Natl. Oilseed Processors Ass'n v. Browner

Full title: NATIONAL OILSEED PROCESSORS ASS'N, Plaintiff, v. Carol BROWNER…

Court: United States District Court, D. Columbia

Date published: Apr 30, 1996

924 F. Supp. 1193 (D.D.C. 1996)

Opinion

Nos. CV 95-763, CV 95-980, CV 95-1673 and CV 95-1910.

April 30, 1996.

David Menotti, John Seymour, Anita Cicero, Shaw, Pittman, Potts Trowbridge, Washington, D.C., for the National Oilseed Processors Associations.

William K. Rawson, Claudia O'Brien, Latham Watkins, Washington, D.C., for Troy Corporation and Chemical Manufacturers Association.

David Zoll, Amit Sachdev, Chemical Manufacturers Association, Washington, D.C., Cynthia Lewis, Karl Bourdeau, Alec I. Ugol, Beveridge Diamond, Washington, D.C., for NMP Producers Group.

Eugene M. FitzMaurice, ARCO Chemical Company, Newtown Square, PA.

Scott Jordan, Mary Edgar, Environmental Defense Section, U.S. Department of Justice, Washington, DC, Amber Aranda, Carl Eichenwald, Julie Simpson, U.S. Environmental Protection Agency, Washington, DC, for Carol Browner and E.P.A.

MEMORANDUM OPINION

KESSLER, District Judge.

I. Introduction

Each of four Plaintiffs, Troy Corporation ("Troy"), NMP Producers Group ("NMP"), National Oilseed Processors Association ("NOPA"), and Chemical Manufacturers Association ("CMA"), brings a separate action against the Environmental Protection Agency ("EPA" or "Agency") pursuant to the Administrative Procedure Act ("APA"), 5 U.S.C. § 706(2)(A), on the grounds that EPA's addition of each Plaintiff's chemical or chemicals to the Toxic Release Inventory ("TRI") list was arbitrary and capricious, an abuse of discretion, and contrary to law.¹ Each Plaintiff seeks a final order vacating the final decision of the EPA to list the chemical or chemicals pursuant to § 313 of the Emergency Planning and Community Right-To-Know Act of 1986 ("EPCRA"), 42 U.S.C. § 11023.

1 Although these cases were not formally consolidated into one action, there were a number of joint status conferences and all counsel worked cooperatively to coordinate and simplify issues to the greatest extent possible. The Court hereby disposes in one opinion of issues which are common to all cases. To the extent the issues in any given case differ, as with chemical-specific arguments, they are dealt with separately. Separate Orders will issue for each Plaintiff.

This matter is before the Court upon Plaintiffs' Motions for Summary Judgment and Defendants' Cross-Motions for Summary Judgment pursuant to Federal Rule of Civil Procedure 56. The Court has considered the Motions, the Oppositions, the Replies, the oral arguments in open court on March 27, 1996, the applicable statutory and case law, and the voluminous record submitted in each of the four cases. For the reasons discussed below, the Court concludes that Plaintiffs' Motions for Summary Judgment must be denied, and that Defendants' Cross-Motions for Summary Judgment must be granted.

II. Statutory Framework

State and local governments, as well as the public at large, are entitled to access information concerning potential chemical hazards in their communities. That is the central premise underlying the Emergency Planning and Community Right-To-Know Act ("EPCRA"), codified at 42 U.S.C. § 11001-11050 and signed into law on October 17, 1986, as Title III of the Superfund Amendments and Reauthorization Act of 1986, Pub.L. No. 99-499, 100 Stat. 1613, 1629. The primary purpose of the law is to encourage state and local planning for spills or releases of toxic or hazardous chemicals. 42 U.S.C. § 11001.

Section 313² of EPCRA mandates that facilities manufacturing, processing or using certain toxic chemicals report annually on the presence of those chemicals at the facility, the uses of the chemicals, an estimate of the maximum amounts of the chemicals present at the facility at any time, methods of disposal and treatment of waste, and the extent to which those chemicals are being released into the environment. 42 U.S.C. § 11023(g). The list of toxic chemicals subject to reporting under § 313 is known as the Toxic Release Inventory ("TRI") list. Notably, EPCRA does not restrict the manufacture, processing, use or disposal of any chemical; it is simply a reporting statute which requires facility owners or operators to provide EPA and state governments with information.³ 42 U.S.C. § 11023(a). EPA and state governments, in turn, make this information available to local governments and citizens in the community, who may then develop appropriate emergency response plans as required under Section 303 of EPCRA. See 42 U.S.C. § 11023(h) and (j); 59 Fed.Reg. 1788 (col. 3) (Jan. 12, 1994).

2 Section 313 of EPCRA applies to facilities with 10 or more employees that are in Standard Industrial Codes 20 through 39 ( i.e., all manufacturing industries). 42 U.S.C. § 11023(b)(1)(A). Reports are generally required for any listed chemical or chemical category if a facility uses at least 10,000 pounds, or manufactures or processes at least 25,000 pounds, of the chemical or chemical category in a year. Reporting under Section 313 is not currently required from non-manufacturing industries, such as agriculture, mining, construction, wholesale and retail trade, and service industries.

3 Even as a reporting statute, Section 313 places certain obligations and burdens on those having to report. While there is great disagreement about the costs and time involved in such reporting, the Court recognizes that, whatever the specifics may be, corporate resources must be devoted to the task.

EPCRA also requires EPA to maintain a publicly accessible computer database containing a national toxic chemical inventory based on the reports submitted under Section 313. 42 U.S.C. § 11023(j). This information has enabled the Federal government, State governments, industry, environmental groups, and the general public to participate in an informed dialogue about the environmental impact of toxic chemicals in order to assess the need to reduce and, where possible, eliminate chemical releases. See, e.g., 42 U.S.C. § 11005(a)(1) (2).

When Congress enacted EPCRA in 1986, it directly placed 309 individual chemicals and 20 chemical categories on the TRI list. 42 U.S.C. § 11023(c). These chemicals came from two existing lists of toxic chemicals: the Maryland Chemical Inventory Report List of Toxic or Hazardous Substances, and the New Jersey Environmental Hazardous Substance List. 59 Fed.Reg. at 1788 (col. 3).

Congress provided that additions may be made to the TRI list in the following manner: First, under § 313(e), 42 U.S.C. § 11023(e), either a private party or a State Governor may petition EPA to add chemicals to or delete chemicals from the TRI list. Within 180 days of receiving such a petition, EPA must either (1) initiate a rulemaking to add or delete the chemical, or (2) publish an explanation of why the petition is denied. 42 U.S.C. § 11023(e)(1).⁴

4 Under Section 313(e)(2), 42 U.S.C. § 11023(e)(2), if a State Governor files a petition to add a chemical to the TRI list, that chemical will be added unless EPA takes affirmative steps granting or denying the petition within 180 days.

Second, on its own initiative, EPA "may by rule add or delete a chemical from the [TRI list] at any time." 42 U.S.C. § 11023(d)(1). Congress left to EPA's discretion the timing and manner of selection of candidate chemicals to be considered for listing under the statutory criteria. The criteria for listing are the same regardless of whether listing is triggered by a petition or an EPA rulemaking. 42 U.S.C. § 11023(d)(2) and (e). The basic requirement for addition of chemicals to the TRI list is a finding by the Administrator that, in her judgment, "there is sufficient evidence to establish any one of" three criteria for listing. 42 U.S.C. § 11023(d)(2).

The first criterion provided in the statute is that EPA may list chemicals that are "known to cause or can reasonably be anticipated to cause significant adverse acute human health effects at concentration levels that are reasonably likely to exist beyond facility site boundaries as a result of continuous, or frequently recurring, releases." 42 U.S.C. § 11023(d)(2)(A).

The second statutory criterion is that EPA may list chemicals that are known to cause or can reasonably be anticipated to cause in humans cancer, teratogenic effects (i.e., developmental malformations), or serious or irreversible reproductive dysfunctions, neurological disorders, heritable genetic mutations, or other chronic health effects. See 42 U.S.C. § 11023(d)(2)(B).

The third statutory criterion is that EPA may list chemicals that are known to cause or can reasonably be anticipated to cause, because of their toxicity alone, or their toxicity and persistence in the environment, or their toxicity and tendency to bioaccumulate in the environment, a significant adverse effect on the environment. See 42 U.S.C. § 11023(d)(2)(C).

In enacting Section 313, Congress provided that EPA may list a chemical even when its effects cannot be proven to a scientific certainty. The statute mandates that EPA base its listing decisions under Section 313 "on accepted scientific principles or laboratory tests, or appropriately designed and conducted epidemiological or other population studies, available to the Administrator." 42 U.S.C. § 11023(d).

III. EPA's Screening Process

After seven years of administering EPCRA, EPA concluded in 1993 that the TRI list covered only a relatively small portion of the chemicals potentially of concern to EPA and the public. See 59 Fed.Reg. at 1788-89. In order to develop candidates for addition to the TRI list, EPA identified ten lists of chemicals and two lists of carcinogens. 59 Fed.Reg. 61,432. Faced with 12 lists containing over 1500 chemicals⁵, as well as the Agency's limited resources, EPA developed and applied screening procedures⁶ to eliminate some chemicals from consideration before beginning the rulemaking.⁷

5 See EPA MSJ (Troy) at 10.

6 Whether these preliminary screening procedures were technically part of the rulemaking process is irrelevant. The goal of the process was to winnow the list of over 1500 chemicals down to a more manageable number. Companies whose chemicals were eliminated, of course, have no complaint about the process. Those chemicals that were not eliminated by the screening process were then subjected to a far more intensive review in accordance with the rulemaking. EPA's final decision to place chemicals on the TRI list was not based on the screening process, but rather on the statutory criteria; the chemicals were considered regardless of their designation as "high priority" or "medium priority" in the screening process. Therefore, any claims, based on the argument that the screening is part of the rulemaking, must fail.

