Pesticides: Reducing Exposure to Residues of Canceled Pesticides (Chapter
Report, 12/28/94, GAO/RCED-95-23).

Because the residues of most pesticides do not linger in the
environment, officials at the Environmental Protection Agency (EPA)
believe that most marketed foods do not contain unsafe levels of
residues from canceled pesticides.  However, the residues of a few
long-canceled chlorinated pesticides continue to appear, especially in
fish.  An EPA study shows that for five canceled chlorinated pesticides,
consumers of some fish may be exposed, over a lifetime, to health risks
that exceed the agency's standard of negligible risk--under which the
risk of an additional case of cancer does not exceed one in one million.
EPA proposed lower action levels in 1991 for residues of the five
canceled pesticides, but the Food and Drug Administration believed that
EPA had not given enough weight to the residues' unavoidability.
Although both agencies believe that the existing action levels should be
lowered, neither has taken further steps to do so.  EPA does not revoke
a pesticide's tolerances at the same time it cancels the pesticide's
registrations for food use. On average, the agency has taken more than
six years to revoke the tolerances for canceled pesticides.  The
establishment of procedures linking revocation to cancellation would
provide for more efficient revocation actions and would reduce the
possibility that consumers might eat imported foods containing residues
of pesticides that EPA no longer considers acceptable for use on food
crops.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  RCED-95-23
     TITLE:  Pesticides: Reducing Exposure to Residues of Canceled 
             Pesticides
      DATE:  12/28/94
   SUBJECT:  Pesticides
             Pesticide regulation
             Toxic substances
             Health hazards
             Safety standards
             Safety regulation
             Interagency relations
             Food inspection
             Fishes
             Consumer protection

             
**************************************************************************
* This file contains an ASCII representation of the text of a GAO        *
* report.  Delineations within the text indicating chapter titles,       *
* headings, and bullets are preserved.  Major divisions and subdivisions *
* of the text, such as Chapters, Sections, and Appendixes, are           *
* identified by double and single lines.  The numbers on the right end   *
* of these lines indicate the position of each of the subsections in the *
* document outline.  These numbers do NOT correspond with the page       *
* numbers of the printed product.                                        *
*                                                                        *
* No attempt has been made to display graphic images, although figure    *
* captions are reproduced. Tables are included, but may not resemble     *
* those in the printed version.                                          *
*                                                                        *
* A printed copy of this report may be obtained from the GAO Document    *
* Distribution Facility by calling (202) 512-6000, by faxing your        *
* request to (301) 258-4066, or by writing to P.O. Box 6015,             *
* Gaithersburg, MD 20884-6015. We are unable to accept electronic orders *
* for printed documents at this time.                                    *
**************************************************************************


Cover
================================================================ COVER


Report to the Chairman, Environment, Energy, and Natural Resources
Subcommittee, Committee on Government Operations, House of
Representatives

December 1994

PESTICIDES - REDUCING EXPOSURE TO
RESIDUES OF CANCELED PESTICIDES

GAO/RCED-95-23

Residues of Canceled Pesticides


Abbreviations
=============================================================== ABBREV

  DDT -
  EPA - Environmental Protection Agency
  FDA - Food and Drug Administration
  FFDCA - Federal Food, Drug, and Cosmetic Act
  FIFRA - Federal Insecticide, Fungicide, and Rodenticide Act
  ppm - parts per million
  USDA - U.S.  Department of Agriculture

Letter
=============================================================== LETTER



B-258289

December 28, 1994

The Honorable Mike Synar
Chairman, Environment, Energy, and
 Natural Resources Subcommittee
Committee on Government Operations
House of Representatives

Dear Mr.  Chairman: 

This report responds to your request that we (1) determine whether
marketed foods contain unsafe levels of residues from canceled
pesticides and (2) evaluate the Environmental Protection Agency's
(EPA) procedures for revoking tolerances for canceled food-use
pesticides.  The report contains recommendations aimed at ensuring
that EPA and the Food and Drug Administration (FDA) develop
appropriate standards for regulating the residues of canceled
pesticides in fish and that EPA revokes the tolerances for canceled
pesticides in a timely manner. 

We are sending copies of this report to the Administrator, EPA, and
the Commissioner, FDA.  We will also make copies available to others
upon request. 

Please contact me on (202) 512-6111 if you or your staff have any
questions about this report.  Major contributors to this report are
listed in appendix V. 

Sincerely yours,

Peter F.  Guerrero
Director, Environmental
 Protection Issues


EXECUTIVE SUMMARY
============================================================ Chapter 0


   PURPOSE
---------------------------------------------------------- Chapter 0:1

Since its creation in 1970, the Environmental Protection Agency (EPA)
has canceled the registrations (licenses) for hundreds of pesticides,
often because it has found that the pesticides pose unreasonable
health risks.  Many of these pesticides were once used in food
production.  A few of them, such as DDT, whose registrations were
canceled about 20 years ago, persist in the environment and continue
to appear in certain foods.  In addition, hundreds of tolerances--or
allowable limits on residues of pesticides in food--remain in effect
for other pesticides whose registrations have been canceled. 

Concerned that canceled pesticides may pose a health risk for U.S. 
consumers, the Chairman, Environment, Energy, and Natural Resources
Subcommittee, House Committee on Government Operations, asked GAO to
(1) determine whether marketed foods contain unsafe levels of
residues from canceled pesticides and (2) evaluate EPA's procedures
for revoking tolerances for canceled food-use pesticides. 


   BACKGROUND
---------------------------------------------------------- Chapter 0:2

Federal responsibility for protecting U.S.  consumers from exposure
to unsafe pesticides is shared by EPA, the Food and Drug
Administration (FDA), and the U.S.  Department of Agriculture (USDA). 
EPA registers pesticides for use on specific crops and, because
pesticides may remain on crops, sets tolerances to limit human
consumption of their residues.  EPA sets tolerances at levels that it
believes will pose no unreasonable risks to consumers.  A single
pesticide may be registered for multiple food and nonfood uses; each
registered use on a food crop must have a tolerance or an exemption
from a tolerance.  FDA and USDA use tolerances to monitor residues of
pesticides in foods and animal feed sold in interstate commerce. 
States may monitor residues in foods that are not sold in interstate
commerce. 

EPA may cancel a pesticide's registrations for some or all uses if it
finds that the pesticide does not meet current standards.  After a
pesticide's registration for a specific use has been canceled, the
pesticide may no longer be sold or distributed for that use in the
United States.  However, as long as the pesticide's associated
tolerances stand, foods containing allowable amounts of the
pesticide's residues may be sold in the United States.  Tolerances
are not automatically revoked when registrations are canceled. 
Instead, EPA takes a separate action to revoke a canceled pesticide's
tolerances. 

In the course of revoking a pesticide's tolerances, EPA assesses the
pesticide's persistence in the environment.  If EPA determines that
the pesticide persists and is unavoidable in certain foods, EPA
recommends action levels to FDA or USDA.  Action levels are
enforcement standards that are used (1) in place of tolerances that
have been revoked or (2) in regulating residues of pesticides that
have never been registered for use on certain foods, such as fish,
but have inadvertently contaminated these foods. 

During the late 1960s and early 1970s, FDA established action levels
for the residues of certain chlorinated pesticides in fish.  These
pesticides, for which tolerances had been set for other foods, were
appearing in fish because they were being transported in agricultural
runoff to the nation's streams and lakes and fish were accumulating
the residues in their tissues.  Although EPA canceled the
registrations for nearly all uses of these pesticides during the
1970s and revoked most of their tolerances in 1986 (see app.  I), the
pesticides have persisted in the environment--that is, they have
remained in the soil and water--and action levels are still required
to regulate their residues in foods, especially in fish. 

In 1982, EPA adopted a policy in which it stated that, in
recommending action levels, it (1) would assess both the health risks
and the extent to which residues were unavoidable in foods and (2)
would lower action levels periodically as residues declined in the
environment.  In this policy, EPA also stated that a pesticide's
tolerances should logically be revoked when the pesticide's
registrations for food uses have been canceled.  FDA and USDA
concurred with EPA's policy. 


   RESULTS IN BRIEF
---------------------------------------------------------- Chapter 0:3

Because the residues of most pesticides do not persist in the
environment, EPA officials believe that most marketed foods do not
contain unsafe levels of residues from canceled pesticides.  However,
the residues of a few long-canceled chlorinated pesticides have
persisted and continue to appear, especially in fish.  Both EPA and
FDA believe that the levels of these residues in fish have generally
declined since the action levels for regulating them were established
about two decades ago.  Nevertheless, a study conducted by EPA shows
that, under the action levels for five canceled chlorinated
pesticides, consumers of some fish may be exposed, over a lifetime,
to health risks that exceed the agency's standard of negligible risk
(under which the risk of an additional case of cancer does not exceed
1 in 1 million).  On the basis of this study, EPA proposed lower
action levels to FDA in 1991 for residues of the five canceled
pesticides in fish.  These proposed action levels reflected EPA's
weighing of both health risks and the extent to which the residues
were unavoidable in fish.  FDA agreed that the action levels for the
five pesticides should be lowered to reflect declines in residues but
believed that EPA had not given sufficient weight to the residues'
unavoidability.  Although both agencies believe that the existing
action levels should be lowered, neither has taken further steps to
reach agreement on the appropriate reduction. 

EPA does not revoke a pesticide's tolerances at the same time as it
cancels the pesticide's registrations for food uses.  On average, the
agency has taken over 6 years to revoke the tolerances for canceled
pesticides.  EPA officials acknowledge that the agency's current
process for revoking tolerances takes too long and makes inefficient
use of scarce resources.  The establishment of procedures linking
revocation to cancellation would provide for more efficient
revocation actions.  It would also reduce the potential for consumers
to be exposed through imported foods to residues of pesticides that
EPA no longer considers acceptable for use on food crops. 


