These cases are before the court upon petitions (filed in May and June 1973) to review an order of the Assistant Secretary of Labor for Occupational Safety and Health, published on May 3, 1973, 38 Fed.Reg. 10929, issuing an Emergency Temporary Standard on Certain Carcinogens, 29 C.F.R. Sec. 1910.93c. The temporary standard prescribes plant operating procedures and equipment, work practices and procedures for preventing exposure to 14 chemicals said to be carcinogens. This court has jurisdiction to review and set aside the standard pursuant to section 6(f) of the Occupational Safety and Health Act of 1970, 29 U.S.C. Sec. 655(f).

In No. 73-1361, petitioners are Dry Color Manufacturers' Association, Inc., Sun Chemical Corporation, Inmont Corporation, H. Kohnstamm & Co., Inc., Allied Chemical Corporation, Chemetron Corporation, The Upjohn Company, Lakeway Chemicals, Inc. and Hercules, Incorporated (hereinafter "Dry Color"). Each is a manufacturer, user, or an association of users of 3,3'-dichlorobenzedine (DCB), one of the 14 chemicals. Petitioner in No. 73-1638, Aerojet-General Corporation (hereinafter "Aerojet") is a user of ethyleneimine (EI), another of the 14 chemicals subject to the Emergency Temporary Standard. The petitioners in these two cases (hereinafter "industry petitioners") raise as objections to the Emergency Temporary Standard that: (1) there is not substantial evidence in the record to show that the use of DCB and EI, respectively, satisfies the provisions of subsection 6(c)(1) of the Occupational Safety and Health Act of 1970, 29 U.S.C. Sec. 655(c)(1), as to the conditions necessary to justify the promulgation of an emergency temporary standard; (2) the findings of fact and statement of reasons for the standard contained in the standard's preamble are inadequate; and (3) the Assistant Secretary violated the National Environmental Policy Act of 1969, 42 U.S.C. Sec. 4332(2)(C), by failing to prepare an environmental impact statement prior to issuance of the standard. In addition, the industry petitioners have challenged the Department of Labor's Certification of Record, charging that a substantial portion of the documents contained therein were never actually considered by the Department in its decision to issue the Emergency Standard and, therefore, should not be included in the record.

Petitioners in No. 73-1383 are Oil, Chemical and Atomic Workers Union, many of whose members work in plants that manufacture or use chemicals subject to the Emergency Temporary Standard, and the Health Research Group, a non-profit organization engaged in public interest research and advocacy on health issues, including occupational health. These petitioners (hereinafter "employee petitioners") contend that the Emergency Temporary Standard falls short of preventing the impairment of worker health which the Act requires and seek to have this court order the implementation of a use permit system that would achieve zero exposure.

The Occupational Safety and Health Act of 1970, 29 U.S.C. Sec. 651 et seq., (hereinafter "the Act") provides for the issuance of both permanent and emergency temporary standards.1 Under subsection 6(b) of the Act, 29 U.S.C. Sec. 655(b), the Secretary may promulgate permanent standards by following rulemaking procedures similar to those prescribed by the Administrative Procedure Act, 5 U.S.C. Sec. 551 et seq., with the added assistance of a statutorily authorized Advisory Committee if deemed necessary. Under subsection 6(c), 29 U.S.C. Sec. 655(c), on the other hand, emergency temporary standards become effective immediately upon publication in the Federal Register; the hearing, public comment, and other provisions of the Administrative Procedure Act do not apply. After issuing an emergency temporary standard, the Secretary must begin a permanent rulemaking process and complete that process by issuing a permanent standard within six months of the publication of the emergency temporary standard.

Subsection 6(c)(1) of the Act commands the Secretary to issue an emergency temporary standard "if he determines (A) that employees are exposed to grave danger from exposure to substances or agents determined to be toxic or physically harmful or from new hazards, and (B) that such emergency standard is necessary to protect employees from such danger," 29 U.S.C. Sec. 655(c)(1). Subsection 6(e) provides that "[w]henever the Secretary promulgates any standard, . . . he shall include a statement of his reasons for such action," 29 U.S.C. Sec. 655(e), and subsection 6(f) directs courts reviewing a standard that "[t]he determinations of the Secretary shall be conclusive if supported by substantial evidence in the record considered as a whole," 29 U.S.C. Sec. 655(f).