7 Twelve chemicals were considered for listing during the rulemaking without having been subjected to the screening process.

First, EPA excluded a number of the chemicals for practical reasons, such as the fact that they were already on the TRI list or that they had previously been considered and rejected for listing in two previous rulemakings triggered by Section 313(e) petitions. 59 Fed.Reg. 1789. Once this was done, EPA had narrowed the list to 1031 potential candidates for the rulemaking process.

Second, EPA subjected most of the remaining 1031 chemicals to a "production volume screen" and a "toxicity screen." 59 Fed.Reg. 1789-90. Under the production volume screen, EPA determined whether each chemical was manufactured, processed, or used in volumes above the statutory reporting thresholds established in Section 313(f). 42 U.S.C. § 11023(f). EPA used the toxicity screen to determine whether the chemicals manifested acute human health effects, cancer, other chronic human health effects, or ecological effects. This screen was based on readily available toxicity data and involved a "quick look" to identify chemicals that did not warrant further consideration. 59 Fed.Reg. 1789. EPA used the toxicity data to place each chemical in one of three categories: high priority, medium priority, or low priority. 59 Fed.Reg. 61,432-33.

As a result of the two screens, there was a significant narrowing of the list of chemicals to be considered by EPA in the rulemaking process. As a result of their "low priority" ranking, the narrowing process culminated in the exclusion of 648 chemicals from the rule-making. 59 Fed.Reg. 1789-90. The remaining 383 chemicals became candidates for listing and entered the rulemaking phase.

IV. The Present Rulemaking

EPA reviewed, in a full "hazard evaluation," each of the remaining 383 chemicals to determine whether the scientific evidence supported its listing under one or more of the three criteria set forth in Section 313(d)(2). 42 U.S.C. § 11023(d)(2). The hazard evaluation involved a review of available studies and other information to determine (1) whether the chemical exhibits one or more of the toxic effects that justify addition to the TRI list, and (2) the dose required to create that effect. 59 Fed.Reg. 1790. In the hazard evaluation, EPA also assessed the scientific studies' reliability by analyzing the assumptions and methodologies used in conducting the studies. Id.

When no single study was sufficient to determine whether a chemical had toxic effects, the hazard evaluation included a "weight-of-evidence" evaluation. That evaluation involved an assessment of the available studies and information, both positive and negative, about the chemical. In undertaking this evaluation, EPA tried to determine whether the collective results of the individual studies would form a reasonable basis for reaching a conclusion on toxicity. 59 Fed. Reg. 1790.

EPA based its hazard evaluation of the chemicals, in most instances, on information in EPA's Integrated Risk Information System ("IRIS"). 59 Fed.Reg. 61,444-45. IRIS is an EPA-created database,⁸ which contains EPA's consensus positions⁹ on the potential adverse human health effects of approximately 500 chemicals. Announcement of Availability of IRIS Background Paper ("IRIS Background Paper"), 58 Fed.Reg. 11,490. Developed in 1986, IRIS offers chemical-specific hazard and risk information, which is the result of periodically updated, comprehensive hazard assessments conducted by panels of senior EPA scientists. See Report of the IRIS Quality Action Team on External Peer Review and Public Involvement at 5-6; IRIS is generally accepted as a reliable source of information on the potential hazardous effects of those chemicals that are included in IRIS. See 58 Fed.Reg. 11,490.¹⁰

8 For each chemical contained in IRIS, the database contains summary information on the studies evaluated, any uncertainties or assumptions made in the studies, a statement of the level of confidence that EPA has in the study, the names of EPA scientists to contact for more information, and complete bibliographic citations. See Report of the IRIS Quality Action Team on External Peer Review and Public Involvement at 5 (July 1994).

9 EPA includes information in IRIS only when no member of EPA's staff is aware of information that would refute the conclusions presented or of analyses that suggest that a different view is more credible. See IRIS Background Paper at 7-9.

10 EPA also used other databases during its hazard evaluations, including the toxicity database submitted by registrants of pesticide products containing IPBC under the Federal Insecticide, Fungicide, and Rodenticide Act ("FIFRA"), 7 U.S.C. § 136 et seq., the Office of Pesticide Programs Tox-One-Liner database, and the Reference Dose Tracking Report.

In January, 1994, after completing the hazard evaluation on the 383 candidate chemicals, EPA published a proposed rule adding chemicals to the TRI list. Of the 383 chemicals considered in the rulemaking, 313 were found to meet the statutory criteria and proposed for listing in EPA's proposed rule. 59 Fed.Reg. 1793.

Following receipt of comments on the proposed rule, EPA considered the arguments and information submitted by the Plaintiffs and others and responded to those comments. As a result of the comments, EPA decided to defer action on 41 chemicals to a future rulemaking. 59 Fed.Reg. 61,439. EPA also decided that there was insufficient information to justify the listing of three chemicals proposed for addition to the TRI list. 59 Fed.Reg. 61,466. Finally, EPA added a category of 20 chemicals, which had included three that had been proposed for listing individually. Thus, of the 313 chemicals that were proposed for listing, 286 chemicals were added to the TRI list in November, 1994 in the final rule.

EPA's administrative record is voluminous. For example, EPA's revised Draft Hazard Assessment Guidelines for Listing Chemicals on the Toxic Release Inventory of May 26, 1992, comprised thirty-two single-spaced pages; EPA's Guidelines for Developmental Toxicity Risk Assessment comprised twenty-eight pages in the Federal Register, see 56 Fed.Reg. 62,798-63,826; the proposed rule itself comprised seventy pages in the Federal Register, see 59 Fed.Reg. 1788-1859; EPA submitted a comprehensive Response to Comments Document; and the final rule comprised 53 pages in the Federal Register, see 59 Fed.Reg. 61,432-85. That sampling only begins to capture the weighty paper trail before the Court.

Further, EPA considered and responded to countless comments throughout the rulemaking process. See, e.g., 59 Fed.Reg. at 61,439-71. EPA responded in detail to comments which included concern about the validity of the screening procedure, the appropriateness of the Draft Hazard Assessment Guidelines, the need for risk assessments and exposure consideration before adding a chemical to the TRI list, the statutory authority to list chemical categories, numerous policy issues, and general technical matters. The record is extensive, ponderous, and sweeping.

Finally, while the actual rulemaking began with the issuance of the proposed rule on January 12, 1994, and ended with the issuance of the final rule on November 30, 1994, consideration of these issues actually began back in May, 1992, when the Revised Draft Hazard Assessment Guidelines for Listing Chemicals on the Toxic Release Inventory was issued. Thus, the issues raised by Plaintiffs in these lawsuits have been under scrutiny in the administrative process for more than two years.

V. Standard of Review

All parties recognize that the Court is bound by a highly deferential standard of review for agency action. Under the Administrative Procedure Act ("APA"), an agency's action may be set aside only if it is "arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law." 5 U.S.C. § 706(2)(A). In making this finding, the Court "must consider whether the decision was based on a consideration of the relevant factors and whether there has been a clear error of judgment." Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 416, 91 S.Ct. 814, 823, 28 L.Ed.2d 136 (1971). The Court may not substitute its judgment for that of the agency. Id. The Court's role is to ensure that the agency's decision was based on relevant factors and not a "clear error of judgment." Id. If the "agency's reasons and policy choices . . . conform to `certain minimal standards of rationality' . . . the rule is reasonable and must be upheld." Small Refiner Lead Phase-Down Task Force v. EPA, 705 F.2d 506, 521 (D.C. Cir. 1983) (citation omitted). This standard presumes the validity of agency action. Ethyl Corp. v. EPA, 541 F.2d 1, 34 (D.C. Cir. 1976) (en banc), cert. denied, 426 U.S. 941, 96 S.Ct. 2662, 49 L.Ed.2d 394 (1976).

An agency's interpretation of a statute should be upheld as long as it is a permissible interpretation. Chevron USA Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 843-45, 104 S.Ct. 2778, 2781-83, 81 L.Ed.2d 694 (1984). Courts also give a high degree of deference to agency actions based on an evaluation of complex scientific data within the agency's technical expertise. See Baltimore Gas Elec. Co. v. NRDC, 462 U.S. 87, 103, 103 S.Ct. 2246, 2255, 76 L.Ed.2d 437 (1983); NRDC v. EPA, 824 F.2d 1211, 1216 (D.C. Cir. 1987) (citing NRDC v. EPA, 812 F.2d 721, 725 (D.C. Cir. 1987)) ("[I]t is not for the judicial branch to undertake comparative evaluations of conflicting scientific evidence.").

It is hard to imagine a case that fits more squarely within these principles than the four cases before the Court. The EPA rulemaking involves consideration of complex scientific data and sophisticated analysis fit primarily for those tutored in the field. Few courts can and no court should engage in an independent comparative evaluation of conflicting scientific evidence. Where the agency decision turns on issues requiring the exercise of technical or scientific judgment, it is essential for judges to "look at the decision not as the chemist, biologist, or statistician that we are qualified neither by training nor experience to be, but as a reviewing court exercising our narrowly defined duty of holding agencies to certain minimal standards of rationality." Ethyl Corp., 541 F.2d at 36-37.

VI. Issues Common to the Rulemaking

A. Exposure

All Plaintiffs contend that the rulemaking is fatally flawed because EPA did not consider exposure information in its listing decisions.¹¹ Plaintiffs' general challenge can be broken down into two major arguments. First, they argue that EPA must, as a matter of law, consider exposure because the language "chronic health effects" in Section 313(d)(2)(B) implies consideration of "risk," which in turn requires weighing exposure in the analysis. Second, Plaintiffs argue that even if EPA does have discretion under the statute whether or not to consider exposure, it must exercise that discretion in a non-arbitrary manner, and fully justify on the record how it exercised its discretion. Plaintiffs contend that in the present rulemaking, EPA has neither articulated a meaningful standard to explain when it will consider exposure, nor provided a record that reveals whether EPA has applied its alleged standard reasonably.