   PRINCIPAL FINDINGS
---------------------------------------------------------- Chapter 0:4


      UNSAFE RESIDUES MAY BE
      PRESENT IN SOME FISH
-------------------------------------------------------- Chapter 0:4.1

According to EPA officials, most marketed foods do not contain unsafe
levels of residues from canceled pesticides because most pesticides
break down fairly quickly in the environment.  Furthermore, EPA
officials do not believe that residues of the few canceled
chlorinated pesticides that have persisted in the environment are
present at significant levels in most foods.  In 1985, when EPA
published its intention to revoke the tolerances for three of these
pesticides--DDT, chlordane, and dieldrin--in foods other than fish,
EPA did not receive comments expressing concerns about health risks. 
Therefore, in 1986, EPA revoked the tolerances and recommended action
levels for residues of these three pesticides in foods other than
fish. 

When EPA proposed action levels for residues of DDT, chlordane, and
dieldrin in foods other than fish, it also proposed to retain the
action levels that FDA had been using since the late 1960s and early
1970s for residues of these pesticides in fish.  In response to these
proposed action levels, EPA received some comments expressing
concerns about health risks.  EPA responded to these comments by
conducting a multiyear study of health risks and residue levels in
fish for these three pesticides, plus two others--heptachlor and
mirex.  This study showed that, under the existing action levels,
consumers of some fish could be exposed, over a lifetime, to health
risks exceeding the agency's standard of negligible risk.  EPA
believed that health risks could be higher for consumers who eat
either large quantities of fish or more highly contaminated fish.  On
the basis of this study, EPA sent a draft letter to FDA proposing
lower action levels.  Unlike the earlier action levels, these took
health risks into account as well as actual residue levels, which EPA
believed had generally declined since the earlier action levels were
established. 

Although FDA agreed that the action levels for fish should be lowered
to reflect the decline in actual residue levels, it maintained that
EPA had not adequately assessed the extent to which residues of the
canceled pesticides were unavoidable in fish.  According to FDA, EPA
had not sufficiently demonstrated the need to lower action levels as
much as EPA had proposed.  FDA asked EPA for more data to explain to
consumers and commercial fisheries the basis for stricter standards
and their potential economic impact.  However, EPA considered its
data adequate and conducted no further studies.  EPA has never
formally recommended the lower action levels to FDA. 

In addition to providing FDA with enforcement standards for
regulating fish in interstate commerce, federal action levels provide
states with standards for regulating fish caught locally by
recreational and subsistence fishermen.  According to an EPA study,
most states use the federal action levels as their basis for
monitoring residues and for advising consumers of any health risks. 
As long as actual residue levels do not exceed the federal action
levels, these states are unlikely to advise consumers of the
potential health risks they may incur in eating the fish. 


      EPA HAS BEEN SLOW TO REVOKE
      TOLERANCES FOR CANCELED
      PESTICIDES
-------------------------------------------------------- Chapter 0:4.2

EPA has not instituted standard procedures for revoking a pesticide's
tolerances after it has canceled the pesticide's registrations for
food uses, and when it has revoked these tolerances, it has taken, on
average, over 6 years to do so.  Typically, the agency has allowed
about 2 years for remaining stocks of the pesticide to be used and
for products legally treated with the pesticide to move through
commerce.  The balance of the delay has occurred because EPA has
assigned low priority to revocation and has not established
guidelines linking revocation to cancellation.  Establishing such
guidelines would be consistent with EPA's 1982 policy on revoking the
tolerances for canceled pesticides, would streamline EPA's process
for revoking tolerances, and would eliminate the inconsistency that
now allows residues of the same pesticides that EPA has deemed
unacceptable for use on crops to appear legally in food. 

EPA has recently made progress in revoking tolerances and, as of July
1994, had done so for 50 canceled food-use pesticides.  Nevertheless,
a potentially large but unknown number of pesticides still have
tolerances awaiting revocation.  (EPA's data bases do not contain the
information needed to determine this number.) As additional
cancellations occur, this backlog can be expected to grow.  An EPA
official estimated that 100 or more canceled food-use pesticides may
still have tolerances awaiting revocation.  GAO identified 10
high-priority pesticides whose registrations for food uses were
canceled from 3 to 13 years ago but whose 185 associated tolerances
had not yet been targeted for revocation. 


   RECOMMENDATIONS
---------------------------------------------------------- Chapter 0:5

To protect consumers from unreasonable exposure to the residues of
canceled pesticides, GAO recommends that the Administrator of EPA and
the Commissioner of FDA work together to determine, on the basis of
the most recent data, the appropriate action levels for residues of
the five canceled chlorinated pesticides in fish.  GAO also
recommends that the Administrator of EPA periodically reevaluate and
lower recommendations for action levels to reflect decreases in
environmental residue levels.  In addition, GAO recommends that the
Administrator establish procedures for concurrently conducting
tolerance revocation and cancellation actions and, when necessary,
set an effective date for revocation that gives growers enough time
to use existing stocks of the canceled pesticide.  Finally, GAO
recommends that the Administrator identify the pesticides whose
registrations for food uses have already been canceled and revoke
their tolerances. 


   AGENCY COMMENTS
---------------------------------------------------------- Chapter 0:6

GAO discussed the facts and analysis presented in this report with
responsible officials from EPA--including the Deputy Assistant
Administrator, Office of Prevention, Pesticides, and Toxic
Substances--and from FDA--including the Deputy Associate Commissioner
for Regulatory Affairs.  These officials generally agreed with the
information presented but suggested a number of technical and
editorial changes that GAO incorporated where appropriate.  In
particular, EPA officials believed that GAO's presentation of EPA's
data on the health risks posed by residues of canceled chlorinated
pesticides in fish overstated the health risks.  GAO agreed and
revised its presentation of EPA's data to point out the uncertainties
in the data and to include only the information that EPA considered
to be the most valid.  EPA agreed with GAO's revised presentation of
the data. 

GAO also discussed the potential effectiveness of the actions
recommended in this report with EPA and FDA officials.  The EPA
officials agreed that such actions are necessary to resolve the
problems that GAO identified in connection with both action levels
and tolerance revocations.  The FDA officials agreed with the thrust
of GAO's recommendations on action levels.  As requested, GAO did not
obtain written agency comments on a draft of this report. 


INTRODUCTION
============================================================ Chapter 1

Pesticides are used extensively in agricultural production throughout
the world to control or kill insects, fungi, or other pests and to
increase crop yields.  But pesticides can also harm human health and
the environment.  As the types and number of pesticides have grown
over the past 30 years, their effects on health and the environment
have come under closer scrutiny.  And as scientific evaluation has
shown that certain food-use pesticides can cause cancer, birth
defects, and other disorders, the use of some of these pesticides has
been banned, or canceled, in the United States. 

But canceling a pesticide may not eliminate all of its risks,
particularly if its residues persist in the environment or appear on
imported foods.  Hence, federal food safety agencies must decide not
only which pesticides to cancel but also how to regulate the residues
of canceled pesticides that continue to appear in foods.  These
decisions made after a pesticide has been canceled can have important
health and economic implications. 


   FEDERAL AGENCIES SHARE
   RESPONSIBILITY FOR ENSURING
   PESTICIDE SAFETY
---------------------------------------------------------- Chapter 1:1

Federal responsibility for protecting public health and the
environment from unsafe pesticides is shared by the Environmental
Protection Agency (EPA), the Food and Drug Administration (FDA), and
the U.S.  Department of Agriculture (USDA).  Broadly speaking, EPA
sets standards for pesticide safety, which FDA and USDA monitor and
enforce. 


      EPA REGISTERS PESTICIDES AND
      SETS TOLERANCES FOR THEIR
      RESIDUES
-------------------------------------------------------- Chapter 1:1.1

Under the Federal Insecticide, Fungicide, and Rodenticide Act
(FIFRA), EPA registers (licenses) pesticide products for use on
specific crops grown in the United States.  EPA may register a
pesticide if it determines, among other things, that the pesticide
will perform its intended function without causing "unreasonable
adverse effects on the environment." FIFRA defines this term to mean
"any unreasonable risk to man or the environment, taking into account
the economic, social, and environmental costs and benefits" of the
pesticide's use.  EPA may register a single pesticide for multiple
food and nonfood uses. 

Residues of a pesticide used on a food or feed crop can remain on the
food or feed and be ingested with it.  EPA is required under the
Federal Food, Drug, and Cosmetic Act (FFDCA) to establish a
tolerance--or an exemption from a tolerance\1 --for any registered
use of a pesticide on food or animal feed.  A tolerance specifies the
maximum amount of the pesticide's residue that may legally remain in
or on the food or feed.  A single pesticide registered for multiple
food or feed uses must have multiple tolerances. 


--------------------
\1 EPA may exempt a registered food-use pesticide from the
requirement for a tolerance when residues of the pesticide remaining
in food appear to pose no hazard to public health. 


      FDA AND USDA MONITOR
      RESIDUES AND ENFORCE
      TOLERANCES
-------------------------------------------------------- Chapter 1:1.2

FDA and USDA monitor pesticides' residues in foods and feed sold in
interstate commerce using the tolerances established by EPA.  If food
or feed products contain residues of pesticides that have not been
granted tolerances (or exemptions from tolerances) or if residues
exceed tolerances, foods are considered adulterated and are subject
to seizure.  FDA monitors most foods sold in interstate commerce
except meat, poultry, and certain egg products, which are monitored
by USDA.  States are responsible for monitoring foods that are not
sold in interstate commerce. 


      EPA CANCELS REGISTRATIONS
      FOR UNSAFE PESTICIDES
-------------------------------------------------------- Chapter 1:1.3

After registering a pesticide, EPA continues to evaluate its safety,
principally through two major programs--reregistration and special
review.  If EPA finds that a pesticide poses unreasonable risks to
humans or the environment, the agency may cancel the registrations
for some or all of its uses.  A manufacturer may also voluntarily
cancel a pesticide's registrations. 