The Emergency Temporary Standard in question was originally published by the Occupational Safety and Health Administration ("OSHA") on May 3, 1973, after about one year of consulting with the National Institute for Occupational Safety and Health ("NIOSH") and receiving data and commentary from other interested groups. The Emergency Temporary Standard was based on the following findings, which are stated in the preamble of the standard and constitute the entire statement of reasons for its issuance:

"On the basis of all the relevant information before us, it is hereby found that: (1) the 14 carcinogens listed in the emergency temporary standard are toxic and physically harmful; (2) that exposure to any of the 14 substances poses a grave danger to employees; (3) that employees are presently being exposed to the substances; and (4) that the emergency temporary standard set forth below is necessary to protect the employees from such exposure." 38 Fed.Reg. 10929.

The standard goes on to prescribe plant operating procedures and equipment, work practices and procedures for preventing exposure of employees to the 14 chemicals found to cause cancer, together with a timetable of effective dates.2 In promulgating the standard, OSHA stated that as soon as possible a draft environmental impact statement would be filed with the Council on Environmental Quality and that a copy of the proposed rules would be sent to the Environmental Protection Agency for its comments under section 309 of the Clean Air Act, 42 U.S.C. Sec. 1857 et seq. A draft environmental impact statement prepared by OSHA was received by the Council on Environmental Quality on May 21, 1973.

On July 27, 1973, OSHA issued a revision of the Emergency Temporary Standard, 38 Fed.Reg. 20074-20076, which became effective July 30, 1973. This revision was prompted by its receipt of "numerous comments, objections, and recommendations," which caused OSHA to determine "that certain changes in the standard are necessary to tailor the requirements for different types of workplaces and work operations, and to clarify the standard." Most of the changes are clarifications of pre-existing requirements; few substantive changes have been made. The revised standard deals more particularly with different work operations, such as isolated environment operations, closed system operations and open vessel operations, and the different control measures necessary to protect employees engaged in each category of operation are specified. In making these changes, OSHA expressly reconfirmed the "original findings concerning the toxicity of all the fourteen substances listed in the standard, the grave danger resulting from exposure to any of them," and that "the standard as revised is necessary and sufficient to protect employees from the grave danger from exposure to any of the substances."

The Emergency Temporary Standard in question here was appropriately issued if there is substantial evidence in the record to support OSHA's determination (A) that the 14 chemicals listed in the standard are carcinogenic and, therefore, that exposure of employees to them presents a "grave danger from exposure to substances or agents determined to be toxic or physically harmful," and (B) that the emergency standard "is necessary to protect employees from such danger," 29 U.S.C. Sec. 655(c)(1) and (f).3

Turning first to the finding of carcinogenicity, the independent studies of Pliss, Saffiotti and Stula all support the conclusion that DCB causes cancer in rodents (rats and hamsters). Dry Color's criticisms of these studies, while significant, do not necessarily reduce their conclusions to the level of insubstantial evidence. The case for the carcinogenicity of EI, however, is weaker. The Walpole study with rats was carried out almost 20 years ago and that of Innes is unclear as to whether the tumors produced in mice were malignant and specifically cautions against extrapolating from his data as to the carcinogenic effect in rodents to that in man. In addition, when on May 23, 1972, OSHA first wrote to NIOSH requesting all available information on nine allegedly carcinogenic chemicals, EI was not on that list.4 And the petition for a temporary emergency standard that employee petitioners filed with OSHA at the end of 1972 did not include EI, despite the fact that it had been listed in NIOSH's publication in the Federal Register in July 1972.

Nevertheless, assuming for the sake of argument that both DCB and EI cause cancer in rodents, the question remains as to what conclusions can be drawn as to their carcinogenicity in man. Dry Color has pointed out that scientists generally agree that the carcinogenic effect of DCB is produced not by the chemical itself, but by a metabolite of DCB produced by rodents. Yet in the independent studies of Meigns and Troll, DCB was found to be resistant to metabolic change in dogs, which have a metabolic system very similar to man. Aerojet has pointed to reports other than Innes' that also stress that the evidence from animal experiments has not yet established that EI is carcinogenic in man. And all industry petitioners emphasize that during the years of industrial experience with these two chemicals, approximately 60 years with DCB and 30 with EI, not a single case of cancer has been reported in an employee that can be attributed to exposure to them.