11 The term "exposure" refers to the likelihood that a person will be exposed to a chemical at a dose high enough to cause a toxic effect. See EPA MSJ (n-hexane) at 15 n. 12.

1. Chevron Analysis: Does EPA Have Discretion to Include or Exclude Consideration of Exposure in Listing Decisions?

The first issue — whether EPA must, as a matter of law, consider exposure in listing decisions — is a question of statutory interpretation, governed by the Supreme Court's landmark decision in Chevron, 467 U.S. 837, 104 S.Ct. 2778. Under the rule stated in Chevron, a court must follow a two-step test. First, a court must look to the plain meaning of the statute to determine whether Congress has spoken to the precise question at issue. Chevron, 467 U.S. at 842, 104 S.Ct. at 2781. If it has, then the court must give effect to the clearly expressed intent of Congress. Id. at 842-43, 104 S.Ct. at 2781. If the statute is silent or ambiguous on the issue, however, the agency's interpretation should be upheld so long as it is a "permissible construction of the statute." Id. at 843, 104 S.Ct. at 2781. A court "need not conclude that the agency construction was the only one it permissibly could have adopted to uphold the construction, or even the reading the court would have reached if the question originally had arisen in a judicial proceeding." Id. at 843 n. 11, 104 S.Ct. at 2781 n. 11.

Under Section 313 of EPCRA, EPA may list a chemical in one of three instances. Under Section (d)(2)(A), the Agency may list a chemical if it "is known to cause or can reasonably be anticipated to cause acute human health effects . . . at concentration levels that are reasonably likely to exist beyond facility site boundaries as a result of continuous, or frequently occurring, releases." 42 U.S.C. § 11023(d)(2)(A). Thus, under the "acute effects" basis for listing, EPA must consider both (1) the effect that the chemical causes and (2) the likelihood of exposure to the chemical ("concentration levels" beyond facility site boundaries).

By contrast, the two other listing criteria (Section (d)(2)(B) for "chronic health effects" and Section (d)(2)(C) for "environmental effects") only require EPA to conclude that the chemical is "known to cause or can reasonably be anticipated to cause" chronic human or environmental effects, that is, they require EPA to consider only the effects the chemical causes. These provisions do not contain the same specific language pertaining to the "concentration levels that are reasonably likely to exist beyond facility site boundaries" or any other language suggesting that EPA must consider the circumstances or likelihood of exposure. 42 U.S.C. § 11023(d)(2)(B) and (C).

While these differences strongly suggest that Congress intended that exposure must be considered under Section (d)(2)(A) but not under Sections (d)(2)(B) or (C), it cannot be said that Congress spoke with such clarity that the issue can be fully resolved by looking only at the plain meaning of the statute. The Court therefore proceeds to the second step of the Chevron analysis.

Under Chevron's second step, the Court must uphold an agency's interpretation so long as it is a "permissible construction of the statute." Chevron, at 843, 104 S.Ct. at 2781. Here, EPA argues that the logical and reasonable interpretation of the differences in statutory language just discussed is that Congress required consideration of exposure when listing a chemical under the "acute effects" prong in subsection (A), but did not mandate such consideration of exposure when listing a chemical under the "chronic effects" or "environmental effects" standards in subsections (B) or (C).

This is both a reasonable and permissible construction of the statute, and under Chevron, the Court must accept the Agency's reading. The Court concludes, therefore, that EPCRA does not mandate that EPA consider exposure under Section 11023(d)(2)(B), and that EPCRA allows the Agency discretion to include or exclude exposure information in making listing decisions under that section.

2. Did EPA Exercise Its Discretion Reasonably?

Even though Congress gave EPA the discretion to not consider exposure in its listing decisions, it must not exercise that discretion in an arbitrary or unreasonable fashion. Chemical Manufacturers Ass'n v. EPA, 899 F.2d 344, 359 (5th Cir. 1990).

Plaintiffs argue that EPA acted arbitrarily and capriciously in this rulemaking because it has departed from its previous policy of considering exposure in listings under Section 313(d)(2)(B). In support, Plaintiffs cite to instances where EPA has considered exposure in past listing and delisting decisions.¹² Plaintiffs also claim that EPA has not adequately explained when it will consider exposure under Section 313(d)(2)(B).

12 58 Fed.Reg. 51,786 (Oct. 5, 1993) (delisting of di-n-octylphthalate); 52 Fed.Reg. 41,624 (October 29, 1987) (denial of a petition to delist orthophenylphenol); 55 Fed.Reg. 25,876 (June 25, 1990) (petition to delist phosphoric acid denied on other grounds); 52 Fed.Reg. 20,142-45 (May 29, 1987) (denial of a petition to add inorganic fluorides, where Agency stated that "potential exposure must be a consideration").

The Agency argues generally that, in the exercise of its discretion, it has elected to consider exposure only in limited circumstances. Specifically, when EPA's hazard assessment shows that a chemical exhibits only low or moderately low toxicity, EPA will consider the potential for exposure in making a listing decision. Conversely, where EPA's hazard assessment reveals that a chemical's toxicity is high or moderately high, EPA does not consider exposure, and will list the chemical based solely on its toxic effect. 59 Fed.Reg. 61,441.¹³

13 EPA follows this policy because it furthers the purpose of the statute, namely, to provide information to state and local governments and citizens about the potential toxic effects of chemicals being discharged into their communities. To fulfill this purpose, EPA chose to focus its listing decisions on the effects of a chemical because it can make a single nationwide determination that a certain dose of a particular chemical will have a toxic effect regardless of where that dose was received. By contrast, exposure is a highly site-specific determination, which is not easily reduced to a single nationwide conclusion. Thus, EPA says, it is more effective for the Agency to list a chemical based on its effects ( i.e., deciding to list Chemical X because it will have a toxic effect if it is received at a certain dose), because local governments will then have access to the relevant information — including the facility's use of the chemical, methods of disposal, treatment of waste, and releases of the chemical — and can then consider exposure on a site-by-site basis. Because no one can predict all of the individual local exposure scenarios that might arise for a chemical, EPA concludes that the most appropriate solution is to list chemicals based on their toxic effects, with consideration of exposure information in limited circumstances only.

This policy, the Agency argues, has been consistently applied since EPCRA's passage. The instances cited by Plaintiff where EPA has considered exposure in listing decisions do not demonstrate otherwise. Rather, argues EPA, those listings and delistings are simply examples where EPA chose to consider exposure, because the chemicals at issue were of low toxicity.

Moreover, EPA asserts that it explained adequately on the record that it chose to not consider exposure in this rulemaking because all of the chemicals proposed for listing under Section 313(d)(2)(B) were of "high to moderately-high" toxicity and therefore consideration of exposure was not appropriate.

After consideration of the extensive arguments on both sides of this issue, the Court concludes that the Agency did not act arbitrarily and capriciously in declining to consider exposure in the listing decisions for this rulemaking. For the past eight years, in "31 of 32 listing/delisting determinations," 59 Fed.Reg. 61,442, the Agency has considered exposure only in limited circumstances, and it has followed that policy in the present rulemaking.¹⁴ While a more clearly and fully articulated policy would be preferable, the Court cannot conclude that EPA was unreasonable in exercising its discretion by continuing to exclude consideration of exposure when chemicals are of high to moderately-high toxicity.

14 EPA concedes there has been one exception to this policy. In 1987, EPA denied a petition to add inorganic fluorides to the TRI, stating that "potential exposure must be a consideration in making decisions to add chemicals to the [TRI] list." 52 Fed.Reg. 20142. Whatever the reason for that statement in 1987, EPA certainly has not followed that policy in the last eight years.

Plaintiffs' examples of listing and delisting decisions where EPA considered exposure do not prove that EPA unreasonably exercised its discretion by departing from past policy. Rather, those decisions reflect EPA's policy in practice, namely, that it will consider exposure when the subject chemical is of low toxicity. The fact that it is EPA's policy to consider exposure in some instances and not in others does not make the policy arbitrary, so long as there is a reasonable rationale for the policy and therefore for the Agency's exercise of its discretion.

The Agency also adequately explained and documented its decision making process in the administrative record. EPA made a finding that all of the chemicals that were being listed under Section 313(d)(2)(B) in this rulemaking were of high or moderately high toxicity, and that consideration of exposure was therefore not appropriate for these chemicals. See 59 Fed.Reg. 61,442. The fact that in its final rule, EPA issued a general finding that applied to the entire group of chemicals listed under Section 313 rather than repeating it with respect to each and every one of them has little significance. What is significant is that EPA stated what its policy for consideration of exposure would be, and then described its application to the chemicals considered in this rulemaking.

Plaintiffs here have failed to show that EPA was unreasonable or arbitrary in exercising its discretion. Because EPA's decision to not consider exposure in this rulemaking was consistent with its policy of using exposure data only in particular circumstances, i.e., where chemicals are of low toxicity, the Court concludes that the Agency was not arbitrary and capricious.¹⁵

15 Plaintiffs' also claim that the different language in the preambles to the proposed and final rules represent a change in policy without reasonable justification or explanation. In the proposed rule, EPA stated that "it is sufficient to consider only the toxicity of the subject chemical to make the section 313(d)(2)(B) determination." 59 Fed.Reg. 1792. In the final rule, EPA said that it would consider exposure only where a subject chemical was "of low toxicity and unrealistic exposures would be necessary for it to pose a risk to communities." 59 Fed.Reg. 61,442.
The different language in the proposed and final rules does not represent a change in policy. In the proposed rule, the Agency stated the boundaries of its statutory authority, i.e., that the statute did not require EPA to consider exposure. In contrast, in the final rule, EPA explained the manner in which it was exercising its discretion in this particular rulemaking.

B. Post-Hoc Rationalization

Plaintiffs' second major complaint about the final rulemaking is that EPA has engaged in post-hoc rationalization to defend its listing decisions. In essence, Plaintiff argues that EPA's lawyers have provided the explanations that EPA's staff and scientists should have provided.

