Opinion
Nos. 10765, 10833.
Argued October 10, 1952.
Decided December 22, 1952. Rehearing Denied January 14, 1953. Writ of Certiorari Denied April 6, 1953. See 73 S.Ct. 783.
Oscar Cox, Washington, D.C. (C.C. Gammons, Kenneth E. Mulford and Roger R. Horton, Wilmington, Del., Lloyd N. Cutler, Louis F. Oberdorfer, Herbert L. Packer and Donald F. Turner, Washington, D.C., on the brief), for Atlas Powder Co.
Michael F. Markel, Washington, D.C. (Wayne K. Hill, Washington, D.C., on the brief), for Glyco Products Co.
William W. Goodrich, Assistant General Counsel, Federal Security Agency, Washington, D.C. (Charles B. Murray, Asst. Atty. Gen., Vincent A. Kleinfeld, Attorney, Department of Justice, Washington, D.C., Selma M. Levine, Attorney, Federal Security Agency, Washington, D.C., on the brief), for respondent.
Before McLAUGHLIN, KALODNER and HASTIE, Circuit Judges.
Atlas Powder Company and Glyco Products Company, Inc., each suing as a "person who will be adversely affected by" an order of the Federal Security Administrator establishing, pursuant to Section 401 of the Food, Drug, and Cosmetic Act, 52 Stat. 1046, 21 U.S.C.A. § 341, a standard of identity for bread, have filed in this court separate petitions for review of that order as authorized by Section 701(f) of the Food, Drug, and Cosmetic Act, 52 Stat. 1055, 21 U.S.C.A. § 371(f), and Section 10 of the Administrative Procedure Act, 60 Stat. 243, 5 U.S.C.A. § 1009.
Both petitioners manufacture, though by different processes, a chemical descriptively denominated polyoxyethylene monostearate. Glyco markets its product under the brand names "Sta-Soft" and "S-541". The principal Atlas product is called "Myrj 45". These products are used by commercial bakers to make bread softer for a longer time after baking than it would be without some such additive.
When, pursuant to Section 401 of the Food, Drug and Cosmetic Act, the Administrator convened a proceeding to establish a definition and standard of identity for bakery products, Atlas and Glyco appeared and asked, and attempted to prove, that the standard of identity for bread should include their softeners as optional ingredients. The Administrator refused thus to amplify the standard and petitioners have sought judicial review.
I
We consider first certain legal contentions made by Glyco. Although this party asked the Administrator to include its products in the standard description of bread, it now makes a principal argument that the matter of including such an ingredient as "Sta-Soft" in bread is not within the purview of Section 401 at all and can not be controlled by administrative action under that section. The propriety of such a challenging of the power of the Administrator over a particular subject matter by the very party who has asked that the Administrator take affirmative action with reference to that subject matter may be questionable. However, the issue before us is the validity and effect of an administrative regulation of nation-wide application to the food industry and of general importance to consumers. In the light of all of the interests involved, we think possible inconsistencies in the position of the petitioner should not prevent us from considering and adjudicating at petitioner's request the question whether the use of chemical softener in bread is a subject matter properly covered and controlled by a standard of identity for bread promulgated under Section 401.
The only appropriate procedure in petitioner's view would be action to bar bread containing "Sta-Soft" from commerce as adulterated under Section 402 of the Act or as containing deleterious substances under Section 406, if either defect could be established.
Section 401 reads as follows:
"Whenever in the judgment of the Secretary such action will promote honesty and fair dealing in the interest of consumers, he shall promulgate regulations fixing and establishing for any food, under its common or usual name so far as practicable, a reasonable definition and standard of identity, a reasonable standard of quality, and/or reasonable standards of fill of container: * * * In prescribing a definition and standard of identity for any food or class of food in which optional ingredients are permitted, the Secretary shall, for the purpose of promoting honesty and fair dealing in the interest of consumers, designate the optional ingredients which shall be named on the label. * * *"
On its face this language seems to empower the Administrator, in his effort to fix a standard of identity for a fabricated food, to consider every ingredient proposed for inclusion in the standard description and to include or exclude each as, after appropriate investigation, may appear reasonable. Beyond this, flexibility in action upon such proposals is afforded by the statutory provision for the inclusion of particular ingredients as optional components in the standard of identity. And we have it on highest authority that "The statutory purpose to fix a definition of identity of an article of food sold under its common or usual name would be defeated if producers were free to add ingredients, however wholesome, which are not within the definition" either as required or permissive components. See Federal Security Administrator v. Quaker Oats Co., 1943, 318 U.S. 218, 232, 63 S.Ct. 589, 597, 87 L. Ed. 724.
