  all  mills  had been using EDB containing products

-------
                                        115
 prior  to the summer  of 1983.   Spot treatment o£ milling equipment has been
 done on  a scheduled  basis  with treatments being applied about 10 to 11
 times  per year  on average. Flour mills requiring treatment are dispersed
 throughout the  U.S.  although  there is  increased concentration in the Corn
 Belt,  Northern  Plains, and Northeast States.

           b.  Alternatives

 Alternative means of insect control  in grain milling  equipment include the
 use of other registered pesticides and increased  diligence  in cleaning  and
 sanitation of the mill machinery. The most  likely chemical alternatives
 to EEC-containing  products would  be  methyl bromide and  aluminum phosphide.
 Other registered  products  containing ethylene dichloride and carbon
 tetrachloride could  also be used, but  these  products  are less preferred as
 the  larger  volume  of liquid applied  can leave residues  that clog the
machinery when  the mills are  restarted following  fumigation.

 In addition to  alternative chemical  control  measures, millers may need  to
 increase efforts  in  cleaning machinery.   All grain product  residues
 should be removed when machinery  is  cleaned  in  order  to remove food
sources and harborage for  insects.   The use  of  improved sanitation
practices and the available alternative chemicals will  probably increase
 the labor requirements for  insect control  in mills.

-------
                                        1  16
          c.  Impact Analysis








Prior to  the Agency's  cancellation of EDB as  a  spot-furnigant  to treat



cereal processing equipment, annual  usage of  EDB was approximately 300,600



pounds of active  ingredient.  Major  millers stopped using  EDB containing



fumigants in August/September of 1983, probably intending  to  await the



final outcome of  the EDB cancellation hearing.  There may  currently be



usage of  EDB by smaller mills, but this usage would have to cone frcm



existing  stocks as none of  the registered producers are currently



producing EDB containing products for spot-treating milling machinery.  It



is estimated that usage has now declined to about 10 percent  of its former



level.  This situation is not likely to change  over the next  six months



since the milling industry  no doubt  is concerned with the  public's concern



with respect to the wholesomeness of their  products due to EDB residues.



Likewise, usage over the next two years would not be expected to be



significant.








The milling industry has the option  of adopting the use of other



registered fumigants and strategies  for insect  control and apparently has



done so since the summer of 1983.  Overall, the adoption of  these



alternative control methods could increase  insect control  costs by $3.5 -



$4.0 million over the  next  six months and by  $15.0 - $16.0 million over



the next  two years.  There  remains seme uncertainty as to  whether the mill



will be able to achieve insect control with the alternative chemical that



is comparable to control with EDB containing  products.

-------
                                        117
 I.   Economics References

 1.  ADI Distributors Inc. 1983.   ADI  1983  Price Book. Carmel, IN.

 2.  Centaur Associates, Inc. 1983.  Final  Report. Analysis of Four
     Industry Segments Vfaich Wbuld Be  Affected By Proposed Standard
     Contract J-9-F-2-0067.  Prepared  for the Occupational Safety and
     Health Administration. October  13,  1983.

 3.  Conrath, W. P. Personal Communication  with Roger C. Holtorf. 23
     January 1984.

 5.  Data Resources, Inc. 1984.  Agricultural Review.  Lexington, MA.

 6.  Development Planning and Research Associates. 1979.  Preliminary
     Benefit Analysis.  Cancellation of  Ethylene Dibromide  in Fumigants
     for Stored Grain. Manhattan, KS.

 7.  Development Planning and Research Associates. 1979. Preliminary
     Benefit Analysis of Cancellation  of Ethylene Dibrcmide  in Fumigants
     for Flour Milling Equipment. Manhattan,  KS.

 8.  E.P.A. 1983.  Ethylene Dibromide  (EDB) Position Document 4.  Office
     of Pesticide Programs.  Washington, D.C.

 9.  Harein, P.K. and W. Sumner,  1982.  A Report on the Use  of CC14 on
     Farm-Stored Drain.  Draft.  Univ. of Minnesota, St. Paul, MN.

10.  Holtorf, R.C. and G. F. Ludvik.  1981.  Grain Fumigants: An Overview
     of Their Significance to U.S. Agriculture and Commerce  and Their
     Pesticide Regulatory Implications.   Office of Pesticide Programs.
     U.S. E.P.A.  Washington, D.C.

11.  Industrial Fumigant Co., 1983.  Products and Services Price List.
     Olathe, KS.

12.  Inglett, G.E. 1974. Wieat: Production  and Utilization.  AVI
     Publishing Co., Inc. Westport,  CN.

13.  	, 1970 Corn:  Culture,  Processing, Products.  AVI
     Publishing Co., Inc. Westport,  CN.

14.  Pueblo Chemical & Supply Co. 1983.  Price List. Hannibal, MO.

15.  Roan, C.C.  Senior Consultant,  Hopes Consulting  Inc.  letter  to
     William D. Ruckelshaus. 17 January 1984.

16.  J.M. Schultz Seed Company. 1983.  Wholesale Price List.  Dieterich,
     IL.

17.  Temple, Barker & Sloane, Inc. 1984. Economics of  Immediate  EDB
     Removal. Lexington, MA.

-------
                                         118
18.  U.S. Department of Commerce.  Bureau of the Census.  1981.  1978
     Census of Agriculture.  Summary of State Data.  Vol. 1, Part 51.
     Washington, D.C.

19.  U.S. Dept. of Commerce. Bureau of Industrial Economics. 1983 U.S.
     Industrial Outlook. Washington, D.C.

20.  USDA EPS. 1983. Feed-Outlook and Situation Report. Washington,
     D.C.

21.  USDA SEA.  1979.  Insect Control in Farm-Stored Grain, Fanners
     Bulletin No. 2269.  Washington, D.C.

22.  USDA SRS Crop Reporting Board.  1984 Grain Stocks.  January 23.
     Washington/ D.C.

23.  GSDA SRS. Crop Reporting Board., 1983.  Agricultural Prices.
     Washington, D.C. December.

-------