Certainly, the voluminous briefs submitted by EPA, in response to Plaintiffs' equally voluminous challenges, are far longer and more detailed than some of the explanations for listing decisions on individual chemicals contained in the administrative record. But when an agency provides further explanation of its decision in the course of litigation, it is not necessarily engaging in post hoc rationalization. The rule is that an agency must defend its actions on the basis on which they were originally taken, not on some new basis that is developed in litigation to justify the decision. Motor Vehicle Mfr. Ass'n v. State Farm Mutual Auto Ins. Co., 463 U.S. 29, 50, 103 S.Ct. 2856, 2870, 77 L.Ed.2d 443 (1983).

The rule against post-hoc rationalizations does not prevent a court from considering a more detailed explanation of an agency's action in response to a legal challenge. As long as the agency does not present a new basis for its action, it may supply a clearer or more detailed explanation. Methodist Hosp. of Sacramento v. Shalala, 38 F.3d 1225, 1233 n. 11 (D.C. Cir. 1994) (court rejected plaintiffs' argument that position taken by agency in litigation was a post hoc rationalization, even though agency "could have placed a finer point" on the issue in its explanation in the record); International Ladies Garment Workers' Union v. Donovan, 722 F.2d 795, 814 (D.C. Cir. 1983) (a court will "uphold a decision of less than ideal clarity if the agency's path may be reasonably discerned.") (quoting Bowman Trans. Inc. v. Arkansas-Best Freight Sys., Inc., 419 U.S. 281, 286, 95 S.Ct. 438, 442, 42 L.Ed.2d 447 (1974)), cert. denied, 469 U.S. 820, 105 S.Ct. 93, 83 L.Ed.2d 39 (1984).

The key distinction is the difference between a post hoc rationalization, which is a new rationale for an agency action, and a post hoc explanation, which is an agency's discussion of the previously-articulated rationale for the challenged action. Post hoc rationalizations are precluded; post hoc explanations are not. See e.g. Sierra Club v. Marsh 976 F.2d 763, 774-75 (1st Cir. 1992) (holding that "the district court properly accepted the post-hoc explanations of the decisionmakers' action" because the agencies' explanations were supported by evidence in the administrative record) (relying on Citizens to Preserve Overton Park v. Volpe, 401 U.S. 402, 420, 91 S.Ct. 814, 825, 28 L.Ed.2d 136 (1971)).

In this case, EPA has not presented new or different or better rationales for its listing decisions. Instead, the Agency has presented in its briefs a rather fulsome description of the basis for each listing and, significantly, in doing so has directed the Court to the relevant supporting portions of the administrative record. The fact that EPA's discussion of its listing decisions is documented with citations to the record is further evidence that the Agency's lawyers are providing clearer and more understandable justifications of highly technical scientific justifications, rather than new and different post hoc rationalizations.¹⁶ See International Bhd. of Teamsters v. United States, 735 F.2d 1525, 1531 n. 3 (D.C. 1984) (whether a discussion in a brief contains citations to the record is a factor in determining whether an argument made in litigation is a post hoc rationalization).

16 As a practical matter, it is hard to believe that EPA's lawyers in the Department of Justice (unless they themselves are PhDs in the relevant scientific fields) could have developed the sophisticated scientific arguments contained in EPA's briefs.

Moreover, as the Court discussed above, the Agency exhaustively documented this rulemaking, filling over 100 pages of the Federal Register with the proposed and final rules, and replying to industry comments in a comprehensive Response to Comments document. The Agency need not explain its thinking in minute detail; it simply must provide a decision-making path that is "reasonably discerned." Motor Vehicle Mfr. Ass'n, 463 U.S. at 44, 103 S.Ct. at 2867 (citations omitted). There is nothing improper with Agency counsel elaborating on EPA's simple but adequate reasoning contained in the administrative record.

For these reasons, the Court concludes that the Agency has not engaged in post hoc rationalization.

VII. EPA's Weight-of-Evidence Review of Specific Chemicals

In addition to the general arguments that Plaintiffs make attacking the validity of the rulemaking, they also claim that there is insufficient evidence in the record to support a finding that the chemicals in question are known to cause or can reasonably be anticipated to cause serious or irreversible chronic health effects in humans. As part of their claim, they argue that the Agency did not conduct a thorough weight-of-the-evidence evaluation and that, had it done so — considering such factors as the biological significance of the health effect, the severity of the effect, the dose level causing the effect, and the quality and quantity of available data — the Agency would not have and could not have concluded that each chemical represents a substantial hazard warranting inclusion on the TRI list.

As explained in detail below, the Court finds that EPA has satisfied the statutory criteria governing inclusion of new chemicals on the TRI list, and that it must defer to the Agency's expertise in weighing and evaluating the merits of individual scientific studies. A. N-Hexane

Commercial hexanes are manufactured by distillation from crude oil or natural gas. Hexane is used in the extraction of oils from seeds such as soybeans, cottonseed, flaxseed, safflower seed, corn germ, and peanuts.¹⁷ Plaintiff NOPA argues that EPA was arbitrary and capricious when it listed the chemical n-hexane (short for "normal-hexane") on the TRI list because it did not explain its decision making process in sufficient detail in the administrative record, and a number of the studies relied on by EPA were flawed.

17 It is also used as a reaction medium and solvent in the manufacture of polyolefins, synthetic rubbers, and pharmaceuticals; and to formulate products such as rubber cements, adhesives, lacquers, and printing inks.

The Agency contends that its decision was adequately documented in the administrative record and cites first to the proposed rule, where EPA presented its rationale for listing n-hexane:

In an epidemiology study, no neurological abnormalities were noted in workers. However, neurophysiological tests showed that the mean motor nerve conduction velocities of the exposed group was significantly decreased over the values of the control group. Also, the residual latency of motor nerve conduction of the posterior tibial nerve in the exposed group was significantly slowed when compared with the nonexposed group. . . . The alterations observed are consistent with n-hexane-induced peripheral neuropathy observed in other studies in humans and in animals. EPA believes that there is sufficient evidence for listing n-hexane [on the TRI list] pursuant to EPCRA section 313(d)(2)(B) based on the neurotoxicity data for this chemical.

59 Fed.Reg. 1817. EPA also noted that IRIS provided more detailed descriptions of the various studies that supported EPA's listing of n-hexane — studies which showed a number of neurotoxic effects caused by n-hexane, including blurred vision, abnormal color vision associated with nerve damage in the eye, loss of coordination, and numbness and tingling in the hands, fingers and feet in humans. See EPA MSJ (n-hexane) at 24-29 and citations to record therein.

EPA explains that after reviewing the available information on n-hexane and the comments by industry, it determined that the weight of the evidence supported the listing of n-hexane, even though no single study was conclusive. 59 Fed.Reg. 61,457. Specifically, in the final rule, EPA concluded:

[T]he review of the comments and data have not changed EPA's position that the weight of evidence supports a finding of chronic neurotoxicity for n-hexane. The weight-of-evidence determination contained in the Agency's IRIS data base is the basis for determining that n-hexane can reasonably be anticipated to cause neurotoxicity in humans. EPA reaffirms that there is sufficient evidence for listing n-hexane on the EPCRA.

59 Fed.Reg. 61,457 (Nov. 30, 1994) (col. 2) (final rule).

The Agency also counters NOPA's claim that a number of the studies relied upon by EPA are seriously flawed, undermining the weight-of-evidence review. The Agency does not dispute that some of the studies have limitations, see Response to Comments at 80-90, but argues that NOPA has not pointed to any relevant information that EPA did not consider, or produce any analysis that demonstrates EPA's conclusions to be scientifically invalid. Instead, argues the Agency, CMA has simply asserted a different conclusion based on the available information.

Based on a review of the administrative record, the Court concludes that the Agency was not arbitrary and capricious in listing n-hexane on the TRI. First, the administrative record was adequate for the Court to determine whether EPA made a reasoned decision. The basis for EPA's decision was made plain in the proposed rule, Response to Comments, and the final rule. It is clear that a number of studies showed the chemical's neurotoxic effects on humans, and that EPA considered these and other studies in IRIS — as well as the weaknesses of these studies — in listing the chemical. Neither the APA nor EPCRA requires an intricately detailed explanation of EPA's decision-making process. "It is simply not the case . . . that all of the essential postulates for an agency rule must be contained in the record." Center for Auto Safety v. Peck, 751 F.2d 1336, 1361 (D.C. Cir. 1985).

Second, the Agency has shown that the studies it relied upon met the statutory "sufficient for listing" standard. While it may be that some of the studies were less reliable than others, NOPA did not bring to EPA's attention — and has not brought to this Court's attention — any studies or information contradicting EPA's essential finding that n-hexane causes neurotoxic effects. Furthermore, neither the APA nor EPCRA requires that every study relied upon be of the highest possible quality. Congress requires only " sufficient evidence," 42 U.S.C. § 11023(d)(2) (emphasis added), to show that a chemical can " reasonably be anticipated to cause," 42 U.S.C. § 11023(d)(2)(B) (emphasis added), the toxic effect. Moreover, as noted earlier, this Court "must look at the decision not as a chemist, biologist, or statistician . . . but as a reviewing court exercising [its] narrowly defined duty to hold agencies to certain minimal standards of rationality." Ethyl Corp., 541 F.2d at 36-37. The record before the Court demonstrates that EPA conducted a weight-of-evidence review sufficient to meet the statutory requirement, and thus EPA was not arbitrary and capricious in listing n-hexane on the TRI.

B. 3-iodo-2-propynyl butyl carbamate ("IPBC")

IPBC is an off-white crystalline solid that is used in oil- and water-based paints, stains, adhesives, metal-working fluids, wood preservatives, textiles, plastics and other products. It is included in these products to prevent the growth of mildew, wood destroying fungi and other fungal organisms.