Under amendments to FIFRA, enacted in 1972 and 1988, EPA is
reevaluating and reregistering thousands of previously registered
pesticide products on the basis of current scientific standards.  In
implementing this mandate, EPA is focusing primarily on about 600
active ingredients that are the main components of the individual
pesticide products.  Although, as we reported in May 1993,\2 EPA will
not complete its reregistration program for many years, many
manufacturers have already canceled the registrations for thousands
of pesticide products containing hundreds of active ingredients
rather than pay the fees or develop the data required to support the
products' reregistration. 

When new evidence indicates that a registered pesticide may pose a
significant health or environmental risk, EPA conducts an extensive
analysis, known as a special review, to determine whether the risks
to human health or the environment exceed the benefits of continued
use.  Through this process, EPA has determined that a number of
active ingredients--and the products containing these active
ingredients--pose unreasonable risks to human health or the
environment and has therefore canceled some or all registrations for
the use of these pesticides.  In many cases, a manufacturer facing
special review has voluntarily canceled a pesticide's registrations. 


--------------------
\2 Pesticides:  Pesticide Reregistration May Not Be Completed Until
2006 (GAO/RCED-93-94, May 21, 1993). 


      EPA TAKES A SEPARATE ACTION
      TO REVOKE TOLERANCES
-------------------------------------------------------- Chapter 1:1.4

In general, after a pesticide's registration for a specific use has
been canceled, the pesticide may no longer be sold or distributed for
that use in the United States.  However, as long as the pesticide's
tolerances remain in effect, foods containing residues of the
pesticide may be sold in the United States.  FDA and USDA may not
classify such foods as adulterated and may not seize and remove them
from commerce. 

Neither FIFRA nor FFDCA requires EPA to revoke a pesticide's
tolerances after it has canceled the pesticide's registrations. 
However, in 1982, after consulting with FDA and USDA, EPA adopted a
policy on revoking the tolerances for canceled pesticides in which it
stated that "when a pesticide's registration for a food or feed use
is canceled because of a concern about the safety of the pesticide,
the associated tolerance .  .  .  is no longer justified and
logically should be revoked." EPA added that "the agencies are
concerned that having formal tolerances remaining in effect for
canceled pesticides may serve to condone use of these pesticides in
this country and/or in or on commodities imported from foreign
countries."


      EPA MAY RECOMMEND ACTION
      LEVELS FOR MONITORING
      UNAVOIDABLE RESIDUES
-------------------------------------------------------- Chapter 1:1.5

Although most pesticides break down fairly quickly in the
environment, some pesticides degrade very slowly and persist in the
environment long after their use has ended.  Hence, residues of these
pesticides may be unavoidable in certain foods.  To provide standards
for regulating these unavoidable residues in foods, EPA recommends
action levels, which FDA and USDA have agreed to establish for use in
monitoring and enforcement, in accordance with EPA's 1982 policy.  In
recommending action levels, EPA's policy requires the agency to
assess health risks as well as the extent to which residues are
unavoidable in foods and to periodically lower action levels as
residues of canceled pesticides decline in the environment. 

Like a tolerance, an action level specifies the maximum amount of a
pesticide's residues that may be allowed in or on a food or feed. 
However, an action level is established only for residues that are
considered unavoidable in a certain food.  An action level may be
established to take the place of a tolerance that has been revoked. 
An action level may also be established for a pesticide's unavoidable
residues in a food for which a tolerance was never set because the
pesticide was never registered for use on that food.  FDA issued a
notice in the Federal Register (55 Fed.  Reg.  14359, Apr.  17, 1990)
explaining how the agency would use action levels.  FDA stated that,
according to FFDCA, "in the absence of a tolerance, any amount of a
pesticide residue in a food or feed is unsafe and therefore renders
the food or feed adulterated." But when a food or feed is unavoidably
contaminated with certain persistent pesticides that do not have
tolerances, FDA said that it would use action levels to provide
guidance for determining when enforcement action was warranted. 

Most of the action levels that EPA has proposed or recommended have
been for a group of chlorinated compounds--including DDT, chlordane,
and dieldrin--that were widely used in U.S.  agriculture during the
1950s and 1960s.  Because these compounds were later found to pose
unacceptable chronic health risks to humans and to affect
reproduction and cause birth defects in wildlife, most of their
registrations were canceled during the 1970s.  However, unlike most
pesticides, these compounds have not readily broken down.  Today,
they are still found in soil, sediment, and water. 

Chlorinated compounds are not highly concentrated in plants, but they
are accumulated in other organisms, particularly in fish, which are
at or near the top of the aquatic food chain.  Unlike the herbivorous
land animals eaten by humans, fish are often predators.  When they
prey on other aquatic animals, they may ingest and accumulate
compounds that their prey have already accumulated.  According to
EPA, aquatic organisms may accumulate environmental contaminants in
concentrations up to 1 million times greater than are found in the
surface water from which the organisms are taken.  Although these
chlorinated pesticides were never registered for use on fish, they
have been found in fish for decades, largely because agricultural
runoff transported the pesticides to the nation's rivers and lakes. 
Since these pesticides did not have tolerances for fish, FDA
established action levels as guidelines for determining when
enforcement action was warranted. 

According to EPA, most foods and feeds either contain no detectable
residues of these canceled pesticides or contain residues that are
well below the recommended action levels.  Therefore, EPA believes
that the dietary risk from these canceled pesticides in most foods is
low.  But because of the relatively high potential for these
persistent pesticides to be concentrated in fish, health risks from
dietary exposure to these canceled pesticides are greater in fish
than in other foods. 


      EPA HAS ESTABLISHED A
      PROCESS FOR REVOKING
      TOLERANCES AND RECOMMENDING
      ACTION LEVELS
-------------------------------------------------------- Chapter 1:1.6

When EPA decides to revoke a pesticide's tolerances, it first
verifies that all of the registrations associated with these
tolerances have been canceled.  It then reviews monitoring data to
determine whether and to what extent residues of the canceled
pesticide remain in foods and whether action levels are needed to
replace the existing tolerances.  EPA also analyzes the economic
impact of revoking tolerances on domestic food producers and on
imported commodities.  Then, EPA prepares and issues a preliminary
notice in the Federal Register stating its intent to revoke certain
tolerances and requesting comments from interested parties.  If
action levels are needed, EPA specifies what levels it intends to
recommend to FDA or USDA.  After the 60-day comment period has
expired, EPA issues a final notice in the Federal Register announcing
the effective date of the tolerances' revocation and, if necessary,
the final recommended action levels. 


   OBJECTIVES, SCOPE, AND
   METHODOLOGY
---------------------------------------------------------- Chapter 1:2

Concerned that residues of canceled pesticides in food continue to
pose a health risk to U.S.  consumers, the Chairman, Environment,
Energy, and Natural Resources Subcommittee, House Committee on
Government Operations, asked GAO to (1) determine whether marketed
foods contain unsafe levels of residues from canceled pesticides and
(2) evaluate EPA's procedures for revoking tolerances for canceled
food-use pesticides. 

To determine whether marketed foods contain unsafe levels of residues
from canceled pesticides, we focused on health risks from fish
contaminated with residues of five canceled pesticides--DDT,
chlordane, dieldrin, heptachlor, and mirex.  We focused on these five
canceled pesticides because--unlike most other pesticides--they are
highly persistent in the environment and because EPA, the National
Academy of Sciences, and other organizations agree that they pose
significant health risks through dietary exposure.  We focused on
fish because fish are more likely than most other foods to accumulate
residues of these canceled pesticides.  In addition, data on the
health risks of residues from other canceled pesticides on food
commodities are sparse.  But data on the health risks of residues
from these five canceled pesticides in fish were readily available
because EPA had assessed these risks in response to concerns
expressed by Members of Congress, EPA regional officials, and
environmental organizations. 

To evaluate EPA's basis for proposing action levels for the residues
of five canceled pesticides in fish, we reviewed EPA documents and
Federal Register announcements on establishing action levels.  We
also reviewed EPA's study of residue levels and health risks for
these pesticides, as well as EPA's analyses of the economic impact of
lowering the action levels for residues of the five canceled
pesticides in fish. 

To determine the trends in residue levels for these canceled
pesticides, we reviewed FDA's pesticide monitoring data for fish and
fishery products and FDA's total diet studies for all food
commodities, from 1984 through 1992, for detections of DDT,
chlordane, and dieldrin.  We compared the current action levels with
the actual residue levels detected in fish tested by FDA. 
Specifically, we compared the current action level for dieldrin
residues in whitefish with the average dieldrin residues that FDA
detected in testing domestic whitefish from 1984 through 1992. 

To demonstrate the effect of federal action levels on consumers of
fish that are not tested by FDA because they do not enter interstate
commerce, we interviewed EPA Office of Water officials and reviewed
EPA studies on (1) the basis that states use to establish fish
advisories, (2) the levels of contamination from pesticides and other
chemicals that are found in fish nationwide, and (3) the guidance
that the Office of Water provides to states on establishing fish
advisories.  We also examined a 1991 National Academy of Sciences
study of seafood safety, which discusses the extent to which fish are
contaminated by pesticides and other chemicals and the actions that
are needed to better inform consumers of the potential risks of
eating certain fish.  In addition, we contacted several state health
and environmental officials to find out how federal action levels
affect their regulation of pesticides in fish and to determine the
extent of their efforts to monitor these pesticides. 

To evaluate EPA's process for revoking the tolerances for canceled
pesticides, we interviewed key EPA officials who either were or had
been involved in the revocation process, and we collected documents
explaining EPA's revocation policy and procedures and showing the
status of EPA's revocation efforts.  We also examined the Federal
Register notices for all pesticides whose tolerances had been revoked
as of July 1994 to determine how many tolerances had been revoked and
how much time had elapsed between cancellation and revocation.  In
addition, we reviewed EPA documents and information systems to
determine how many canceled pesticides still have tolerances.  When
we reviewed data from EPA's information systems on canceled food-use
pesticides, we found that much of the information was unreliable and,
therefore, could not be used.  To identify canceled food-use
pesticides that still have tolerances, we examined EPA planning
documents and reregistration status reports and asked EPA officials
to verify the information.  In addition, we examined FDA fiscal year
1992 monitoring data to determine whether some canceled pesticides
that still had tolerances were appearing in the U.S.  food supply. 