It seems to be petitioner's contention here that the foregoing generalization does not apply to the case at hand. It is argued that a standard of identity for a named class of food is properly concerned only with the inclusion and proportions of ingredients which are shown to be definitive or characteristic of the species. The accomplishment of standardization is said not to be affected in any significant way by the inclusion or exclusion of less consequential ingredients which do not change the essential description or properties of the food. In this case the supposed limitation upon the function and effect of statutory standardization is reflected in a claim that though the administrative order standardizing bread does not approve the addition of "Sta-Soft", the addition of that ingredient involves no departure from the standard because the food is still bread and essentially the same kind of bread whether the softening chemical is added or left out.
But this rationalization is self-defeating if only because it covers too much. Under it the addition of vitamins at will would have to be permitted under any proper statutory standard of identity established under Section 401 for a cereal. But in the Quaker Oats case, supra, the Supreme Court has made it clear that this is not true. So too, the addition of a harmless preservative to a manufactured food, for example, a bit of sodium benzoate to tomato catsup, would have to be permitted under any standard of identity for catsup. The contrary was held by the Court of Appeals for the Second Circuit in Libby, McNeil and Libby v. United States, 1945, 148 F.2d 71. Similarly, the Court of Appeals for the Eighth Circuit has viewed a standard for oleomargarine as properly controlling the addition of artificial coloring, artificial flavoring, vitamins and preservatives. Land O'Lakes Creameries, Inc. v. McNutt, 1943, 132 F.2d 653. Beyond inconsistency with these cases, it seems to us that petitioner's reasoning lends itself to the reductio ad absurdum type of refutation. For logically petitioner would not let the Administrator challenge even the addition of a bit of arsenic to bread as inconsistent with any proper standard of identity or subject to control under the procedure for food standardization prescribed in Section 401.
We conceive of no workable way of determining what ingredients, in petitioner's conception, are relevant to standardization of a food and what ingredients in what amounts can be added at will to a food as beyond the reach of standardization achieved under Section 401. We believe that Congress has not complicated the practical administration of standards by any distinction so pointless and impractical. Rather we are convinced that all ingredients proposed for use in the fabrication of a food are comprehended by the administrative process of standardization. Whether the standard evolved shall be strictly particularized and invariant or generalized and comprehensive of many different compositions of matter will depend upon proper administrative evaluation of the circumstances of each case.
But if the power of the Administrator under Section 401 is thus broadly construed, the petitioner next argues, then the statute itself contains a delegation of power so deficient in guides to and limitations upon its exercise as to invalidate the legislation and any administrative action thereunder. Petitioner is concerned particularly that the statutory scheme for standardization permits administrative approval of optional ingredients which a manufacturer may include or omit at will and still comply with the standard. This is said to show that administrative discretion in the designation of such optional ingredients is exercised at will without meaningful legislative control and guidance. At the same time, however, it seems to be conceded that for the administrative determination of required ingredients, Section 401 affords adequate legislative guidance.
In order to reach a "reasonable" conclusion upon the matter of required ingredients and to determine what action "will promote honesty and fair dealing in the interest of consumers", as the statute prescribes, the Administrator normally will consider food chemistry, commercial practice, consumer reaction and other relevant factors. But the very data measured by the same legislative yardsticks that lead to rigid standardization for one food may as clearly show the propriety of flexibility in the form of permissive variations of composition in another. And whether variation is in percentage of ingredients or substitution of ingredients or option to include or omit a particular ingredient at will, the result is achieved by the same administrative process and governed by the same criteria of reasonableness and protection of consumer interests.
Moreover, we cannot properly consider this a matter of first impression. If Section 401 were fatally deficient in its attempted regulation of optional ingredients, administrative standardization under Section 401 could not have been effective to deprive the petitioner in the Quaker Oats case of the privilege of adding vitamins at will to its product. And while it does not appear that this issue was raised, it is not to be assumed that the Supreme Court would have overlooked or disregarded any such fundamental inadequacy in controlling legislation. Both reason and authority sustain the validity of Section 401 and the construction which makes it effective to control the use of softening agents in bread.
II
But even if the statute is valid and applicable to this case, both petitioners contend that analysis of the record will show that the Administrator acted unreasonably and unfairly in excluding their products from the bread standard.
The Administrator has been trying for more than ten years to arrive at a definition and standard of identity for bread and bakery products. To that end he first duly convened a proceeding and held hearings in 1941 and 1943, but war conditions caused adjournment before the inquiry could be completed. New hearings began late in 1948 and continued for ten months. A 17,000 page record was compiled. On this record the Administrator prepared and in August 1950 announced tentative findings and the order he proposed to issue fixing a definition and standard of identity for bread. Among many other things he found that the type of emulsifier recommended as an ingredient of bread by parties including the petitioners had substantial utility only as a softening agent and in that regard tended to deceive consumers as to the age of bread. He found further that the use of such chemicals in bread was new and that their safety had not been established. In these circumstances he refused to designate or approve emulsifiers of this group as optional ingredients within the standard of identity.