EPA's hazard evaluation of IPBC consisted of a review of the comprehensive toxicity database submitted by registrants of pesticide products containing IPBC under the Federal Insecticide, Fungicide, and Rodenticide Act ("FIFRA"), 7 U.S.C. § 136 et seq. Upon consideration of the information in that database, other scientific information and analysis, as well as Plaintiff Troy's comments, EPA concluded that IPBC "can reasonably be anticipated to cause in humans . . . serious . . . chronic health effects" under Section 313(d)(2)(B), 42 U.S.C. § 11023(d)(2)(B) and added IPBC to the TRI list. 59 Fed.Reg. 61,457.

EPA relied primarily on two animal studies showing that exposure to IPBC causes serious stomach lesions that are not readily reversible. In both a 90-day study and a two-year study, rats that were fed relatively low doses of IPBC developed stomach lesions and other abnormalities of stomach tissues. 59 Fed.Reg. 61,457.

Troy objects to the addition of IPBC on the TRI list, in part, because of its concern that EPA did not conduct a weight-of-the-evidence assessment of the entire database for IPBC. Instead, Troy claims, EPA merely relied on the two rat studies and concluded summarily that the listing criteria were met. In effect, Troy argues that the rulemaking record does not contain sufficient evidence or analysis to explain why the results of the 90-day and two-year IPBC rat studies are sufficient to establish that IPBC is known to cause or can reasonably be anticipated to cause serious or irreversible chronic health effects in humans.

The Court must defer to the evaluation conducted and conclusions drawn by senior EPA scientists, who reviewed many scientific studies¹⁸ on the effects of IPBC, evaluated the conclusions of the available studies, considered the assumptions and methodology used in conducting the studies, and assessed the validity and relevance of the results. 59 Fed.Reg. 1,790. Based on these considerations, as well as relying heavily on the findings of the two rat studies, EPA concluded that IPBC could reasonably be anticipated to cause lesions and other tissue abnormalities in humans and that these were chronic effects of sufficient seriousness to meet the statutory criteria under Section 313(d)(2)(B), 42 U.S.C. § 11023(d)(2)(B).

18 EPA reviewed 121 studies submitted to EPA's Office of Pesticide Programs to support registration of pesticide products containing IPBC under FIFRA.

While it is true that EPA might have been more thorough in presenting both its evidence and its rationale for drawing its conclusions, the Court is satisfied that EPA has offered sufficient evidence to establish that IPBC is known to cause or can reasonably be anticipated to cause serious or irreversible chronic health effects in humans — thereby satisfying the statutory requirement of EPCRA § 313(d)(2)(B), 42 U.S.C. § 11023(d)(2)(B).¹⁹ EPA has here, once again, met the minimal standard of rationality and has conducted a weight-of-evidence review sufficient to meet the statutory requirement.

19 Troy maintains that EPA should have explained (1) why it considers stomach lesions to be a serious health effect; (2) why stomach lesions in rats constitute an irreversible health effect; (3) why stomach lesions in rats have relevance for humans; (4) why stomach irritation in rats has relevance for a human health hazard assessment; and (5) why, given that stomach irritation was seen in only one species, it is sufficient to support the addition of IPBC to the § 313 list.
In its briefs, EPA offers a detailed response to each of these points. First, EPA notes that EPA did in fact directly address the issue of why it considers stomach lesions to be a serious health effect. See, e.g., 59 Fed.Reg. 61,457. Second, EPA notes that it did not conclude that the lesions were irreversible, nor was it required to do so. See 42 U.S.C. § 11023(d)(2)(B) (a listing is authorized based on a finding of "serious or irreversible" chronic health effects) (emphasis added). Third, EPA claims that since neither Troy nor its technical consultant raised, during the rulemaking process, the argument about the relevance for humans of rat studies (given that humans do not have a forestomach), EPA is under no obligation to address that issue now. Fourth, EPA concluded that it may properly rely on animal studies as "surrogates for toxicity testing to predict potential hazards to humans, except in a very few rare cases where effects have been determined to be species-specific." See 59 Fed.Reg. 61,459. And, fifth, EPA explained that its Draft Guidelines acknowledge that EPA may list substances based on studies on only one animal species. See Draft Guidelines at 29.

C. The Diisocyanates Category; IPDI; 2,2,4 TMDI; and 2,4,4 TMDI

Plaintiff CMA argues that EPA's listing of three out of twenty chemicals added to the TRI as part of the diisocyanates category was arbitrary and capricious. The three diisocyanates are: isophorone diisocyanate ("IPDI"), 2,2,4 trimethlyhexanethylene diisocyanate ("2,2,4 TMDI") and 2,4,4, trimethlyhexanethylene diisocyanate ("2,4,4 TMDI"). IPDI is a yellowish liquid that is used in the manufacture of specialty paints, coatings and adhesives to enhance their weather resistance. 2,2,4-TMDI and 2,4,4-TMDI are produced together as an isotonic mixture, that has similar, but more specialized uses than IPDI.

CMA asserts that EPA lacked sufficient evidence to list these chemicals within the diisocyanates category and, more generally, that EPA may not list chemicals as part of a category without demonstrating that each individual chemical meets the statutory criteria.

The first issue is whether Section 313(d) allows EPA to add a category of chemicals, as opposed to individual chemicals, to the TRI list. CMA's primary argument is that Congress chose to use the phrase "a chemical," in EPCRA, rather than "a chemical or chemical category." EPA counters by asserting that the phrase "a chemical" means only that EPA may not define categories so broadly that individual chemicals within the category cannot reasonably be expected to cause health effects.

As the Agency points out, Congress itself included 20 chemical categories on the initial TRI list when it enacted EPCRA. See List of Toxic Chemicals, Senate Committee on Environment and Public Works at 3 (1986) (original list includes, for example, antimony compounds, glycol ethers, and polybrominated biphenyls). Furthermore, in defining the TRI list in Section 313(c) of EPCRA, Congress used the word "chemicals" to refer to both the individual chemicals and the chemical categories that Congress included in its initial TRI list:

The toxic chemicals subject to the requirements of this section are those chemicals on the list in Committee Print Number 99-169 of the Senate Committee on Environment and Public Works, titled "Toxic Chemicals Subject to Section 313 of [EPCRA]" (including any revised version of the list as may be made pursuant to subsection (d) or (e) of this section).

42 U.S.C. § 11023(c). Because Congress used the word "chemicals" to refer to both individual chemicals and chemical categories, it is eminently reasonable for EPA to interpret the phrase in the same way the Congress itself used it, namely, that use of the term "a chemical" referred to either an individual chemical or a chemical category.

Moreover, Congress used the phrase "a chemical" in Section 313(d) to define what may be added to or deleted from the TRI list. Thus, if the phrase "a chemical" were interpreted to mean only specific individual chemicals, one implication would be that the chemical categories that Congress itself included in the original TRI list could never be removed, and that only the specific individual chemicals on Congress' initial list were subject to further review and removal under Section 313(d). There is no reason to think Congress intended such an outcome. It is more logical to conclude that Congress intended EPA to be able to reevaluate and remove both specific chemicals and chemical categories if and when such removal satisfied the statutory criteria.

For these reasons, the Court concludes that Section 313(d)(1) permits the Agency to add categories of chemicals to the TRI list, where appropriate, without demonstrating that each individual chemical meets the statutory criteria.

The second issue is whether EPA had sufficient evidence and provided a reasoned explanation for listing the diisocyanates category and IPDI, 2,2,4-TMDI and 2,4,4-TMDI as part of that category. Resolution of this issue centers upon the parties' opposing interpretations of extraordinarily complex scientific evidence. We are thus guided by the advice given by our Court of Appeals in NRDC, 824 F.2d at 1216, that it is "not for the judicial branch to undertake comparative evaluations of conflicting scientific evidence." This Court must not undertake an independent review of EPA's scientific judgments; our inquiry focuses only on whether the Agency has met the statutory requirement for "sufficient evidence."

CMA argues that EPA was arbitrary and capricious because there is no data in the record specifically on 2,2,4-TMDI or 2,4,4-TMDI, and the only study on IPDI did not find chronic pulmonary irritation. Moreover, argues CMA, the studies relied upon by the Agency to list the three chemicals are based on diisocyanates with different physical and chemical properties than IPDI, 2,2,4-TMDI and 2,4,4-TMDI. Thus, concludes CMA, EPA simply "speculated" that IPDI, 2,2,4-TMDI and 2,4,4-TMDI would cause the same effects as the tested chemicals.

EPA explains that to support the listing of both the diisocyanates category and the individual chemicals that comprise it, the Agency reviewed a number of studies of individual diisocyanates. The studies showed that six chemicals with the diisocyanate structure caused pulmonary irritation as a result of inhalation, including such effects as coughing, difficulty breathing, and death. EPA argues that the six chemicals showed these effects despite differences in chemical structure, physical state, molecular weight and vapor pressure. From this evidence EPA concluded that their toxicity was due to the common characteristic shared by the six chemicals, namely, presence of the diisocyanate portion of the molecule, and that this toxicity occurs despite the differences in their other chemical and physical characteristics. 59 Fed.Reg. 61,453-54. Given this finding, EPA concluded that other chemicals with the diisocyanate structure could also reasonably be anticipated to cause pulmonary irritation because, despite difference in various characteristics, they each share the key characteristic: the diisocyanate portion of the molecule. 59 Fed.Reg. 61,453; Response to Comments at 48-49; 51-52. Thus, EPA listed the diisocyanates category, which included 20 individual chemicals, each with the diisocyanate structure.²⁰

20 Of the 20 individual chemicals in the diisocyanates category, 19 were listed for the first time in this rulemaking and one (methylenebis (phenlisocyanate)) was already on the list and was moved into the diisocyanates category. 59 Fed. Reg. at 61,454 (cols. 2-3).

In short, the Agency believes that IPDI, 2,2,4-TMDI and 2,4,4-TMDI are properly part of the diisocyanates category because all three contain the diisocyanate structure. Moreover, although IPDI, 2,2,4-TMDI and 2,4,4-TMDI have varying physical and chemical characteristics, comparing the three diisocyanates to the six diisocyanates discussed in the studies shows that their characteristics fall within the ranges created by the six tested diisocyanates. The Agency provides extensive scientific evidence for this claim. See EPA MSJ (CMA) at 47-49 and sources cited therein.