We conducted our review between April 1993 and September 1994 in
accordance with generally accepted government auditing standards. 

We discussed the facts and analysis presented in this report with
responsible officials from EPA--including the Deputy Assistant
Administrator, Office of Prevention, Pesticides, and Toxic
Substances; the Deputy Director, Office of Pesticide Programs; and
the Director, Policy and Special Projects Staff, Office of Pesticide
Programs--and from FDA--including the Deputy Associate Commissioner
for Regulatory Affairs; the Director, Office of Policy, Planning, and
Strategic Initiatives, Center for Food Safety and Applied Nutrition;
and the Director, Contaminants Policy Staff, Office of Regulatory
Affairs.  These officials generally agreed with the information
presented but suggested a number of technical and editorial changes
that we incorporated where appropriate.  As requested, we did not
obtain written agency comments on a draft of this report. 


UNSAFE RESIDUES MAY BE PRESENT IN
SOME FISH
============================================================ Chapter 2

According to EPA, most pesticides break down fairly quickly in the
environment and therefore do not appear at significant levels in most
foods.  But a few pesticides whose registrations were canceled about
20 years ago have persisted in the environment.  EPA believes that
residues of these pesticides are present at low levels in most foods. 
However, they are found in some fish at levels that exceed EPA's
usual negligible risk standard.  The action levels currently used to
regulate residues of these canceled pesticides in fish do not meet
criteria in EPA's 1982 policy because they (1) are not based on an
assessment of health risks and (2) have never been adjusted to
reflect declines in residue levels that have occurred since FDA first
set the action levels in the late 1960s and early 1970s. 

In 1991, after conducting a study that evaluated health and economic
effects and more recent residue data, EPA proposed lower action
levels to FDA.  While FDA agreed that action levels should be lowered
to reflect declines in residues of these pesticides in fish, it
believed that EPA's proposed levels would represent too great a
reduction.  Despite their shared belief that action levels should be
lowered, neither agency has since taken any action to reach agreement
on appropriate lower action levels. 


   EPA'S POLICY REQUIRES
   ASSESSMENT OF HEALTH RISKS AND
   DECLINES IN RESIDUES
---------------------------------------------------------- Chapter 2:1

In its 1982 policy on revoking the tolerances for canceled
pesticides, EPA established principles for recommending action levels
to FDA and USDA that emphasized the importance of assessing health
risks and actual residue levels in food.  Under the policy, action
levels would "be set limiting the quantity of a pesticide in or on
food commodities to the extent necessary to protect the public
health." Although action levels would "tak[e] into account the extent
to which the contaminant is unavoidable," they would be "sufficient
to protect the public health." In some instances, according to the
policy, the health risk for a given pesticide could be so great that
no residue level would be acceptable.  In these instances, the policy
calls for EPA to recommend action levels that do not exceed levels
that FDA can detect using its current testing methods.  Finally, the
policy stated that EPA would periodically review action levels and
lower them as residues of canceled pesticides in food declined. 


   CURRENT ACTION LEVELS ARE NOT
   BASED ON EPA'S POLICY CRITERIA
---------------------------------------------------------- Chapter 2:2

In 1985, EPA placed preliminary notices in the Federal Register
announcing its intention to revoke the tolerances for DDT, chlordane,
and dieldrin, whose registrations for food uses it had canceled
during the 1970s.  Because these pesticides' residues persisted in
the environment, EPA, in accordance with its 1982 policy statement,
also proposed action levels to replace the revoked tolerances.  For
fish, EPA proposed to retain the action levels that FDA had been
using since the late 1960s and early 1970s. 


      EPA BASED 1985 ACTION LEVEL
      PROPOSALS ON OLD RESIDUE
      DATA
-------------------------------------------------------- Chapter 2:2.1

In developing the action levels that it proposed for DDT, chlordane,
and dieldrin in 1985, EPA primarily reviewed residue data from the
1970s, which indicated that residue levels in fish had not declined
much since the action levels were originally established.  In
addition, EPA did not assess the health risks posed by these
residues, as directed in its 1982 policy.  In response to concerns
about DDT expressed in a 1986 congressional hearing, EPA said that it
had not assessed health risks when it proposed action levels to
replace the tolerances or existing action levels for this pesticide. 
Instead, EPA reviewed FDA's pesticide monitoring data and proposed
action levels that reflected the actual levels of DDT residue found
in foods monitored during the late 1970s. 

EPA officials told us that the proposed action levels for DDT and
other chlorinated pesticides were set at a level high enough so that
most--about 95 percent--of the residues found in foods would be at or
below the action levels.  EPA reasoned that the residues of these
canceled pesticides were unavoidable and had not entered the food
supply through the misuse of pesticides.  Therefore, the agency did
not want to penalize food producers for past legal uses of the
pesticides. 

In response to its preliminary notices, EPA received no significant
comments on the adequacy of the action levels it had proposed for
foods other than fish.  But a number of commenters--including two EPA
regions--questioned the safety of the action levels proposed for
fish.  EPA's Region VII noted that EPA apparently had not reviewed
available health effects data, as required by its 1982 policy, to
assess the safety of the proposed action levels.  Similarly, EPA's
Region V commented that EPA had not assessed the effects of the
proposed action levels on human health.  According to several
commenters, the risk of cancer under the proposed action levels was
far greater than the agency's risk standards usually allowed.  For
example, although EPA typically applies a negligible risk standard
for cancer when regulating pesticides in food,\3 EPA's Region VII
stated that the risk of cancer under the proposed action level for
chlordane was 1 in 22,000, and the National Wildlife Federation
estimated that the risk of cancer for dieldrin was 1 in 1,000. 

In commenting on the proposed action levels for fish, FDA maintained
that the existing action levels might need to remain in effect
because high levels of DDT residue were still being found in fish
from at least one part of the United States.  But FDA also said that
EPA needed to study the health effects of these pesticides in light
of the comments it received on its proposed action levels. 

In 1986, EPA issued final notices in the Federal Register in which it
revoked the tolerances for DDT, chlordane, and dieldrin in foods
other than fish and recommended action levels to replace the
tolerances.  But in response to concerns over the level of risk that
would still be allowed under the proposed action levels for residues
of these three canceled pesticides in fish, EPA announced that it
would wait to recommend action levels for fish until it could obtain
updated residue data and assess the health effects of alternative
action levels.  Later, EPA added two other canceled
pesticides--heptachlor and mirex--to its study of action levels for
fish (see app.  I). 


--------------------
\3 In practice, EPA has generally defined negligible risk from
dietary exposure to pesticides as an incremental increase of 1 in 1
million or lower in the lifetime risk of cancer as calculated
according to a conservative risk-assessment methodology. 


      EPA CONDUCTED A STUDY TO
      ASSESS RECENT RESIDUE DATA,
      HEALTH RISKS, AND ECONOMIC
      EFFECTS
-------------------------------------------------------- Chapter 2:2.2

To determine whether action levels should be revised, EPA conducted a
study in which it reviewed recent residue data and evaluated health
risks, as prescribed in its 1982 policy.  In addition, EPA evaluated
the economic effects of lower action levels. 

To conduct its study, EPA obtained recent exposure information by
collecting national and regional data on residues of DDT, chlordane,
dieldrin, heptachlor, and mirex in fish and compiled a data base from
tests of about 11,000 samples conducted between 1983 and 1987 by FDA,
EPA regions, state agencies and other federal agencies.  Using these
data, EPA estimated the risk of cancer to consumers of fish at
various action levels. 

To assess the economic effects of alternative action levels for fish,
EPA projected the percentage of the fish catch that would exceed
lower action levels and estimated the costs to commercial fisheries
of not being able to sell these fish. 


         EPA'S STUDY SHOWED LOWER
         RESIDUE LEVELS
------------------------------------------------------ Chapter 2:2.2.1

EPA's analysis of the monitoring data showed that, for the five
canceled pesticides, residue levels in fish generally appeared to be
declining.  However, in certain locations, residue levels appeared to
remain constant or to increase over time.  EPA attributed the
apparent increases in residue levels in fish to the occasional
stirring up of sediment and releasing of residues from the sediment
into the water or to methodological issues, such as variation in the
size of the fish sampled and nonrepresentative sampling.  But EPA
said that action levels could be lowered to reflect the generally
declining levels of these pesticides' residues in fish. 

We examined FDA's records of dieldrin detections in whitefish from
1984 through 1992 to determine the trends in residue levels.  As
shown in figure 2.1, FDA's records indicate that residue levels
declined from an average of 0.313 parts per million (ppm) in 1984 to
0.120 ppm in 1992.  Since 1984, average dieldrin residue levels have
been consistently below the current action level of 0.3 ppm. 
However, FDA's records also show that between 1990 and 1992 residues
of dieldrin in fish have remained steady or have increased slightly. 
According to EPA officials, this slower rate of decline indicates
that despite the general decline in residues of canceled chlorinated
pesticides in fish, these residues may continue to appear at
significant levels in some fish for a number of years to come. 

   Figure 2.1:  Average Dieldrin
   Residue Levels in Domestic
   Whitefish, 1984-92

   (See figure in printed
   edition.)

Source:  GAO's presentation of FDA's data. 


         EPA'S STUDY INDICATED
         HEALTH RISKS EXCEEDING
         THE AGENCY'S USUAL
         STANDARD
------------------------------------------------------ Chapter 2:2.2.2

EPA's Office of Pesticide Programs calculated the risks of cancer at
current action levels to consumers of average amounts of fish that
are sold in interstate commerce and that contain average levels of
the five canceled pesticides' residues.  These calculations,
presented in table 2.1, were based on national and regional data
collected by EPA from federal agencies, states, and EPA regions. 
EPA's analysis of the data showed that, at current action levels for
the five pesticides, the dietary risks of cancer exceed the agency's
usual standard of negligible risk (1 in 1 million). 