Atlas and Glyco separately filed exceptions to the proposed adverse findings and order, urging particularly that unlawful and invalidating discrimination against their products appeared on the face of the record in that the order both disapproved their products and sanctioned the use of other surface active agents which competed with their products and were not shown to be any safer or less deceptive. Atlas also asked the Administrator to reopen the record, stating among other grounds its desire to present evidence of investigations and experiments made during a two year period since the hearings had been concluded in 1949. This new evidence was said to establish affirmatively the safety of its products. This motion was denied. The Administrator modified his findings in some details and entered an order still disapproving the Atlas and Glyco type softeners and approving the rival products as optional ingredients in the bread standard. These petitions for review followed.
The issue of safety and the evidence relevant thereto will be considered first.
In the words of the chief chemist of Atlas, the recommendation urged upon the Administrator was that he approve as optional ingredients in the proposed standard for bread "Emulsifiers or mixtures thereof, selected from the following:
"(a) Partial esters of fat forming fatty acids and mono and dihydrides of sorbitol, and polyoxyethylene derivatives thereof.
"(b) Long chain polyoxyethylene esters of fat forming fatty acids."
Many products would satisfy these descriptions, but the one of principal commercial importance in bread making, produced by different processes by both Atlas and Glyco, is called polyoxyethylene monostearate and sold under trade names "Myrj 45", "Sta-Soft" and "S-541". Most of the evidence concerns this product, although formally the proponents were urging the approval of a variety of esters which for convenience, with no effort at technical accuracy in nomenclature, we will call the polyoxides.
Another type of surface active agent which has some softening effect is the group of monoglycerides and diglycerides of fat forming fatty acids, which we will call the glycerides. It is the approval of the use of the glycerides in limited proportion in bread of which petitioners complain as constituting arbitrary discrimination against the polyoxides.
The Administrator found the use of polyoxides in food, and particularly in bread, to be new. Polyoxyethylene monostearate had first been offered to the baking trade only about eighteen months before the convening of the 1948 hearings. This product and its prototypes do not occur in nature and are not derived from biological sources. They are complex reaction products rather than simple unvarying substances. Polyoxyethylene monostearate itself is an esterfication product of stearic acid reacted with ethylene oxide. In analysis, and apparently in the course of human digestion, it breaks down into a polyoxyethylene glycol fraction and a fatty acid fraction. It is chemically closely related to certain poisonous glycols. Indeed, analysis has shown very small amounts of such toxic material in polyoxyethylene monostearate although the evidence at hand does not make it seem likely that these small quantities are hurtful to human beings.
In these circumstances, when polyoxyethylene monostearate first appeared on the market in 1947, the American Institute of Baking, described in this record as "the scientific and educational organization of the Baking Industry", issued a bulletin warning the trade that the composition of the new commercial offering suggested that it might be injurious and warning against its use in bread until its safety should be fully and affirmatively established. Later, the Institute appeared in this administrative proceeding and took the same position. The Administrator was also confronted with resolutions of the Council of Foods and Nutrition of the American Medical Association and of the Food and Nutrition Board of the National Research Council, both viewing with concern the possible toxicity of the polyoxides and urging that they not be approved for use in bread until their safety should be established affirmatively. This position was summarized in the present hearing by the Chairman of the Committee on Cereals of the Food and Nutrition Board who testified "that the Food and Nutrition Board endorses the rigorous exclusion so far as possible of additions to bread and other foodstuffs of ingredients of nonbiological origin which have not been fully demonstrated to be nontoxic, wholesome, and not adverse to the nutritive quality and economy of the foodstuff."
In addition, the Administrator received detailed testimony of chemists and other experts concerning a considerable amount of scientific investigation, including the experimental feeding of animals, designed to test the safety of polyoxyethylene monostearate. He found this evidence to be indecisive. We have examined the data and can not see how any more positive conclusion could fairly have been reached.
On the whole record the Administrator properly concluded that the safety of the questioned products had not been established, although he was unable to conclude that they were deleterious. But, even conceding this, the petitioners take the position that it is unreasonable and unfair to permit an indecisive conclusion that safety has not been proved to militate against the approval of their products. Only an affirmative finding that the products are injurious, it is argued, may count against them.
But, as has already been pointed out in the foregoing review of evidence, this negative finding was given effect in the light of a number of significant facts. The lack of long term experience with daily human consumption of these products, their complex and not wholly clear chemistry, the alarm with which that chemistry was viewed by competent and disinterested scientific groups and the actual presence of toxic material even in very small quantities all combined to justify their rejection as an item of daily human diet until satisfactory affirmative proof of safety should be forthcoming.