Upon review of the administrative record, the Court concludes that the Agency was not arbitrary and capricious in listing the diisocyanates category or including IPDI, 2,2,4-TMDI and 2,4,4-TMDI as part of the category. With respect to the category listing, EPA clearly has sufficient evidence to support the listing, and CMA offers no substantive argument that the diisocyanates category is improperly listed on the TRI.

The same conclusion applies to the three individual chemicals. While none of the studies relied on by EPA specifically involved IPDI, 2,2,4-TMDI or 2,4,4-TMDI, EPA has provided a reasoned explanation for their inclusion in the category: they all possess the diisocyanate structure, which is most likely the source of the chemicals' toxicity. Moreover, EPA's explanation was documented fully in the record. CMA believes EPA has wrongly analyzed available scientific data, and CMA has vigorously attacked EPA's scientific conclusions ( i.e., arguing that because the three chemicals have different chemical, physical and molecular make-ups than the tested chemicals they should not be part of the category). But CMA has failed to show that EPA's conclusions are unreasonable, and it is certainly not the job of this Court to make an independent evaluation of the scientific evidence.

Given the record before the Court, EPA had "sufficient evidence" to conclude that the diisocyanates category should be listed on the TRI and that IPDI, 2,2,4-TMDI and 2,4,4-TMDI should be listed as part of the category.²¹

21 CMA also asserts that EPA failed to consider a subchronic study of IPDI which concluded that the chemical did not cause pulmonary irritation. The Agency argues, however, that the study was not considered because it was (1) short term (4 weeks) and thus did not disprove EPA's conclusion that pulmonary irritation may arise following long-term exposure; and (2) it is not clear what those conducting the study did to check for pulmonary irritation. EPA's explanation is reasonable, and the Agency certainly has the discretion to exclude a study it considers insignificant or flawed. Portland Cement Ass'n v. Ruckelshaus, 486 F.2d 375, 394 (D.C. Cir. 1973).
CMA also argues that when EPA explained its listing decision for the purposes of this lawsuit, it improperly relied on a document called the Dynamac Report, a 350-page document prepared in 1987 that summarizes numerous studies on diisocyanates. CMA argues that in the proposed rule, EPA did not cite to this document or reference any individual studies that it believed supported its listing decision. CMA argues that EPA's citation in the final rule to "available data on the chronic pulmonary toxicity for several members of the diisocyanates category" is insufficient to provide an adequate administrative record.
While it may be true that EPA did not explicitly identify the Dynamac Report in the proposed rule, the Agency certainly identified it in response to the comments of CMA and others on EPA's proposed listing of the diisocyanates category. See Response to Comments at 51-52. Further, in the proposed rule, EPA's discussion of the proposed listing of the diisocyanates category identifies "Reference 7," which is EPA's support document for the addition of chemicals from the Clean Air Act (the "CAA Support Document"). 59 Fed.Reg. at 1816, 1845. The CAA Support Document in turn identifies the Dynamac Report as the underlying source of information.

D. 2,6-DMP

CMA challenges the Agency's listing of 2,6-Dimethylphenol ("2,6-DMP"),²² arguing that the two studies relied on by EPA were inadequate to meet the statutory standard. The issue reduces itself to whether EPA may rely on two Russian studies conducted in the 1960s to support its listing of 2,6-DMP.

22 The parties did not provide a description of 2,6 DMP.

CMA contends that EPA improperly relied upon the Russian studies because they do not provide a sufficient basis for evaluating human health hazards. CMA notes that documents in the administrative record state that confidence in the 2,6-DMP studies is low because details of the experiments are not available. For example, the Russian studies did not provide information about (1) animal strain, (2) number of animals tested at each dose level, (3) number of animals in the control group, or (4) purity of test articles. Furthermore, in 1990, the Interagency Testing Committee ("ITC") concluded that there was "insufficient evidence" to evaluate both Russian studies.²³ CMA concludes that without this missing information, it is not possible to determine if the 2,6-DMP experiments are valid.

23 The Interagency Testing Committee is responsible for recommending chemicals for testing under the federal Toxic Substances Control Act ("TSCA"). 15 U.S.C. § 2603(e). The ITC is comprised of representatives from several federal agencies, including EPA.

The Agency argues that the Russian studies provided sufficient evidence to support the listing. In the proposed rule, EPA explained the findings of those studies:

Oral administration of [2,6-DMP] to rats for 8 months produced histologic lesions . . . in the liver, kidneys and spleen. Another supporting oral study in rats . . . also reported histological lesions in the liver . . . of rats following subchronic oral administration of [2,6-DMP]

59 Fed.Reg. 1810.²⁴ EPA also claims that it considered confirming toxicological information in the Hazardous Substances Data Bank ("HSDB") file on 2,6-DMP. The HSDB is a peer-reviewed database maintained by the National Library of Medicine on the toxicology of potentially hazardous chemicals.²⁵

24 EPA's summary of the two studies in the preamble to the proposed rule mistakenly stated that both studies showed effects on the kidneys. In fact, only the eight-month study showed effects on the kidneys. EPA admitted this error in its motion papers. See EPA MSJ (CMA) at 56 n. 38.

25 Upon publication of the proposed rule, EPA placed a file in the public docket that contained the evidence regarding 2,6-DMP reviewed by EPA in the course of its hazard assessment for this chemical. The HSDB print-out for 2,6-DMP was included in that file and available to potential commenters.

The Agency argues that CMA's list of the studies' alleged shortcomings conveniently ignores key details about the studies which have been reported. The reports on these studies reveal the number of animals tested, the method of dosing, the dosage levels and duration, and the existence of control groups. EPA stresses that the reports reveal that "pronounced" effects on the liver and kidneys resulted at "relatively low dose levels." 59 Fed.Reg. 61,455. More importantly, says EPA, while the Russian studies might not be ideal, no study undertaken in the last 25 years has cast doubt on their results.

Furthermore, EPA acknowledged in its Response to Comments that the absence of information on some of the details of the two studies had led the ITC to characterize the studies as meriting only "low confidence." EPA's position, however, is that although more detail regarding the design of the studies would be desirable for the type of quantitative analysis undertaken in a risk assessment under other statutes, sufficient detail was available for EPA to make the qualitative analysis involved in the hazard assessment required under Section 313 of EPCRA. See 59 Fed.Reg. at 61,455 (col. 1); Response to Comments at 58, 208. The Agency argues that a "low confidence" rating is very different from a "no confidence" rating, and that the Agency does have some confidence in the studies, as evidenced by its use of them in other contexts. For example, EPA uses these studies as evidence of the potential for adverse human health effects in proposing testing for other toxicity endpoints under TSCA. In addition, although the Agency has not set a Reportable Quantity for 2,6-DMP under Section 102 of the Comprehensive Environmental Response, Compensation and Liability Act of 1980 ("CERCLA"), 42 U.S.C. § 9602, EPA used the two Russian studies to prepare a Health and Environmental Effects Profile of 2,6-DMP for purposes of setting a Reportable Quantity of 100 pounds of this substance. In sum, it is EPA's position that while the studies may have been less than ideal, they provided sufficient evidence to justify the listing of 2,6-DMP.

Based on its evaluation of the Russian studies and information from the HSDB, EPA concluded that there was sufficient evidence to list 2,6-DMP based on hepatoxicity (toxicity to the liver) and nephrotoxicity (toxicity to the kidneys). Id.

The Court concludes that in light of the relevant data that is known about the Russian studies, the fact that their results have not been invalidated in 25 years, consideration of the HSDB, and the substance of EPA's explanation for its listing decision, the Agency's decision to list 2,6-DMP was not arbitrary and capricious. While it would be desirable to know more about the Russian studies, EPA has provided a reasonable explanation for why they are sufficient to meet the statutory standard and justify listing 2,6 DMP on the TRI.²⁶

26 CMA also objected to EPA's citation to a report on 2,6-DMP contained in the HSDB because, CMA claims, the Agency did not include the report in the administrative record. While the study was not mentioned in the preamble to the final rule, the published report of this study was summarized in two documents that are in the administrative record and were among materials regarding 2,6-DMP that EPA scientists reviewed: (1) a profile of 2,6-DMP prepared by EPA's Health and Environmental Review Division (the "HERD Profile"); and (2) a print-out of a file on the toxicology of 2,6-DMP maintained in the peer-reviewed HSDB.

E. Bronopol

Plaintiff CMA argues that EPA was arbitrary and capricious in listing 2-bromo-2-nitropropane-1,3-diol ("Bronopol"). Bronopol is a water-soluble, white, crystalline solid that acts as a broad-spectrum antimicrobial agent and has a wide range of industrial uses, including as a biocide for paper treatment, recirculating cooling systems, air scrubbers and air conditioning systems. Bronopol is used to prevent bacterial spoilage of latex paints, water-based printing inks and fountain solutions, and adhesives.

EPA added Bronopol to the TRI list based primarily upon a two-year study in rats which showed irritation of the gastrointestinal tract, ulceration and stomach lesions. 59 Fed.Reg. 1800. The Agency also summarized the results of four other animal studies that confirmed the two-year study's finding of "severe" effects on the gastrointestinal tract following ingestion of Bronopol. Id. ²⁷ EPA concluded that the five studies together showed sufficient evidence to warrant listing on the TRI. 59 Fed.Reg. 61,450.

27 The five studies were submitted to EPA under FIFRA by registrants of pesticide products containing Bronopol. EPA scientists specializing in the topic of each study, reviewed the studies and prepared one or more Data Evaluation Reports ("DERs") on each. The DERs summarize the study findings, evaluate the quality of the respective study, and make certain conclusions regarding the utility and nature of the study findings. EPA then summarizes the study findings in a database called a "Tox-One-Liner," which references the studies and the DERs by document numbers.