For its part, the Department of Labor has argued that the industrial experience is not conclusive because of the small size of the statistical sample and the relatively short time span of the studies relied on by industry petitioners. The Department also has pointed out that while it has not yet been demonstrated that dogs or humans produce harmful metabolites of DCB, neither has it been shown that they do not. The Department admits, however, that there is no reliable data on the carcinogenicity of DCB or EI in humans. Thus, its entire case depends on an extrapolation from the data gathered in rodent experiments, and to justify that extrapolation the Department relies heavily on the April 22, 1970, Report of the Ad Hoc Committee on the Evaluation of Low Levels of Environmental Chemical Carcinogens to the Surgeon General.5 That Report recommends that "[a]ny substance which is shown conclusively to cause tumors in animals should be considered carcinogenic and therefore a potential cancer hazard for man."6 Although this recommendation does not command unanimous agreement,7 it does appear to constitute substantial evidence in support of the Department's position. The problem, however, is that the Report is not part of the record, and there is no evidence in the published notice of May 3, 1973, that this Report was relied on.8

Based on this summary of the evidence, we conclude that the most that can be said is that DCB and EI pose a "potential" cancer hazard to man. Although the danger of cancer is surely "grave," subsection 6(c)(1) of the Act requires a grave danger of exposure to substances "determined to be toxic or physically harmful." The amount and quality of the evidence necessary to provide "substantial" support for such a finding will necessarily vary depending on the facts of each case. OSHA need not wait until workers have actually been harmed to issue an emergency temporary standard. Extrapolation from animal experiments may in appropriate cases be used to establish a sufficient probability of harm to man.9 However, while the Act does not require an absolute certainty as to the deleterious effect of a substance on man, an emergency temporary standard must be supported by evidence that shows more than some possibility that a substance may cause cancer in man.^9a On this record, the evidence supplies no more than some possibility that DCB and EI may cause cancer in man.

From the foregoing analysis, it is apparent that we have doubts-more serious as to EI than to DCB-as to whether there is substantial evidence in the record to support OSHA's finding that DCB and EI are carcinogenic in man and, therefore, subject employees "to grave danger from exposure to substances or agents determined to be toxic or physically harmful . . . ." Nevertheless, it is not necessary to rule definitively on this question at this time. As discussed below, we find that the statement of reasons for the issuance of the Emergency Temporary Standard in question is inadequate, and we are troubled by doubts as to whether all the documents included by OSHA in its certification of the record were actually considered in the course of making its decision to issue the Emergency Temporary Standard. In the absence of a more complete and detailed statement of reasons from OSHA and a record upon which we can confidently rely, we have concluded that a final ruling on the question of whether there is substantial evidence to support the finding that DCB and EI are carcinogenic would be inconsistent with the statutory language and settled principles of judicial review in the field of administrative law. See below at pages 105-106. For this reason, we also find it unnecessary to rule on the claim of industry petitioners that there is not substantial evidence in the record to show that the Emergency Temporary Standard in question was "necessary to protect employees" from the danger found to exist. We must note, however, that in order to provide immediate protection to employees exposed to a grave danger, emergency temporary regulations may necessarily be somewhat general. Some room must be left for further refinement in the subsequent permanent standard. It cannot be expected that every procedure or practice will be strictly necessary as to every substance, type of use, or plant operation. Nevertheless, it is expected that even an emergency temporary standard not overlook those obvious distinctions among the chemicals to be regulated, uses and plant practices that make certain regulations that are appropriate in one category of cases entirely unnecessary in another.