                          Table 2.1
           
             Estimated Risks of Cancer at Current
                        Action Levels

                                           Risk of cancer at
                                              current action
Pesticide                                            level\a
----------------------------------------  ------------------
Dieldrin                                    100 in 1 million
Heptachlor                                   39 in 1 million
Chlordane                                    16 in 1 million
DDT                                           9 in 1 million
Mirex                                         2 in 1 million
------------------------------------------------------------
Note:  To calculate these risks, EPA used the average level of
residue in fish containing these pesticides at or below the current
action levels.  Three species of fish--tuna, cod, and salmon--account
for most of the calculated risk either because consumption of these
fish is high or because the level of residue in these fish is high. 
The data analyzed were not representative of national fish
consumption and were derived, to a great extent, from fish that were
likely to be more contaminated.  For samples containing no detectable
residues, it was assumed that residues were present at the limit that
can be detected using current testing methods. 

\a Number of additional cases of cancer expected per 1 million
persons assuming consumption of 15 grams per day for 70 years and an
average body weight of 70 kilograms. 

Source:  GAO's presentation of EPA's data. 

According to EPA, the figures shown in table 2.1 could either
overestimate or underestimate risks, depending on the extent to which
actual exposure differs from the assumptions used in the calculations
(see note following table 2.1).  For consumers of average amounts of
fish sold in interstate commerce, the figures may significantly
overstate risks.  For example, for dieldrin, EPA said that if it had
used less conservative assumptions for samples in which no residues
were detected, the risks calculated for this pesticide would have
been only 2 in 1 million rather than 100 in 1 million.  According to
EPA, more accurate estimates of risk were not possible using
available data. 

But EPA also noted that the actual risks could be considerably higher
than the average risks shown in the table for consumers of larger
amounts of fish or of fish that are more highly contaminated with
residues of these pesticides.  For example, although the calculations
of risk in table 2.1 assume consumption of 15 grams of fish per day,
EPA has estimated that typical recreational fishermen consume 30
grams per day and subsistence fishermen consume 140 grams per day, on
average.  These levels of consumption are about two to nine times
greater than the levels EPA used to calculate the risks shown in
table 2.1.  Because risks are proportional to consumption, consumers
of larger amounts of fish could be exposed to proportionately higher
risks than are shown in the table. 


         EPA ESTIMATED THE
         ECONOMIC EFFECTS OF LOWER
         ACTION LEVELS
------------------------------------------------------ Chapter 2:2.2.3

EPA computed the economic costs to commercial fisheries of
implementing lower action levels, taking into account the estimated
loss of nationally and regionally important fish species.  The agency
calculated a potential annual economic loss to commercial fisheries
of either $74.3 million or $272.7 million, depending upon the action
levels considered and assuming that FDA would identify and remove
from commerce all fish that exceeded the action levels.  For fish
species considered to have national or local economic importance, EPA
also estimated the percentage of fish that would exceed the current
action levels and the lower action levels.  These estimates indicated
that a significantly greater percentage of fish would exceed the
proposed lower action levels than would exceed the current action
levels.  For example, although none of the herring catch would exceed
the current action level for dieldrin, 17 percent of the catch would
exceed the lower action level for dieldrin that EPA proposed in 1991. 
Similarly, while 17 percent of the sablefish catch would exceed the
current action level for DDT, 25 percent would exceed the proposed
lower action level. 


      EPA'S STUDY INDICATED LOWER
      ACTION LEVELS
-------------------------------------------------------- Chapter 2:2.3

In 1991, after reviewing its data on residues of the five pesticides
in about 11,000 fish samples, assessing the health risks of each
pesticide, and calculating the economic costs to commercial fisheries
of implementing alternative action levels, EPA sent FDA a draft
proposal to lower the action levels for residues of DDT, chlordane,
dieldrin, heptachlor, and mirex in fish.  These lower action levels
are presented in table 2.2 along with the current action levels used
by FDA.  EPA officials told us that these action levels represent the
agency's balancing of health and economic effects, taking into
account the unavoidability of these residues in fish. 



                          Table 2.2
           
              Comparison of Current and Proposed
                        Action Levels

             (Action levels in parts per million
                            (ppm))


                                                          to
                                              Currentl   FDA
                                                y used    in
Pesticide                                       by FDA  1991
--------------------------------------------  --------  ----
DDT                                                5.0  0.50
Dieldrin                                           0.3  0.10
Chlordane                                          0.3  0.10
Mirex                                              0.1  0.05
Heptachlor                                         0.3  0.02
------------------------------------------------------------
Source:  GAO's presentation of EPA's data. 


      EPA AND FDA HAVE NOT AGREED
      ON LOWER ACTION LEVELS FOR
      FISH
-------------------------------------------------------- Chapter 2:2.4

One year after EPA sent the draft proposal, EPA and FDA officials met
to discuss these action levels.  According to EPA's Deputy Assistant
Administrator for Prevention, Pesticides, and Toxic Substances, staff
from both agencies believed that residues of the five pesticides had
been declining in the environment and that lower action levels would
therefore be appropriate.  FDA agreed to review its pesticide
monitoring data to see how much residues had declined and whether its
data could support the lower action levels proposed by EPA. 

In September 1992, FDA concluded, after reviewing its monitoring data
for 1989 to 1991, that EPA's lower action levels would greatly
decrease the allowable catch from the Great Lakes and a number of
southern and western states.  FDA said that EPA would therefore have
to demonstrate and document the need for the lower action levels to
protect consumers and show that the lower levels took into account
the unavoidability of residues in fish. 

In May 1994, the Director of FDA's Contaminants Policy Staff told us
that EPA did not justify its proposed action levels to FDA.  He said
that FDA believes that residues of canceled chlorinated pesticides in
fish have generally declined since the action levels were originally
established and that the action levels should be lowered to reflect
this decline.  But, according to the official, EPA's proposed lower
action levels were significantly lower than they would be if they
were based only on declines in residues in fish.  Therefore, FDA
needed adequate justification to explain to consumers and commercial
fisheries the basis for the stricter standards and their potential
economic impact. 

In May 1994, the Director of EPA's Pesticide Registration Division
told us that because of budgetary constraints, EPA has no foreseeable
plans to obtain additional documentation to satisfy FDA's concerns. 
He said that EPA considers its data sufficient to justify the lower
action levels.  However, EPA has not formally recommended the lower
action levels to FDA.  Hence, despite their agreement that the action
levels should be lowered, neither agency has taken the initiative to
reach agreement on appropriate lower action levels. 


   MANY STATES USE FEDERAL ACTION
   LEVELS TO REGULATE FISH
---------------------------------------------------------- Chapter 2:3

Although federal action levels are based on national rather than
regional or local data, many states use the federal action levels as
their basis for determining when to issue fish consumption
advisories.\4 In 1990, EPA's Office of Water reported that two-thirds
of the states (34) were using federal action levels as their basis
for evaluating the safety of chemical contaminants in fish.  Other
states were using a risk-assessment approach derived from EPA's
criteria or had developed their own approach. 

According to a 1991 National Academy of Sciences report on seafood
safety, fish caught for recreation or subsistence may pose greater
health risks than fish sold in interstate commerce because such fish
are more likely to be caught near areas contaminated with hazardous
chemicals (including pesticides)\5

and may be consumed in greater quantities by certain subpopulations. 
The Academy reported that recreationally harvested fish may represent
over one-fifth of the fish consumed in the United States and that
these fish are caught by an estimated 17 million recreational
fishermen. 

The Academy noted that state regulatory agencies are almost
exclusively responsible for issuing seafood health advisories.  But
it said that states depend heavily on federal guidance in regulating
seafood, and this guidance may not take into account specific
regional variations in seafood safety.  The Academy suggested that "a
more consistent and focused effort in the determination and
communication of public health risks from contaminated seafood should
be developed" and that "a more pronounced and consistently defined
federal role in the risk characterizations leading to these [seafood
health] advisories would be [beneficial]."

In 1992, EPA's Office of Water completed a study of chemical residues
in fish that revealed widespread contamination by pesticides and
other chemicals in fish.  Concerned about this contamination and
about states' inconsistent procedures for sampling fish and issuing
fish consumption advisories, the Office of Water issued guidance to
the states in 1993 and 1994 to assist them in developing a risk-based
approach for monitoring fish and determining when fish advisories
should be issued (see app.  II for more detailed information on the
study and guidance).  Office of Water representatives told us that it
was too soon to evaluate the impact of this guidance.  They did not
know of any states that had used the guidance to strengthen their
monitoring standards for pesticides' residues in fish. 

An Office of Water official also said that a number of states are not
active in monitoring fish and issuing fish advisories, principally
because they lack funding.  He said that other states, such as New
York, recognize that the federal action levels are designed for FDA's
regulation of fish in interstate commerce but nevertheless continue
to use the federal action levels in their own regulatory programs. 

A New York State environmental health official told us that although
her agency believes that a risk-based monitoring approach would
protect consumers' health better than action levels, the agency is
reluctant to move toward risk-based monitoring.  The official
explained that New York has a number of commercial fisheries whose
catches are subject to FDA's monitoring.  Because pesticides and
other chemicals in these fish do not generally exceed the federal
action levels, the fish are sold legally in interstate commerce.  The
official said that, in view of FDA's monitoring, the state believes
that it would face an untenable position if it were to adopt more
stringent risk-based monitoring standards for fish caught and
consumed within the state.  At the same time as the state was trying
to justify stricter standards for fish that would be caught and
consumed within New York, the official said, FDA would be allowing
the same species of fish, with the same levels of chemical
contamination, to be sold nationwide without any warnings or
advisories. 

EPA's Region V also noted that a number of states are reluctant to
apply different state and federal standards in monitoring the safety
of pesticides' residues in fish.  The region is concerned that if the
action levels for pesticides' residues in fish are not lowered, then
the states will not issue more protective fish consumption
advisories. 