Plaintiff CMA's primary argument is that EPA was arbitrary and capricious when it listed Bronopol because the three effects cited by EPA — lesions of the gastrointestinal tract, inflammation of the gastrointestinal tract and a thickening of the walls of the gastrointestinal tract — are acute, not chronic, as required by the statute.²⁸

28 Plaintiff also argues that EPA's discussion of the Bronopol studies contained inaccuracies and unsupported conclusions. The agency has provided a reasonable explanation to every point raised by Plaintiff, and the Court need not repeat the extensive, point-by-point, argument here. See EPA Reply at 56-61.

The Agency does not dispute that Bronopol can cause acute health effects. Indeed, in the rulemaking, EPA acknowledged that Bronopol causes gastric effects after short-term exposure to high doses ( i.e., an "acute" response). But, EPA argues, it also found that gastric effects appeared after long-term exposure to low-doses (i.e. a "chronic" response). EPA says that after examining the relative dose levels, duration of treatment, and the incidence and types of effects found in the studies, it concluded that the evidence that Bronopol caused certain acute effects did not negate its conclusion that Bronopol also caused chronic human health effects. 59 Fed.Reg. 61,450. In the final rule EPA explained:

Although the Agency agrees with the comment that [bronopol] presents a moderate acute hazard, EPA does not agree that the chemical is not a chronic toxicant. The effects noted in both acute and chronic studies, cited in the proposed rule, indicate irritation due to exposure in the compound. However, differing expressions of irritation are obtained depending upon the level of material to which the test animals were exposed and the duration of the exposure. In the acute studies cited in the proposed rule, the acute gastric effects were seen at relatively high doses. In the chronic studies, cited in the proposed rule, the effects . . . were noted following repeated oral exposure to lower doses of [bronopol].

Id.

EPA contends that CMA is incorrect that a chemical can cause only acute toxic effects or chronic toxic effects, but not both. In EPA's view, a finding of acute toxicity and a finding of chronic toxicity are not mutually exclusive. Thus, when CMA points to various acute toxicity studies to prove that Bronopol does not exhibit chronic toxicity, it does not disprove the findings of the chronic two-year study, and cannot disprove those findings unless one assumes that a chemical cannot exhibit both acute and chronic toxicity.

In consideration of these arguments, the Court concludes that EPA was not arbitrary and capricious in listing Bronopol. Put simply, CMA has offered little to disprove the findings of the primary study relied upon by the Agency. EPA has certainly provided a reasoned explanation for its conclusion that Bronopol causes chronic health effects. See EPA MSJ (CMA) at 73-78 and citations to the record therein.²⁹ EPA has also adequately documented its decision-making process with respect to listing Bronopol. EPA scientists did not, as CMA suggests, review only the Tox-One-Liner database for Bronopol. In the Response to Comments, EPA said:

29 EPA also fully responded to the comments that Bronopol should not be added to the TRI list because the FDA has approved it for certain uses. See Response to Comments at 23; EPA MSJ (CMA) at 78-79.

Although the 1988 Tox-One-Liners were used as part of the Agency's evaluation of the toxicity of a candidate chemical, a number of other sources were also used. These include . . . data evaluation records for studies used in support of listing. Therefore, evaluations of the toxicity of individual chemicals has been made on the entire body of available information on the specific chemicals and did not rely on the 1988 Tox-One-Liners data base.

Response to Comments at 199-200. Moreover, although the studies and the DERs referenced in the Tox-One-Liner database were not in the administrative record, they were specifically referenced in the Tox-One-Liner database. In short, the substance of EPA's evaluation was well documented in the Tox-One-Liner, the preambles to the proposed and final rules, and in EPA's Response to Comments.

For these reasons, the Court holds that EPA was not arbitrary and capricious in listing Bronopol on the TRI.

F. Chlorinated Alpha-Olefins

Plaintiff CMA challenges EPA's decision to include chlorinated alpha-olefins within a category of chemicals called polychlorinated alkanes. Polychlorinated alkanes are chlorinated linear hydrocarbons; that is, they consist of chains of carbon, hydrogen and chlorine atoms. The listed category of polychlorinated alkanes consists of those with 10, 11, 12 or 13 carbon atoms in each molecule, which are commonly referred to as "C₁₀-C₁₃", "C₁₀-₁₃" or "short chain" polychlorinated alkanes. See 59 Fed.Reg. 61,463.

As part of this category — short-chain polychlorinated alkanes — EPA included chlorinated alpha-olefins, which are short-chain polychlorinated alkanes derived from a particular feedstock. CMA argues that chlorinated alpha-olefins are physically and chemically distinct from short-chain polychlorinated alkanes (the listed category) and thus EPA acted arbitrarily when it included chlorinated alkanes in the polychlorinated alkane category. More specifically, CMA attacks EPA's decision because the Agency's data on the short-chain polychlorinated alkanes did not include studies employing alpha-olefin derived polychlorinated alkanes. CMA claims that the toxicity of a short-chain polychlorinated alkane (the listed category) is attributable to the portion of the mixture with carbon chain lengths of 9 and 10, and therefore cannot be used to predict the toxicity of alpha-olefin derived polychlorinated alkanes, which generally have a carbon chain length of 12.

The Agency explains that its decision to list the short-chain polychlorinated alkanes category rested on studies in which a mixture of short-chain (10-13 carbons) alkanes chlorinated to 58 percent was shown to be carcinogenic in rats and mice, causing liver, kidney and thyroid tumors. 59 Fed.Reg. 1802. The International Agency for Research on Cancer ("IARC") has also classified short-chain polychlorinated alkanes as a probable human carcinogen. 59 Fed.Reg. 61,462. Thus, EPA concluded there was sufficient evidence to list short-chain chlorinated paraffins (a common name for an alkane) as a category on the TRI list.³⁰ 59 Fed.Reg. 61,461.

30 The Agency originally proposed to add to the TRI list a category called "chlorinated paraffins," which was the same as the listed category, short-chain polychlorinated alkanes, except that it also included "long chain" chlorinated paraffins i.e., it consisted of linear chlorinated hydrocarbons with an average chain length of 10 to 30 carbon atoms, instead of the "short chain" 10 to 13 version.
Commenters challenged EPA's proposal primarily on two grounds: (1) that EPA had insufficient evidence regarding the carcinogenity of long-chain polychlorinated alkanes, i.e. those with more than 13 carbon atoms, and (2) that EPA's category definition improperly included polychlorinated alkanes obtained through chlorination of olefin feedstocks when the laboratory tests on which EPA based the proposed listing employed chlorinated paraffins obtained through chlorination of a paraffin mixture, not an olefin-based mixture. EPA considered these comments and, in response, decided not to add the longer-chain substances to the TRI list. 59 Fed.Reg. at 61,462.
EPA did not change its decision, however, to include within the listed category those short-chain linear chlorinated hydrocarbons that are derived from olefin feedstocks ( i.e., to include alpha-olefins in the category). EPA's consideration of the comments on this issue instead led the Agency to change the name of the listed category from "chlorinated paraffins" to "polychlorinated alkanes."

In the final rule, EPA stated that olefin-derived polychlorinated alkanes were included in the category because the toxicity data for chlorinated paraffins was sufficient for all polychlorinated alkanes that fall within the same carbon number and chlorine content — regardless of what materials were used to manufacture them. 59 Fed.Reg. 61,461. While it is more than a bit difficult to parse EPA's scientific explanation, it appears that, basically, EPA found alpha-olefins were similar enough to the tested short-chain alkanes so that alpha-olefins could be included within the short-chain polychlorinated alkanes category.

EPA also counters CMA's assertion that the toxicity of short-chain polychlorinated alkanes (the listed category) is attributable to the portion of the mixture with carbon chain lengths of 9 and 10. The agency argues that no evidence in the record supports CMA's thesis that the C₉ and C₁₀ molecules are the cause of the observed toxicity of the tested short-chain polychlorinated alkane mixtures. See EPA MSJ (CMA) at 92-96; Reply at 63-64. In particular, EPA argues that the Serrone paper, cited by CMA to support its position, does not show or even suggest that C₉ or C₁₀ polychlorinated alkane molecules are significantly more toxic than C₁₂ polychlorinated molecules. See id. at 96; Reply at 64 n. 41.

As the preceding discussion shows, this controversy is highly technical,³¹ and rests almost completely on the interpretation of arcane scientific data. The parties spend a good part of their motions papers discussing this scientific evidence, see e.g. CMA MSJ at 9-14; EPA MSJ (CMA) at 85-91; CMA Reply 7-16; and as this Court has already noted, we must "look at the decision not as a chemist, biologist, or statistician that we are qualified neither by training nor experience to be, but as a reviewing court exercising our narrowly defined duty of holding agencies to certain minimal standards of rationality." Ethyl Corp., 541 F.2d at 36-37.

31 The Court appends to this Opinion, only two diagrams — of many submitted — as an example of the complex arguments presented by the parties.

The issue here is whether, given the evidence in the record, EPA's judgment is reasonable, not whether EPA's evaluation of the scientific data and arguments is correct. Given the administrative record and the exhaustive elaboration of that record in the parties' briefs, the Court concludes that EPA was not arbitrary and capricious in listing chlorinated alpha-olefins within the polychlorinated alkanes category. EPA adequately explained why CMA's primary attack on EPA's listing — that C₉ or C₁₀ polychlorinated alkane molecules are significantly more toxic than C₁₂ polychlorinated molecules — failed, and there is insufficient evidence to prove CMA's related argument that toxicity of the tested short-chain polychlorinated alkane mixtures cannot reasonably be attributed to chlorinated alpha-olefins. See EPA Reply at 64-66. EPA has also reasonably addressed CMA's comments regarding the SAR memorandum. See EPA Reply at 66-68.

