This case is before the court upon petition to review an order of the Assistant Secretary of Labor for Occupational Safety and Health, published on January 29, 1974, 39 Fed.Reg. 3756, issuing permanent occupational health standards which regulate, inter alia, employee exposure to solid or liquid mixtures containing one percent or more of ethyleneimine ('EI). See 29 C.F.R. 1910.93lc.1 This court has jurisdiction to review and set aside these standards pursuant to 6(f) of the Occupational Safety and Health Act of 1970, 29 U.S.C. 655(f)('Act').

The standards in question are part of a group of permanent standards simultaneously issued pursuant of 6(b) of the Act. 29 U.S.C. 655(b). They were derived from similar emergency temporary standards issued on May 3, 1973, 38 Fed.Reg. 10929. This court has previously vacated and remanded to the Occupational Safety and Health Administration ('OSHA') the temporary standards in so for as they applied to EI and one other chemical. Dry Color Manufacturers' Association, Inc. v. Department of Labor, 486 F.2d 98 (C.A. 3, 1973). We must now consider the permanent standards relating to EI.

On review the petitioners request that the court set aside OSHA's standards applicable to EI. They challenge the validity of the standards on three principal grounds. They maintain that there is not substantial evidence to show that EI is carcinogenic; that the record does not support the particular EI standards; and that the Secretary failed to make sufficient findings of fact or to provide sufficient statements of reasons for the standards.

We deal first with the petitioners' last contention. Subsection 6(e) of the Act provides:

'Whenever 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. 655(e).

As this court has previously stated, to satisfy subsection 6(e) the statement of reasons should indicate which data in the record is being relied upon, why that data shows the substances regulated are harmful, and why the particular standards were chosen. Dry Color Manufacturers' Assn., Inc. v. Department of Labor, 486 F.2d at 106. In that case a temporary emergency standard for the control of EI was invalidated by this court because the Secretary, in promulgating it, had not provided an adequate statement of reasons.

In the instant case, the statement of reasons offered by OSHA, published January 29, 1974, 39 Fed. Reg. 3756, consisted in part of the finding that EI was potentially carcinogenic to humans. In addition, according to OSHA, the case for EI's carcinogenicity rested on the extrapolation to humans of the findings of two animal studies. The Secretary justifies such extrapolation on the basis of the April 22, 1970 Report of the Ad Hoc Committee on the Evaluation of Low Levels of Environmental Chemical Carcinogens to the Surgeon General. We have carefully examined the statements as well as those relating to the adopted standards themselves and conclude that they satisfy the requisites of subsection 6(e). We turn now to a consideration of the evidence in the record.

Subsection 6(f) directs courts reviewing a standard that 'the determinations of the Secretary shall be conclusive if supported by substantial evidence in the record considered as a whole.' 29 U.S.C. 655(f). See Industrial Union Department, AFL-CIO v. Hodgson, 499 F.2d 467 (D.C.Cir. 1974); Associated Industries of New York State, Inc. v. United States Department of Labor, 487 F.2d 342 (C.A. 2, 1973). We find it extraordinarily difficult to apply this standard in this case. Section 6(a) of the Act directs the Secretary of Labor to promulgate by rule an occupational safety or health 'standard which assures the greatest protection of the safety or health of the affected employees.' 29 U.S.C. 655(a). Section 6(b)(5) of the Act, 29 U.S.C. 655(b)(5) directs that

'The Secretary, in promulgating standards dealing with toxic materials or harmful physical agents . . . shall set the standard which most adequately assures, to the extent feasible, on the basis of the best available evidence, that no employee will suffer material impairment of health or functional capacity even if such employee has regular exposure to the hazard dealt with by such standard for the period of his working life.'

Because the Dry Color case turned on noncompliance with 6(e), it did not present us with an occasion to discuss the standard for review of informal rule making by the Secretary under the Act. However, the instant case requires that we do so.