The Director of FDA's Contaminants Policy Staff agreed that states
might have difficulty explaining differences between federal and
state monitoring levels to local consumers and commercial fisheries. 
Nevertheless, he said that FDA could not set enforcement limits for
local conditions because the action levels enforced by FDA apply
nationwide to fish in interstate commerce.  The Director noted that a
state or locality might issue guidance on the amount of contaminated
fish that consumers might eat without appreciable risk to their
health. 


--------------------
\4 Fish consumption advisories are issued by many states to provide
information to the public about contaminated fish from particular
bodies of water that may have unacceptable effects on health if
consumed above certain levels. 

\5 According to EPA, most recreational fishing is done in freshwater
where contamination of fish with pesticides is generally greater than
in saltwater. 


   CONCLUSIONS
---------------------------------------------------------- Chapter 2:4

Although EPA believes that residues of the five persistent pesticides
that it studied do not appear in most foods at significant levels, it
has found that they appear in some fish at relatively high levels. 
The current action levels for these pesticides in fish, which FDA
established about two decades ago, are based on residue levels found
in the environment at that time.  They have not been adjusted to
reflect health risk assessments or subsequent declines in residue
levels.  Consequently, they are not consistent with the 1982 policy
that calls upon EPA, when recommending action levels, to assess
health risks as well as unavoidable residues and to revise its
recommendations periodically as residue levels decline. 

The action levels that EPA proposed to FDA in 1991 are based on EPA's
assessment of residue data, health risks, and economic effects. 
Hence, these action levels were developed in accordance with the
requirements of the 1982 policy.  Both EPA and FDA agree that the
action levels should be lowered but disagree on the extent to which
they should be lowered on the basis of available data.  We do not
believe that the differences between EPA and FDA over the sufficiency
of EPA's data should block attempts by the agencies to reach
agreement on appropriate action levels.  Reaching agreement on
appropriate action levels would help to ensure that consumers of both
federally monitored and state-monitored fish are being adequately
protected. 


   RECOMMENDATIONS
---------------------------------------------------------- Chapter 2:5

To protect consumers from unreasonable exposure to the residues of
canceled pesticides, we recommend that the Administrator of EPA and
the Commissioner of FDA work together to determine, on the basis of
the most recent data, the appropriate action levels for residues of
the five canceled chlorinated pesticides in fish.  We also recommend
that the Administrator of EPA periodically reevaluate and lower
action level recommendations to reflect decreases in environmental
residue levels. 


   AGENCY COMMENTS
---------------------------------------------------------- Chapter 2:6

EPA and FDA officials generally agreed with the information presented
in this chapter but suggested a number of technical and editorial
changes that we incorporated where appropriate.  In particular, EPA
officials believed that our presentation of EPA's data on health
risks posed by residues of canceled chlorinated pesticides in fish
overstated the health risks.  We revised our presentation of EPA's
data to highlight the uncertainties in the data and to include only
the information that EPA considered to be the most valid. 

We also discussed with these officials the potential effectiveness of
the actions that we recommend in this report.  The EPA officials
agreed that actions such as we recommend are necessary to resolve the
problems we identified in connection with action levels.  The FDA
officials agreed with the thrust of our recommendations on action
levels. 


EPA IS SLOW TO REVOKE TOLERANCES
FOR CANCELED PESTICIDES
============================================================ Chapter 3

In canceling the registrations for many food-use pesticides during
the past two decades, EPA has not concurrently revoked the related
tolerances for these pesticides.  Although EPA has recently taken
action to revoke the tolerances for some older canceled pesticides,
an undetermined but potentially large number of canceled pesticides
still have tolerances. 

On average, EPA has taken over 6 years to revoke a pesticide's
tolerances after canceling the pesticide's registrations.  Although
part of this delay is intended to allow food treated with remaining
stocks of a canceled pesticide to clear the channels of trade, a
greater part is attributable to the low priority that EPA has
assigned to revocation and to the absence of procedures linking
revocation to cancellation. 

As long as the tolerances for canceled pesticides remain in effect,
foods containing allowable amounts of these pesticides' residues can
legally enter the U.S.  food supply.  FDA and USDA cannot consider
such foods adulterated and cannot take enforcement action against
them. 


   MANY TOLERANCES HAVE BEEN
   REVOKED, BUT MANY MORE REMAIN
---------------------------------------------------------- Chapter 3:1

Over the past few years, EPA has stepped up efforts to revoke the
tolerances for pesticides whose registrations for food uses were, for
the most part, canceled during the 1980s.  As of July 1994, EPA had
revoked the tolerances for 50 canceled pesticides\6 and had formally
proposed to revoke the tolerances for 31 canceled pesticides (see
apps.  III and IV).  According to EPA officials, these revocations
have dealt with the major pesticides that pose a dietary risk to the
public, such as DDT, chlordane, and toxaphene.  Most of these
revocation actions occurred during the past 2 years. 

EPA has not been able to determine how many more canceled pesticides
have tolerances that should be revoked because its data bases do not
identify all pesticides whose registrations for some or all food uses
have been canceled.  But an EPA official responsible for revocations
estimated that over 100 pesticides may fall into this category and
that hundreds of associated tolerances remain in effect for these
canceled pesticides.  This official believes that the food-use
registrations for most of these pesticides have been canceled for
over 2 years. 

In the past, EPA tried to define the universe of canceled pesticides
that still had tolerances to be revoked.  For example, in early 1992,
EPA identified 98 pesticides whose tolerances it considered
"probable" candidates for revocation.  Over 2 years later, EPA has
not begun to revoke the tolerances for nearly half of these
pesticides, and the agency is again attempting to identify the
canceled pesticides that still have tolerances to be revoked. 

Though difficult to determine because of deficiencies in EPA's data
bases, the number of canceled pesticides that still have tolerances
may be sizable.  From EPA's List A--the group of pesticides assigned
the highest priority for reregistration--we identified 10 pesticides
whose registrations for food uses were all canceled between 3 and 13
years ago.  There are 185 tolerances that remain in effect for these
canceled pesticides.  For one of the pesticides, cyhexatin, all of
the registrations were canceled voluntarily by its manufacturers in
1987 after EPA considered initiating a special review in response to
concerns over the pesticide's potential to cause birth defects. 
However, as of July 1994, EPA had not begun to revoke cyhexatin's
tolerances. 

As of April 1994, EPA had evaluated approximately 100 out of about
600 active ingredients that are undergoing reregistration.  As it
continues its evaluations through reregistration and special review,
the registrations for other pesticides are likely to be canceled and
their tolerances will then need to be revoked. 


--------------------
\6 In addition to these revocations, EPA had, as of July 1994,
revoked the tolerances for 27 other pesticides for reasons other than
the cancellation of their food-use registrations or for reasons that
could not be determined because of the pesticides' uncertain
regulatory history.  EPA had also revoked the tolerance exemptions
for seven inert ingredients used in pesticide products. 


   EPA ASSIGNS REVOCATION A LOW
   PRIORITY AND HAS NO PROCEDURES
   LINKING REVOCATION TO
   CANCELLATION
---------------------------------------------------------- Chapter 3:2

Since 1982, when it issued its policy on revoking tolerances, EPA has
taken over 6 years, on average, to revoke a pesticide's tolerances
after canceling the pesticide's registrations.  Typically, the agency
has allowed some time--usually about 2 years--for remaining stocks of
a pesticide to be used and for products legally treated with the
pesticide to move through commerce.  But most of the delay in
revocation can be attributed to both the low priority that EPA has
assigned to revocation and the absence of procedures linking
revocation to cancellation.  The regulatory history of the pesticide
bufencarb illustrates this pattern:  Although EPA had canceled all
registrations for food uses of bufencarb by April 1986, it did not
propose to revoke the related tolerances until August 1992, and it
did not complete their revocation until June 1993, over 7 years
later. 


      EPA ASSIGNS LOW PRIORITY TO
      REVOCATION
-------------------------------------------------------- Chapter 3:2.1

EPA assigns low priority to revoking tolerances, in part because
revocation is not required by law.  When the agency first canceled
registrations for food-use pesticides during the 1970s, it had no
mandate or guidance directing it to revoke the related tolerances. 
Although EPA's 1982 policy established the principle that revocation
should coincide with cancellation, it created no time frames for
revocation or impetus for linking revocation to cancellation. 

EPA officials told us that the agency assigns low priority to
revoking tolerances because many of the food-use pesticides have been
canceled voluntarily or pose little or no dietary risk.  But the fact
that a pesticide has been canceled voluntarily does not necessarily
mean that it poses no dietary risk.  In addition, EPA has delayed
revocation for almost all pesticides, including those considered to
pose dietary risks.  For example, long-term exposure through the diet
to the pesticide heptachlor could cause cancer in humans, according
to EPA, yet the agency revoked the tolerances for this pesticide more
than 11 years after canceling its registrations for food uses. 

The low priority assigned to revocation is reflected in the limited
resources allocated to it.  The EPA official responsible for
revocation said that none of the 13 staff in her unit works full-time
on revocation actions and slightly fewer than 3 full-time-equivalent
staff per year are allocated for such actions.  According to this
official, her unit has so many other higher-priority responsibilities
that it can handle only a limited number of revocation actions at one
time.  While recognizing that the tolerances for many other canceled
pesticides still required revocation, she said that she currently did
not have the time available to identify these pesticides.  Most of
the EPA personnel involved in revocation told us that revocation
activities have a lower priority than their other responsibilities. 

To conserve the limited resources that it has allocated to
revocation, EPA usually delays the revocation of a pesticide's
tolerances until it has canceled all of the pesticide's registrations
for all of its food uses.  Thus, it avoids taking multiple revocation
actions for a single pesticide.  For example, EPA canceled all
registrations for insecticidal uses of the pesticide sodium arsenite
in 1988 because of concerns about the pesticide's toxic effects on
workers and the general public.  But the agency did not propose to
revoke the tolerances for these uses until the registration for one
remaining food-use--as a fungicide on grapes--was canceled in 1992,
about 4 years later. 