For these reasons, the Court concludes that the Agency was not arbitrary and capricious in listing the chlorinated alpha-olefins within the polychlorinated alkanes category on the TRI.³²

32 In addition to challenging EPA's listing decisions on the Diisocyanates category, 2,6-DMP, Bronopol and Chlorinated Alpha-Olefins, Plaintiff CMA also broadly challenges EPA's listing of 152 chemicals on procedural grounds. See CMA Statement of Points and Authorities In Support Of Its Motion For Summary Judgment On Claim I. CMA argues that EPA (1) listed 152 chemicals based solely on the Agency's toxicity screening criteria, rather than the statutory criteria; (2) improperly relied on information from one or more databases; and (3) improperly failed to consider exposure in the listing decisions. For reasons stated fully in the Court's Opinion, CMA's broad-based claim fails. See Sections III, IV and VI(A).

G. NMP

NMP is a general purpose solvent, which is used as a paint stripper, as well as in lube oil extraction, electronics, magnet wire coatings, engineering resins, and agricultural chemicals.

EPA undertook a detailed review for over nine years of NMP's developmental and reproductive toxicity under the Toxic Substances Control Act ("TSCA"), 15 U.S.C. § 2601-2692, before the present rulemaking began. NPG's members³³ have participated at each step of this laborious process, and have had numerous opportunities to comment on EPA's conclusion that NMP exhibits developmental and reproductive toxicity.

33 NPG is the NMP Producer's Group.

In the course of its TSCA review, EPA examined six studies that had been done on NMP and summarized this review in a memorandum dated April 27, 1989. Section 8(e) Memorandum at 1-2 (NPG App. 3). This memorandum concluded that NMP had been shown to cause toxic effects in rats and mice.

Based on the demonstrated potential for developmental toxicity, and the potential for human exposure to NMP, EPA proposed to require testing designed to investigate further NMP's developmental toxicity, as well as its potential to cause reproductive and other types of toxic effects. EPA requested comments on its findings regarding toxicity and exposure, and specifically requested the submission of any existing studies that had not been made available to EPA.

EPA received numerous comments on the proposed test rule from manufacturers and processors of NMP. After the close of the comment period for the proposed test rule, EPA initiated a detailed review of two 1991 studies performed by NPG. EPA concluded that "[t]hese studies indicated the potential for reproductive and developmental effects resulting from exposure to NMP." Lifecycle Analysis at 11 (NPG App. 3). Further, "[b]ecause the potential for such effects appeared to be significant," EPA decided "that the reproductive and developmental risks from NMP should be assessed and appropriate risk management activities initiated on an expedited basis." Lifecycle Analysis at 11 (NPG App. 3).

In April 1992, EPA notified the NPG members of its concerns regarding NMP's developmental and reproductive toxicity and its intent to expedite a risk assessment regarding NMP-based paint strippers. EPA then proceeded to perform an expedited risk assessment of NMP in paint stripping products, which it concluded on September 8, 1993, with the issuance of the "Lifecycle Analysis and Pollution Prevention Assessment for [NMP] in Paint Stripping." NPG App. 3. In the final Lifecycle Analysis, EPA claimed that "the available studies establish the potential for significant risk of reproductive and developmental effects in humans unless dermal exposures are prevented." Lifecycle Analysis at 15 (NPG App. 3.)

Ultimately, EPA concluded:

Reproductive and developmental studies in animals exposed to NMP showed effects including reduced fertility and reduced mean offspring body weight at birth and throughout lactation. These effects in animals suggest that similar effects may occur in humans.

Lifecycle Analysis at 47.

EPA's conclusion, that NMP exhibits reproductive and developmental toxicity, was based upon the detailed analysis and study EPA conducted under TSCA. Given that existing detailed assessment of the hazards of NMP, EPA decided it was unnecessary to use the screening procedures being applied in this proceeding and, instead, determined that NMP should be one of the 383 candidates for listing in the rulemaking process.

EPA considered NMP, along with the other 382 candidate chemicals, to determine whether the scientific evidence supported its listing under one or more of the three criteria set forth in Section 313(d)(2). 42 U.S.C. § 11023(d)(2). Having already performed a detailed hazard evaluation for NMP under TSCA, EPA reviewed all available studies, including the 1991 Rabbit Developmental Study and the 1991 Rat Reproductive Study, and determined that NMP exhibits toxic effects justifying inclusion on the TRI list. See 59 Fed.Reg. 1790-93; see also 59 Fed.Reg. at 1823.

EPA received comments on its proposal to list NMP from NPG and others. These comments provided no new data on the developmental or reproductive toxicity of NMP. On November 30, 1994, EPA published its final Rule, adding NMP and 285 other chemicals that it originally considered for the TRI list. See 59 Fed.Reg. 61,432.

Plaintiff NPG contends that EPA failed to meet the Section 313 statutory listing criteria, given that EPA did not conduct any hazard assessment before making its final decision to add NMP to the TRI list. Accordingly, there was less than a full weight-of-the-evidence evaluation of all available data and, therefore, EPA did not meets its obligations under the statutory guidelines.

Contrary to NPG's claims, EPA did conduct a hazard assessment, a summary of which is found in the preambles to the proposed and final rules. See 59 Fed.Reg. 1823; 59 Fed.Reg. 61,458-59. That assessment is also fully documented in EPA's September 1993 "Lifecycle Analysis and Pollution Prevention Assessment for N-Methylpyrrolidone (NMP) in Paint Stripping."

The Lifecycle Evaluation reflects the results of EPA's review of nine studies on the reproductive and developmental toxicity of NMP.³⁴ Among the developmental malformations found in these studies were low birth weight (Lifecycle Analysis at 94), reduced postnatal survival ( id.), misshapen skull bones and cardiovascular problems ( id. at 93), and other dysfunctions of skeletal systems and major organs (Section 8(e) Memorandum). Neither NPG nor any other commenter offered any developmental toxicity studies that disproved EPA's conclusion.

34 These are the three studies discussed in Appendix C to the Lifecycle Analysis (the 1991 Rabbit Development Study, the 1991 Rat Reproductive Study, and a two-generation reproductive and developmental toxicity study submitted by DuPont) and the six developmental studies discussed in the Section 8(e) Memorandum, which is attached to Appendix C of the Lifecycle Analysis.

NMP's developmental toxicity is sufficient to warrant the addition of NMP to the TRI list, given that only one of the listing criteria need be met in order to list a chemical or chemical category under EPCRA § 313. See 59 Fed.Reg. 1793. Nonetheless, EPA also found evidence of reproductive toxicity of NMP, based primarily on the 1991 Rat Reproductive Study ( see 59 Fed.Reg. 1823), and found no valid study indicating the contrary.

EPA considered the comments received from NPG's members at numerous stages in the Agency's long evaluation of NMP under TSCA, and reviewed all of its findings again during the present rulemaking. On the basis of all the information and analyses before it, EPA concluded that NMP "can reasonably be anticipated to cause in humans . . . teratogenic effects, or . . . serious or irreversible . . . reproductive dysfunctions." 42 U.S.C. § 11023(d)(2)(B). NPG's claim that EPA did not perform a hazard evaluation and that, therefore, its conclusions are insufficiently supported, is belied by the evidence. The Court concludes that EPA's hazard evaluation was sufficient to be the basis for its conclusion that NMP should be added to the TRI list.³⁵

35 NPG is essentially asking the court to independently evaluate the scientific evidence. This we may not do. NRDC, 824 F.2d 1211, 1216 (D.C. Cir. 1987). While the Court has avoided the temptation to engage in scientific debates arising from conflicting scientific analyses, it has undertaken the judicially proper function of addressing the threshold issue of whether EPA has performed a sufficient hazard evaluation necessary to sustain its conclusions and subsequent listing of NMP.

VIII. Conclusion

To carry out its statutory mandate under EPCRA to provide the public and state and local governments with maximum access to information about potential chemical hazards in their communities, EPA undertook a massive administrative effort, starting in 1992, to determine what additional chemicals belonged on the TRI list. Beginning with consideration of more than 1500 chemicals, EPA applied screening procedures and exclusion policies which winnowed that number down to 383 chemicals which, in 1994, became the candidates for placement on the TRI list in this rulemaking. Ultimately, in November of 1994, EPA concluded that 279 chemicals and 7 chemical categories satisfied the statutory requirements for inclusion on the TRI list.

The history of this rulemaking demonstrates that EPA scientists and technical staff devoted enormous amounts of time to the scientific evaluations that formed the basis of the listing decisions. In fact, the administrative record encompasses a seventy-page proposed rule, an extensive round of comments, a fifty-three page final rule, and a fifty-nine page support document, in addition to the hundreds of citations to various scientific articles, studies, papers, and databases on which the staff relied. In light of this record, Plaintiffs' attempt to paint the rulemaking process as a rush to judgment fails.

That is not to say that the process was perfect, nor that every technical conclusion was documented with as much detail as, from hindsight, would prove desirable, nor that every policy judgment was articulated as fully as the Court might prefer. But the principles of administrative law do not demand perfection in the administrative process. The Court has explained in great detail why it did not find the Agency's conclusions or procedures to be arbitrary, capricious, an abuse of discretion, or violative of the statute. To the contrary, the Court concludes that EPA went to great lengths to separately evaluate each and every chemical on the basis of the relevant data, and to make scientific and technical judgments which are clearly outside the expertise of a reviewing court. While some of the explanations of the Agency are too sparse for this Court's liking, there is no question that the basic steps on the Agency's decision-making path can be reasonably discerned.

In short, EPA has satisfied the procedural demands of EPCRA and the APA; it has based its listing decisions on accepted scientific principles, laboratory tests, and appropriately designed and conducted epidemiological and population studies, as required by EPCRA; it has exercised its statutory discretion reasonably; and it has explained its decisions so the Court may determine whether they were "based on a consideration of the relevant factors and whether there has been a clear error of judgment." Citizens to Preserve Overton Park, 401 U.S. at 416, 91 S.Ct. at 823.

For these reasons, the Agency's Motions for Summary Judgment must be granted, and the Plaintiffs' Motions for Summary Judgment must be denied. Orders consistent with this opinion will issue in each case.