Subsection 6(e) of the Act specifically mandates that "[w]henever the Secretary promulgates any standard, . . . he shall include a statement of the reasons for such action, which shall be published in the Federal Register." 29 U.S.C. Sec. 655(e).10 This requirement is designed to serve several general functions: it provides an internal check on arbitrary agency action by insuring that prior to taking action an agency can clearly articulate the reasons for its decision; it makes possible informed public criticism of a decision by making known its underlying rationale; and it facilitates judicial review of agency action by providing an important part of the record of the decision. Furthermore, the statement-of-reasons requirement of subsection 6(e) complements the requirement of subsection 6(c)(1) that certain factual findings be made by the Secretary of Labor prior to promulgation of an emergency temporary standard. Congress contemplated that emergency temporary standards would be developed and issued without the benefit of ordinary standard-setting procedures involving public comment and open hearings in the interest of permitting rapid action to meet emergencies. But, by the language of subsection 6(e), it clearly refused to eliminate the general requirement of articulation of the reasons for such action as an essential safeguard to emergency temporary standard-setting.

The question before us is whether the statement of reasons included in the preamble to the Emergency Temporary Standard in question is adequate to satisfy subsection 6(e). There are no cases applying this provision of the Act to guide our decision. However, it is an increasingly accepted principle of administrative law, even when there is no statutory requirement for a statement of reasons, that "the basis for . . . decision should appear clearly on the record, not in conclusory terms but in sufficient detail to permit prompt and effective review."11 Furthermore, the informal or expeditious nature of the administrative procedure leading to an action or decision should not induce courts to relent in their demand for an adequate statement of reasons. Thus, in Kennecott Copper Corp. v. Environmental Protection Agency, 149 U.S.App.D.C. 231, 462 F.2d 846 (1972), the court remanded the record to the Director of the Environmental Protection Agency "to supply an implementing statement that will enlighten the court as to the basis on which he reached the . . . [particular] standard [he issued]." Id. at 850.12

The only statement of reasons offered by OSHA and contained in the preamble to the Emergency Temporary Standard consists of the finding that the 14 chemicals listed in the standard are carcinogens and the conclusion, reciting the language of subsection 6(c)(1) of the Act, that the conditions necessary for the issuance of an emergency temporary standard have been met. We find this statement of reasons inadequate. In the context of a voluminous factual record, such a conclusory statement of reasons places too great a burden on interested persons to determine and challenge the basis for the standard, and makes possible in any subsequent judicial review the use of posthoc rationalizations that do not necessarily reflect the reasoning of the agency at the time the standard was issued.

In particular, we find the statement of reasons here insufficient in two respects. First, it fails to set forth the basis for its finding that the 14 chemicals listed in the standard are carcinogens. To satisfy subsection 6(e), the statement of reasons should indicate which data in the record is being principally relied on and why that data suffices to show that the substances covered by the standard are harmful and pose a grave danger of exposure to employees. This could have been accomplished here by a brief statement in the May 3, 1973, notice that certain scientific data (citing the record documents) showed that DCB and EI produced cancer in rodents and supported the conclusion that they were therefore carcinogenic in man. Second, the statement of reasons failed to offer any explanation as to why this particular standard is "necessary to protect the employees from such exposure." We do not mean to say that every procedure must be justified as to every substance, type of use or production technique. But we do read subsection 6(e) as requiring at least a general explanation as to why the procedures prescribed were chosen in light of the recommendations of scientific experts and other governmental bodies, the types of industrial practices with these chemicals, and the alternative kinds of regulations considered by OSHA.

Because of these deficiencies in the statement of reasons, the Emergency Temporary Standard will be vacated and remanded to OSHA as to DCB and EI.13 This court has been asked to review the adequacy of the statement of reasons only with respect to DCB and EI, and hence no relief is requested in the proceedings now before the court on these petitions with respect to the other 12 chemicals listed in the standard. However, if OSHA decides to issue another emergency temporary standard on carcinogenic chemicals dealing with DCB and EI, rather than await issuance of the permanent standard to be made on or before November 3, 1973, it should on remand issue a statement of reasons that satisfies the requirement of subsection 6(e) as to these chemicals. However, we do not contemplate that this decision will interfere with the current proceedings before OSHA to promulgate a permanent standard for carcinogens.