      EPA DOES NOT HAVE FORMAL
      PROCEDURES FOR LINKING
      REVOCATION TO CANCELLATION
-------------------------------------------------------- Chapter 3:2.2

Despite its policy supporting a link between revocation and
cancellation, EPA has not developed written procedures or guidelines
specifying when it should revoke a pesticide's tolerances for
canceled food uses.  As a result, the agency has taken anywhere from
a few months to over 14 years to revoke the tolerances for individual
pesticides.  Without such procedures, EPA is under no pressure to
revoke tolerances in a predictable or consistent way. 

In addition, EPA has no written procedures or guidelines requiring
the officials responsible for handling cancellations to notify the
officials responsible for revoking the related tolerances of any
cancellations.  Consequently, the personnel responsible for
revocations often do not receive information about cancellations on a
timely and consistent basis and in a standard format that provides
all of the information needed to revoke tolerances.  Without reliable
channels of information and communication between the personnel
responsible for cancellations and the personnel responsible for
revocations, EPA cannot effectively implement its policy linking
revocation to cancellation. 

EPA officials acknowledge that the agency's current process for
revoking tolerances takes too long and is inefficient.  According to
the officials, the establishment of a process linking revocation to
cancellation would help ensure that the tolerances for canceled
pesticides are revoked in a timely manner.  Furthermore, the
officials said that taking revocation and cancellation action
concurrently would generally result in the more efficient use of
EPA's resources than the current revocation process.  They emphasized
that although the revocation action could occur at the same time as
the cancellation action, EPA would need, when establishing an
effective date of revocation, to give growers enough time to use
exisitng stocks of the canceled pesticide. 


   ENFORCEMENT AGENCIES CANNOT
   TAKE ACTION AGAINST FOODS
   CONTAINING RESIDUES OF CANCELED
   PESTICIDES
---------------------------------------------------------- Chapter 3:3

As long as the tolerances for residues of canceled pesticides remain
in effect, FDA and USDA can do nothing to prevent foods containing
allowable amounts of these chemicals from entering the U.S.  food
supply.  Although a pesticide's residues usually decline or disappear
in domestic foods within a few years after the pesticide's
registrations have been canceled--except when the pesticide persists
in the environment--the pesticide's residues may continue to appear
in imported foods.  U.S.  and foreign manufacturers may continue to
sell the canceled pesticide for use on crops abroad, and the food
grown abroad may be sold in the United States as long as the
pesticide's residues do not exceed the tolerances. 

Determining the extent to which canceled pesticides that still have
tolerances appear in foods is difficult, not only because EPA has not
identified all of these pesticides but also because FDA monitors
residues selectively.  Nevertheless, in reviewing FDA's 1992
monitoring data, we found that FDA had detected demeton, a pesticide
whose registrations were canceled nearly 5 years ago, seven times on
four different food commodities that still had tolerances.  Because
the tolerances for demeton had not been revoked, FDA did not consider
the foods containing the pesticide to be adulterated. 

Similarly, assessing the health risks posed by canceled pesticides
that still have tolerances is difficult.  Information on the health
risks of many canceled pesticides is limited because EPA's data bases
are incomplete or because the registrant voluntarily canceled the
registrations before EPA finished assessing the pesticide's health
risks.  For example, after the registrations for food-uses of the
pesticide tetrachlorvinphos (a possible human carcinogen) were
canceled in 1987, EPA stated that it did not have sufficient
scientific data to evaluate the safety of the pesticide's
tolerances--which were established under less stringent scientific
standards in the early 1970s.  Despite this uncertainty about the
pesticide's dietary risks, EPA did not propose to revoke the
pesticide's tolerances until 1994--about 7 years after the associated
food-use registrations were canceled. 

For some canceled pesticides that still have tolerances, available
data indicate a potential to harm humans.  For example, the pesticide
monocrotophos is a potent cholinesterase inhibitor (linked to nervous
system problems) and is toxic to fetuses.  Another pesticide,
captafol, is classified by EPA as a probable human carcinogen.  Both
of these pesticides were voluntarily canceled 6 or more years ago,
but EPA has not yet revoked their tolerances, although it proposed to
do so in June 1993. 


   CONCLUSIONS
---------------------------------------------------------- Chapter 3:4

Because EPA has not consistently or expeditiously revoked the
tolerances for many canceled pesticides, residues of these pesticides
have been allowed to appear legally in the food supply, often for
many years after the pesticides' domestic uses were prohibited. 
Although EPA believes that it has revoked the tolerances for most of
the older, higher-risk canceled pesticides, it still needs to
identify and revoke the tolerances for a substantial number of other
canceled pesticides. 

By revoking the tolerances for canceled pesticides and, in the
future, conducting revocation and cancellation actions concurrently,
EPA would be acting consistently with its 1982 policy statement.  In
addition, it would be streamlining its process for revoking
tolerances and eliminating the potential for residues of the same
pesticides that it has deemed unacceptable for use on crops to appear
legally in food. 


   RECOMMENDATIONS
---------------------------------------------------------- Chapter 3:5

To expedite the revocation of tolerances for canceled pesticides and
make more efficient use of scarce resources, we recommend that the
Administrator, EPA,

establish procedures for concurrently conducting tolerance revocation
and cancellation actions and, when necessary, set an effective date
for revocation that gives growers enough time to use existing stocks
of the canceled pesticide and

identify the pesticides whose registrations for food uses have
already been canceled and revoke their tolerances. 


   AGENCY COMMENTS
---------------------------------------------------------- Chapter 3:6

EPA officials generally agreed with the information presented in this
chapter but suggested a number of technical and editorial changes
that we incorporated where appropriate.  The officials agreed that
actions such as we recommend are necessary to resolve the problems we
identified regarding tolerance revocations. 


SIX CANCELED PESTICIDES WITH
ACTION LEVELS FOR FISH
=========================================================== Appendix I

Pesticid
e         Regulatory action                   Effects
--------  ----------------------------------  ----------------------------------
Aldrin/   Registrations for all food uses of  Classified as probable human
Dieldrin  both pesticides canceled by EPA in  carcinogens. Neurotoxicity
\a        1975.                               observed in humans with chronic
                                              exposure. Adverse effects on the
                                              liver and the reproductive system
                                              indicated by animal studies.

Chlordan  Registrations canceled by EPA in    Classified as a probable human
e         1978 for all but a few food uses,   carcinogen. Adverse effects on the
          which were phased out by 1980.      immune and nervous systems,
                                              particularly during prenatal
                                              exposure, indicated by animal
                                              studies.

DDT       Registrations for all food uses     Classified as a probable human
          canceled by EPA in 1972.            carcinogen. Adverse effects on the
                                              reproductive system, liver, and
                                              immune system indicated by animal
                                              studies. Chromosomal damage also
                                              indicated.

Heptachl  Registrations for most food uses    Classified as a probable human
or        canceled by EPA in 1978; remaining  carcinogen. Increased chromosomal
          food uses phased out by 1983.       aberrations, blood disorders,
                                              decreased fertility, and decreased
                                              survival of newborns indicated by
                                              animal studies. Exposure in humans
                                              associated with stillbirths.

Mirex     Registrations for all uses          Classified as a probable human
          canceled by EPA in 1976.            carcinogen. Adverse developmental
                                              effects including heart defects,
                                              decreased fertility, undescended
                                              testes, and decreased brain and
                                              liver weights indicated by animal
                                              studies.
--------------------------------------------------------------------------------
\a Because the breakdown product of the pesticide aldrin is dieldrin,
residues of the two pesticides cannot be distinguished. 


OFFICE OF WATER'S STUDY OF
CHEMICAL RESIDUES IN FISH
========================================================== Appendix II

In September 1992, EPA's Office of Water issued a study of chemical
residues in fish that revealed extensive contamination by pesticides
in fish.  Because this study followed up on a national dioxin study,
the Office of Water sampled fish from waters near many sites that had
been targeted as likely sources of dioxins.  Samples were taken from
waters near 388 sites, 314 of which were point and nonpoint sources
of pollution, such as pulp and paper mills, and Superfund sites.  The
study found the following: 

DDT residues were found in fish at 99 percent of the sites.  For
example, catfish from the Alamo River, sampled near Calipatria,
California, had DDT levels yielding an estimated cancer risk of 8.9
in 100,000, and lake trout from Lake Michigan, sampled near Waukegan,
Illinois, had DDT levels yielding an estimated cancer risk of 6 in
100,000. 

Chlordane residues were found in fish at 64 percent of the sites. 
For example, lake trout from Lake Michigan, sampled near Waukegan,
Illinois, had chlordane levels yielding an estimated cancer risk of
9.3 in 100,000, and catfish from the Delaware River, sampled near
Torresdale, Pennsylvania, had chlordane levels yielding an estimated
cancer risk of 3.5 in 100,000. 

Dieldrin residues were found in fish at 60 percent of the sites.  For
example, lake trout from Lake Michigan, sampled near Waukegan,
Illinois, had dieldrin levels yielding an estimated cancer risk of
6.0 in 10,000, and carp from the Mississippi River, sampled near
Quincy, Illinois, had dieldrin levels yielding an estimated cancer
risk of 2.8 in 10,000. 

Mirex residues were found in fish at 38 percent of the sites.  For
example, chinook salmon from Lake Ontario, sampled near Olcott, New
York, had mirex levels yielding an estimated cancer risk of 3.8 in
100,000, and brown trout from Lake Ontario, sampled near Rochester,
New York, had mirex levels yielding an estimated cancer risk of 2.2
in 100,000. 

Heptachlor residues were found in fish at 16 percent of the sites. 
For example, lake trout from Lake Michigan, sampled near Waukegan,
Illinois, had heptachlor levels yielding an estimated cancer risk of
3.4 in 100,000, and carp from the Mississippi River, sampled near
Quincy, Illinois, had heptachlor levels yielding an estimated cancer
risk of 2.2 in 100,000. 