It is made clear in 6(b) that 6(a) standards may be adopted by notice and comment rule making analogous to that provided for by 4 of the Administrative Procedure Act, 5 U.S.C. 553. Yet the quoted language of 6(f) appears to limit judicial review to the scope of review which ordinarily applies to administrative determinations resulting from formal adjudications. See 5 U.S.C. 706(2)(E). Compounding the confusion, 10 of the Act, 29 U.S.C. 659, provides for adjudications, 11, 29 U.S.C. 660, provides for judicial review of adjudications, and the wording of the standard for review of adjudicatory findings set forth in 11(a) is virtually the same as that for review of informal rule making set forth in 6(f). See Brennan v. Occupational Safety & Health Review Commission (Hanovia Lamp), 502 F.2d 946 (C.A.3, filed-- 1974). It is clear that application of the substantial evidence test to informal rule making was a deliberate legislative compromise between the House and Senate versions of the bill which became law. See the legislative history in Judge Friendly's opinion in Associated Industries of New York State, Inc. v. United States Department of Labor, 487 F.2d 342, 348-349 (C.A. 2, 1973).

How a reviewing court applies the substantial evidence test in reviewing quasi-legislative informal rule making is an intriguing problem which is just beginning to generate what will prove to be, we suspect, an extensive literature. In Associated Industries of New York State, Inc. v. United States Department of Labor, supra, Judge Friendly suggests that the difference between the substantial evidence test and the more traditional abuse of discretion test in the context of informal rule making may be largely semantic. 487 F.2d at 349-350. In Florida Peach Growers Association, Inc. v. United States Department of Labor, 489 F.2d 120, 129 (C.A. 5, 1974), Judge Roney writes:

'It seems clear that even with the required substantial evidence test, our review basically must determine whether the Secretary carried out his essentially legislative task in a manner reasonable under the state of the record before him.'

In Industrial Union Department, AFL-CIO v. Hodgson, 499 F.2d 467 (D.C.Cir. 1974), Judge McGowan distinguished those rule making issues which were susceptible of resolution on a factual record, as to which the court would review for substantial evidence and those issues which by their nature were not capable of such resolution but represented policy judgments. As to the latter, Judge McGowan suggested that the Secretary's policy judgment would be affirmed if he indicated that factual certainty was not possible and identified rationally the controlling considerations. But cf. Environmental Defense Fund, Inc. v. Environmental Protection Agency, 489 F.2d 1247 (D.C.Cir. 1973) and International Harvester Co. v. Ruckelshaus, 155 U.S.App.D.C. 411, 478 F.2d 615 (1973).

This case is a good illustration of the difficulty of attempting to measure a legislative policy decision against a factual yardstick. OSHA's position with respect to EI is bottomed on an extrapolation from data gathered in two animal studies. The first is a study carried out in 1954 by Walpole which concluded that EI may be regarded as carcinogenic in rats and mice.2 The Innes study, the second upon which OSHA relied, concluded that mice under experimental exposure to EI developed tumors with 'malignant potentiality.'3 The extrapolation from the data gathered in these rodent experiments to humans is justified by the Report of the Ad Hoc Committee on the Evaluation of Law Levels of Environmental Chemical Carcinogens to the Surgeon General. This report states:

'Any substance which is shown conclusively to cause tumors in animals should be considered carcinogenic and therefore a potential cancer hazard for man.'

If the issue to be reviewed were merely whether EI was carcinogenic in rats and mice, we believe that we could point to the Walpole and Innes studies and safely conclude that the Secretary's determination of animal carcinogenicity was supported by substantial evidence. But the extrapolation of that determination from animals to humans is not really a factual matter. The Secretary's 6(e) statement puts it thus:

'Extrapolation of results obtained by animal experimentation is alleged to be vitiated by several considerations: (a) That certain cancers are specific only to some species; (b) that the conditions of animal experiments are out of proportion to, and not consistent with, conditions prevailing in industrial exposure; and (c) that no cancers have yet been detected in humans exposed to the substances.

'The objections raise the much broader issue of human exposure to a chemical which is only known to have caused cancers in experimental animals.

'We think it improper to afford less protection to workers when exposed to substances found to be carcinogenic only in experimental animals. Once the carcinogenicity of a substance has been demonstrated in animal experiments, the practical regulatory alternatives are to consider them either non-carcinogenic or corcinogenic to humans, until evidence to the contrary is produced. The first alternative would logically require, not relaxed controls on exposure, but exclusion from regulation. The other alternative logically leads to the treatment of a substance as if it was known to be carcinogenic in man.