The remaining issues raised in this case may be dealt with briefly. Industry petitioners contend that OSHA violated the National Environmental Policy Act of 1969 ("NEPA"), 42 U.S.C. Sec. 4321 et seq., by failing to prepare an environmental impact statement prior to the promulgation of the standard. Paragraph 102(2)(C) of NEPA, 42 U.S.C. Sec. 4332(2) (C), and the Guidelines of the Council on Environmental Quality on the Preparation of Environmental Impact Statements, 36 Fed.Reg. 7724 (April 23, 1971), require all federal agencies to prepare an environmental impact statement in connection with major federal actions significantly affecting the quality of the human environment and to complete that statement before action is taken. We see no reason to create a general exemption for OSHA from this requirement. The decision in Portland Cement Association v. Ruckelshaus, 158 U.S.App.D.C. -, 486 F.2d 375 (1973), was a carefully limited one, creating a narrow exemption from the requirements of NEPA for the environmentally protective regulatory activities of the Environmental Protective Agency under a statute-the Clean Air Act-which provides effective procedures to accomplish NEPA's purposes. Nevertheless, while the NEPA requirement of an environmental impact statement applies to ordinary standards promulgated under the Occupational Safety and Health Act, an exception should be made for the Emergency Temporary Standard involved in this case.14 The process by which NEPA statements are produced and circulated is a lengthy one. To require its completion before the promulgation of an emergency temporary standard would impair the purpose of subsection 6(c) to provide speedy protection from grave dangers to the health of employees. No doubt any exemption of emergency temporary standards under this Act from the requirement of an environmental impact statement involves some sacrifice of the policy behind NEPA. However, that sacrifice is mitigated somewhat by the fact that an emergency temporary standard must be replaced within six months by a permanent standard, for which a completed impact statement is required, and it is, in any case, justified by the need to accommodate the provisions of NEPA with those of the Occupational Safety and Health Act. We therefore have concluded that where, as here, OSHA begins the process contemplated by NEPA by issuing a draft environmental impact statement within a reasonable time after issuing an emergency temporary standard, the requirements of NEPA are sufficiently satisfied.15

Industry petitioners have also moved to strike certain documents certified in the record by OSHA, charging that the Department of Labor never had before it and did not consider many of the items upon which they and their brief rely to support the standard. The Department of Labor replied in oral argument that since NIOSH (although an institute of HEW) was by statute the scientific advisor to OSHA, OSHA could rely on summaries and recommendations from NIOSH and certify as part of the record any item considered by NIOSH.16 In view of our remand of the standard to OSHA, we need not rule on the motions to strike and amend the certification of the record,17 since on remand OSHA will have to create a new record for any standard governing the use of DCB and EI. However, it would be the better practice, in the absence of unusual circumstances, for OSHA, in certifying the record on any petition filed with this court, to designate specifically any items certified which have not been read by it prior to its publication of standards, such as those where it relies on summaries prepared by NIOSH.

Finally, employee petitioners argue that the Emergency Temporary Standard falls short of preventing the impairment of worker health which the Act requires and seek to have this court order the implementation of a use permit system that would achieve zero exposure. Although there appears to be substantial evidence to support the adequacy of the procedures prescribed by the standard to protect workers' health,18 it is unnecessary to decide this issue at this time in light of our disposition of the case.

The Emergency Temporary Standard will be vacated and remanded to OSHA insofar as it concerns DCB and EI, and the petition in No. 73-1383 will be denied without prejudice in accordance with the foregoing opinion. Petitioners' motion in No. 73-1361 to strike certain items from the certification of record and respondents' motion in all cases for leave to amend the certification of record will be denied as moot.

McLAUGHLIN, Circuit Judge (dissenting).

In 1970, the Occupational Health and Safety legislation became law. The Congress of the United States declared it to be their purpose and policy to assure, so far as possible, every working man and woman in the nation safe and healthful working conditions. To achieve this goal, Congress empowered the United States Secretary of Labor to implement regulatory standards.