In June 1994, the Office of Water issued a guidance document for
states entitled Guidance for Assessing Chemical Contaminant Data for
Use in Fish Advisories:  Volume II - Risk Assessment and Fish
Consumption Limits.  This document was designed to assist states in
developing risk-based consumption limits for 23 high-priority
chemical contaminants in fish.  For the five pesticides included in
the Office of Pesticide Programs' study, the guidance document
indicated that consumers who eat slightly less than 1/2 pound of fish
per month and who desire no more than a negligible risk of cancer (1
in 1 million) should not eat fish when fish contain pesticides at or
near the action levels proposed by EPA to FDA in 1991. 

The Office of Water's guidance document presented alternative
consumption rates and risk levels for states to consider.  Table II.2
shows the number of meals per month that the document considers
acceptable when (1) fish contain pesticides at (or near) EPA's
proposed action levels, (2) the meal size varies from 4 ounces per
month to 16 ounces per month, and (3) the risk of cancer varies from
1 in 10,000 to 1 in 1 million.  For example, when fish contain
chlordane at the proposed action level (0.1 ppm), consumers can eat
14 4-ounce fish meals per month if they are willing to incur a cancer
risk of 1 in 10,000, but they cannot eat any fish meals if their
standard of safety is negligible risk (1 in 1 million). 



                          Table II.2
           
            Office of Water's Monthly Consumption
            Limits for Fish Containing Residues at
                or Near Proposed Action Levels


                                    1 in      1 in    1 in 1
                        Ounces    10,000   100,000   million
--------------------  --------  --------  --------  --------
Dieldrin
 0.10 ppm                    4       1.0      None      None
                             8       0.5      None      None
                            16      None      None      None
DDT\a
 0.50 ppm                    4      13.0       1.0      None
                             8       6.0       0.5      None
                            16       3.0      None      None
Chlordane
 0.10 ppm                    4      14.0       1.0      None
                             8       7.0       0.5      None
                            16       3.0      None      None
Heptachlor
 epoxide\b
 0.02 ppm                    4      10.0       1.0      None
                             8       5.0       0.5      None
                            16       2.0      None      None
Mirex\c
 0.05                        4  Unlimite       2.0      None
                                       d
                             8        13       1.0      None
                            16       6.0       0.5      None
------------------------------------------------------------
\a The Office of Water did not calculate consumption limits for fish
containing DDT residues at the proposed 0.5 action level but instead
calculated limits at the 0.4 and 0.6 residue levels.  In this table,
we used the limits calculated at the 0.4 residue level.  Consumption
limits at the 0.5 residue level (the proposed EPA action level) would
be expected to be equal to or lower than the limits at 0.4. 

\b Heptachlor epoxide is a metabolic breakdown product of heptachlor. 
Action levels set for heptachlor also apply to heptachlor epoxide. 

\c The Office of Water did not calculate consumption limits for fish
containing mirex residues at the proposed 0.05 action level but
instead calculated limits at the 0.04 and 0.06 residue levels.  In
this table, we used the limits calculated at the 0.04 residue level. 
Consumption limits at the 0.05 residue level (the proposed EPA action
level) would be expected to be equal to or lower than the limits at
0.04. 


CANCELED PESTICIDES WHOSE
TOLERANCES EPA HAD REVOKED AS OF
JULY 1994
========================================================= Appendix III

                                 (Delay in years)

                    Date of
                    cancellation        Date of final
Pesticide           action\a            revocation action     in revocation\b
------------------  ------------------  ------------------  --------------------
Acetic acid         January 1991        September 1993              2.6

Aldicarb            March 1992          June 1993                   1.2

Aldrin              May 1975            December 1986               11.6

Arsenic acid        May 1993            January 1994                0.7

BHC                 July 1978           July 1986                   8.0

Bufencarb           April 1986          June 1993                   7.1

Calcium arsenate    June 1988           April 1991                  2.8

Captan              February 1989       August 1992                 3.5

Carbon disulfide    December 1986       February 1989               2.1

Carbon              November 1986       February 1989               2.2
tetrachloride

Carbophenothion     October 1989        July 1994                   4.7

Chlordane           March 1978          December 1986               8.8

Chlordimeform       February 1989       October 1991                2.7

Chlordimeform       As of 1978          October 1989                10.8

Chlorobenzilate     February 1979       March 1986                  7.1

Chloroform          October 1983        February 1989               5.3

Copper arsenate     April 1977          May 1988                    11.1

Crufomate           October 1988        June 1993                   4.7

Daminozide          November 1989       March 1990                  0.3

Daminozide          August 1984         July 1987                   2.9

DBCP                September 1978      January 1986                7.3

DDT                 July 1972           December 1986               14.5

Diallate            January 1991        January 1994                3.0

Dichlorvos          October 1989        June 1991                   1.7

Dieldrin            May 1975            December 1986               11.6

Dinoseb             October 1986        September 1993              6.9

EDB                 October 1983        January 1985                1.3

EDB                 October 1983        December 1993               10.2

Endrin              As of               June 1993                   22.5
                    December 1970

EPN                 July 1987           June 1993                   5.9

Ethylene            December 1986       February 1989               2.1
dichloride

Fensulfothion       October 1988        November 1993               5.1

Glyphosine          March 1984          May 1988                    4.1

Heptachlor          March 1978          August 1989                 11.4

Isopropalin         As of               June 1991                   9.9
                    August 1981

Lead arsenate       June 1988           April 1991                  2.8

Mirex               December 1976       December 1986               10.0

Nitrapyrin          January 1986        June 1993                   7.4

Nitrofen [TOK]      September 1983      September 1985              2.0

Oryzalin            July 1984           June 1991                   6.9

Perfluidone         July 1984           June 1993                   8.9

Pirimicarb          As of               May 1988                    6.3
                    December 1981

Profluralin         April 1984          June 1993                   9.1

Ronnel              January 1986        March 1994                  8.2

Schradan [OMPA]     May 1976            January 1986                9.6

Silvex              January 1985        June 1993                   8.5

Sodium arsenite     June 1988           July 1993                   5.1

Sodium diacetate    January 1991        September 1993              2.6

Strobane            June 1976           January 1986                9.6

Tetrasul            January 1984        May 1988                    4.3

Thionazin           As of               May 1988                    16.4
                    December 1971

Toxaphene           November 1982       September 1993              10.8

Zineb               January 1991        December 1992               1.9

Average delay
between                                                             6.6
cancellation and
revocation
--------------------------------------------------------------------------------
\a Not all of the pesticide's food uses may have been canceled by the
action.  Some pesticides may have had more than one cancellation
action. 

\b This delay represents the time elapsed between cancellation and
final revocation.  When the day of the month could not be determined
for a cancellation or revocation action, the last day of the month
was used in calculating the delay.  When the month of the year could
not be determined for either of the actions, the last month of the
year was used in the calculation. 


CANCELED PESTICIDES WHOSE
TOLERANCES EPA HAD PROPOSED TO
REVOKE AS OF JULY 1994
========================================================== Appendix IV

For the following pesticides, EPA has issued a notice in the Federal
Register proposing the revocation of tolerances associated with their
canceled uses and requesting comments on the proposed actions.  But
the agency has not yet taken final action to revoke these tolerances. 

                       (Delay in years)

                                                    Delay in
                                                    revocati
                                                    on as of
                          Date of cancellation          July
Pesticide                 action\a                    1994\b
------------------------  ------------------------  --------
1,2,4,5-Tetrachloro-3-    October 1989                   4.8
 nitrobenzene [TCNB]
2-                        October 1989                   4.8
 Chloroallyldiethyldithi
 o-carbamate
 [Sulfallate]
Alachlor                  March 1988                     6.3
Barban                    October 1989                   4.8
Bifenox                   January 1991                   3.5
Captafol                  April 1987                     7.2
Chlorfenvinphos           January 1991                   3.5
Chlorobenzilate           December 1988                  5.6
Crotoxyphos               January 1991                   3.5
Cyclohexamide             October 1989                   4.8
Demeton                   October 1989                   4.8
O,O-Dimethyl O-p-         October 1989                   4.8
 (Dimethylsulfamoyl)phen
 yl. [Cythioate]
Dinitramine               January 1991                   3.5
Dipropetryn               December 1989                  4.6
Flucythrinate             January 1991                   3.5
Hexachlorophene           January 1991                   3.5
Mancozeb                  March 1992                     2.4
Maneb                     March 1992                     2.4
Metiram                   March 1992                     2.4
Monocrotophos             July 1988                      6.0
Norea                     October 1989                   4.8
Perthane                  June 1980                     14.1
Phenothiazine             January 1991                   3.5
Pirimphos-ethyl           January 1991                   3.5
Sodium trichloroacetate   October 1989                   4.8
Sulfur dioxide            December 1991                  2.5
Terbutryn                 As of                          4.8
                           October 1989
Terrazole                 Mid-1980s                      6.5
Tetrachlorvinphos         August 1987                    6.9
Tetradifon                January 1990                   4.5
Tributylphosphorotrithio  January 1991                   3.5
 ite [Merphos]
Average delay between
 cancellation and                                        4.7
 revocation
------------------------------------------------------------
\a Not all of the pesticide's food uses may have been canceled. 

\b Because EPA has not yet finally revoked the tolerances for these
pesticides, this delay represents the time elapsed between the date
of cancellation and the date we obtained these data--July 1994.  The
delay for these pesticides can be expected to increase until EPA
finally revokes their tolerances.  When the day of the month could
not be determined for a cancellation action, the last day of the
month was used in calculating the delay.  When the month of the year
could not be determined for a cancellation action, the last month of
the year was used in the calculation. 


MAJOR CONTRIBUTORS TO THIS REPORT
=========================================================== Appendix V

RESOURCES, COMMUNITY, AND ECONOMIC
DEVELOPMENT DIVISION, WASHINGTON,
D.C. 

J.  Kevin Donohue, Assistant Director
Robert J.  Tice, Evaluator-in-Charge
Mark Trapani, Staff Evaluator
Richard A.  Frankel, Senior Scientist/Evaluator
Elizabeth R.  Eisenstadt, Communications Analyst