'We agree with the director of NIOSH and the report of the Ad Hoc Committee on the Evaluation of Low Levels of Environmental Chemical Carcinogens to the Surgeon General, U.S. Public Health Service, April 22, 1970 that the second alternative is the responsible and correct one.' (Appendix at 845-46.)

It seems to us that what the Secretary has done in extrapolating from animal studies to humans is to make a legal rather than a factual determination. He has said in effect that if carcinogenicity in two animal species is established, as a matter of law 6(a) and 6(b)(5) require that they be treated as carcinogenic in man. This is in the nature of a recommendation for prudent legislative action.

This raises the problem of our role under 6(f) in reviewing the Secretary's nonfactual policy decisions. The deliberate congressional choice of the substantial evidence standard for review might be interpreted to exclude us from review of any but factual determinations and to require that we afford absolute deference to the Secretary's interpretation of the statute. However, no court that has considered the scope of judicial review has so read 6(f). By one formulation or another each seems to have found room under 6(f) for judicial review for consistency between the Secretary's rule and the statutory language and purpose. See Industrial Union Department, AFL-CIO v. Hodgson, supra; cf. Florida Peach Growers Association, Inc. v. United States Department of Labor, supra; Dry Color Manufacturers' Assn., Inc., v. Department of Labor, supra; Associated Industries of New York State, Inc. v. United States Department of Labor, supra. Undoubtedly, this approach is sound, for the last sentence of 6(f) cannot be read in a vacuum. Meaning must be given to all of 6. The last sentence of 6(f) must be read together with the requirement in 6(e) that the Secretary 'include a statement of the reasons for such action, which shall be published in the Federal Register.' He is not directed to file findings of fact. His reasons for action may include policy determinations as well as factual findings. The requirement that he state reasons strongly suggests that those reasons, whether policy judgments or factual findings, may be subjected to judicial review. Moreover, there are different legal criteria for the adoption of emergency temporary standards than for the adoption of permanent standards, but 6(f) provides for judicial review of both. Section 6(c)(1) authorizes adoption of emergency temporary standards when the Secretary '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. 655(c)(1). Permanent standards must assure 'the greatest protection of the safety or health of the affected employees,' 29 U.S.C. 655(a), and, with respect to toxic substances, must 'most adequately assures, to the extent feasible, on the basis of the best available evidence, that no employee will suffer material impairment of health or functional capacity even if such employee has regular exposure to the hazard dealt with by such standard for the period of his working life.' 29 U.S.C. 655(b)(5). The very difference in the legal criteria suggests that the court in reviewing pursuant to 6(f) must at a minimum determine if the Secretary applied the correct ones, and certainly, as we held in Dry Color Manufacturers' Assn., Inc. v. Department of Labor, supra, judicial review under 6(f) encompasses our evaluation of the procedural requirements of 6 even though our interpretation differs from that of the Secretary.

It seems, then, that judicial review of a 6 standard properly includes at least the following:

(1) determining whether the Secretary's notice of proposed rule making adequately informed interested persons of the action taken;

(2) determining whether the Secretary's promulgation adequately sets forth reasons for his action;

(3) determining whether the statement of reasons reflects consideration of factors relevant under the statute;

(4) determining whether presently available alternatives were at least considered; and

(5) if the Secretary's determination is based in whole or in part on factual matters subject to evidentiary development, whether substantial evidence in the record as a whole supports the determination.

In the context of this case, all five steps are involved.

We note that there was a lack of notice regarding OSHA's plans to make special provisions regarding EI's use in research laboratories. The Secretary's notice did not advise that the agency planned to make special provisions regarding EI use in laboratories. Because of this, we hold that to the extent that the standards in question relate to research laboratories, they will be vacated and remanded to OSHA. Further, as we concluded above, the statements of reasons reflect considerations relevant under 6(a) and 6(b)(5) and are therefore sufficient under subsection 6(e). Lastly, we hold that there does exist substantial evidence in the record as a whole to support the Secretary's finding that EI is carcinogenic in rats and mice and in the absence of evidence of carcinogenicity in humans, the Secretary properly weighed the only available alternatives. We, therefore, reject the petitioners' challenge to the EI standards except in so far as they relate to laboratories as outlined above.

The cause will be remanded to OSHA with directions to proceed in conformity with this opinion.