On May 3, 1973, pursuant to the congressional mandate set forth in the Occupational Safety and Health Act, 29 U.S.C. Secs. 651-678 (the Act), the Secretary of Labor promulgated an Emergency Temporary Standard regulating employee exposure to fourteen (14) substances which the Secretary determined to be carcinogenic (cancer-causing). Several manufacturers and users of two (2) of these substances, dichlorobenzedine (DCB) and ethyleneimine (EI), petitioned this court to vacate the Emergency Temporary Standard in so far as it applied to these substances.

These petitioners contend that there was no substantial evidence in the Secretary's findings to show that employees are subjected to grave danger from exposure to substances or agents determined to be toxic or physically harmful as required by 29 U.S.C. Sec. 655(c)(1) and that the Secretary's reasons for issuing the Emergency Temporary Standard were inadequate. The majority opinion agrees with these propositions and thereby vacates the Emergency Temporary Standard's provisions relative to DCB and EI. I must respectfully dissent.

The record before us contains the results of various scientific experiments which stand for the proposition that ingestion of DCB and EI may cause cancerous tumors to develop in the anatomical structures of mice and rats. These findings led the Secretary to conclude that DCB and EI could possibly produce similar results in the human anatomy and he therefore issued the Emergency Temporary Standard. The Act requires that, upon judicial review of any standard promulgated pursuant to its provisions, "determinations of the Secretary shall be conclusive if supported by substantial evidence in the record considered as a whole." 29 U.S.C. Sec. 655(f). Stated another way, there must be substantial evidence in the record to justify setting aside the Standard. Here, the Secretary has taken affirmative action to protect and preserve human life. Petitioners have asked this court to invalidate that action. Obviously the petitioners must bear the burden of proof, i. e. they must be able to show by substantial evidence that their conduct does not jeopardize the health of their workers. They, however, merely assert that since there is inconclusive scientific data proving DCB and EI to be carcinogenic in man, the Standard should be set aside. Those assertions cannot be reasonably accepted as substantial evidence and clearly the Standard should remain.

Subsection (c)(1)(A) of Section 655 of the Act provides that "The Secretary shall provide * * * for an emergency temporary standard to take immediate effect upon publication in the Federal Register if he determines (A) that employees are exposed to grave danger from exposure to substances or agents determined to be toxic or physically harmful * * *." The majority takes the position that the phrase "determined to be toxic or physically harmful" requires the Secretary to establish, by substantial evidence, that a substance is potentially harmful and that a showing of some possibility of harm would not justify the issuance of an Emergency Temporary Standard. To decide whether Congress empowered the Secretary to regulate a substance or agent because it is possibly harmful, probably harmful, or actually harmful is to engage in a futile exercise in semantics. My reading of the Act and its legislative history makes one thing abundantly clear: Congress intended to protect the health and safety of the American worker. Therefore, I would hold that even a scintilla of evidence which tends to prove a substance carcinogenic in man or animal justifies the issuance of an Emergency Temporary Standard.

Subsection (e) of Section 655 of the Act provides that "Whenever the Secretary promulgates any standard * * * under this chapter, he shall include a statement of the reasons for such action, which shall be published in the Federal Register." The majority would have it that the Secretary's statement of reasons in the Emergency Temporary Standard published in the May 3, 1973 issue of the Federal Register is inadequate. I cannot possibly agree. The unmistakable intent of subsection (e) is to require the Secretary to inform the public of the reasons for his actions. On May 3, 1973, users and manufacturers of DCB and EI were put on notice that their products were to be regulated by the federal government. They were not left to second guess the Secretary's reasons or motives for taking such action. They were informed on said date that the Secretary had determined the substances to be carcinogenic. Nothing more should be required. If we were to try and force the Secretary to issue an exhaustive statement explaining and supporting his motives, the Emergency Temporary Standard would become an ineffective mechanism. All of the relevant data utilized by the Secretary in making his determination was available to interested parties upon their request. The petitioners had notice of the Secretary's reasons for the issuance of the Emergency Temporary Standard and those same petitioners had access to the scientific data which was the foundation for the Secretary's decision. The Secretary's action promptly and properly fulfilled the statutory requirement, it should not be disturbed.

For the reasons set forth above, the order of the Assistant Secretary of Labor for Occupational Safety and Health as published in the May 3, 1973 issue of the Federal Register, 38 Fed.Reg. 10929 should be upheld.