But even if this reasoning is proper, the petitioners argue that the Administrator made an arbitrary distinction in disapproving their products and at the same time sanctioning the use of the competing emulsifiers in the glyceride group. However, certain differences between the two types of emulsifiers appear clearly in the record.
Reference has already been made to the chemistry of the polyoxides. In contrast, this record indicates that mono and diglycerides are substances of a type which occurs in natural fats, indeed, that they appear as intermediates in the digestion of natural fats. Beyond this it appears that the glycerides have been used in food, particularly in shortening, for some twenty years. Neither their chemistry nor experience in their use has suggested that they may be deleterious. If their safety has not been proved conclusively, no reason for apprehension has appeared as it did in the case of the polyoxides, and there are substantially greater affirmative indications of safety than in the case of the polyoxides. In these circumstances, we can not characterize as arbitrary the administrative action which permits the use of one product forthwith but rejects the other until there shall be more convincing proof of safety.
The second principal finding of the Administrator adverse to the polyoxides was that their use tended to deceive consumers with reference to the age of bread. It is not disputed that the most important function of these products in bread is to make the loaf softer longer. It is as softeners that "Myrj 45" and "Sta-Soft" are marketed. And though the manufacturers now claim additional virtues for their products, their original claims were limited to a softening effect. The Administrator found no other significant function or effect of the polyoxides. We can not agree with the petitioner that this finding was clearly wrong. We view it as a permissible resolution of conflicting evidence concerning the function and effect of these ingredients. The Administrator found further and on substantial evidence that consumers prefer newly baked bread and tend to test such "freshness" by the feeling of softness. It is upon these findings that the Administrator predicates his conclusion as to deception.
But is this "deception" objectionable? Clearly so if of two equally soft loaves the more recently baked without polyoxyethylene monostearate is superior to the older which contains that emulsifier. But in thousands of pages of testimony we find very little that is helpful on this point. There is no evidence that the "fresher" loaf is more nutritious. The record contains various indecisive statements about apparent differences in taste and the bases thereof, both physiological and psychological. There is some indication that the more recently baked loaf crumbles less readily than the older and equally soft loaf which contains polyoxyethylene monostearate. Opinions are expressed among the trade that bread should be sold to consumers before it is two days old. But on the whole record it is far from clear that consumers prefer a "fresher" loaf for any quality not present in older bread made equally soft by petitioner's product.
Yet, for whatever reason, the evidence indicates that to many consumers time elapsed since baking is an important factor in determining the acceptability of bread. Even if they are mistaken as to the significance of the factor or if properties which go with it can be produced or preserved by an additive, we think an administrative officer fixing reasonable standards of identity may properly question whether it is fair and reasonable to induce consumers to take something which may be just as good as a newly baked loaf by creating in their minds the erroneous impression that the loaf is newly baked. And when the additive which accomplishes this deception is something new in food which has no other attending and compensating virtues, we think there is room for a reasonable man in the Administrator's position to conclude that the deception, even if not notably harmful, militates against the approval of the ingredient.
Here too, petitioners urge that the glycerides are equally deceptive. But the Administrator found as a fact that the glycerides have a "shortening" effect in bread beyond any softening effect. Indeed, their use in and as shortening antedates the commercial development and exploitation of substances as softening agents. While the description and exposition of the "shortening" effect of the glycerides in the record is less than one might desire for clarity, we think that at least there is substantial evidence that these additives make bread "tenderer to the bite". Moreover, in recognition of this distinction, the Administrator permitted the use of glycerides only at a stated level which in his judgment would preserve the advantages of shortening while minimizing the dangers of deceptive softening. Petitioners insist that the evidence relied upon to support this distinction does not do so when correctly analyzed. While this interpretation is arguable, we are not persuaded that the Administrator acted arbitrarily either in his contrary interpretation of the evidence or in his determination of an unobjectionable proportion of glycerides.
This still does not answer the petitioners' contention that their products should have been approved at a comparable nondeceptive level. However, if, as the Administrator reasonably found, the polyoxides served no significant purpose except softening, there is no reason to approve or even to seek to discover that proportion of polyoxyethylene monostearate which will have no significant softening effect.
In summary, we are satisfied that the findings of the Administrator and the action he has taken to exclude some surface active agents from the bread standard while including others are supported by substantial evidence. Far from being arbitrary, they reflect a carefully reasoned effort to do what Congress has authorized in the way Congress has indicated. Our own review has constantly been guided by awareness that the Administrator in deciding what in the future shall be the ingredients of commercial bakers' bread has made decisions and imposed regulations of legislative type and that he was entitled to evaluate data accordingly. See Federal Security Administrator v. Quaker Oats Co., 1943, 318 U.S. 218, 228, 63 S.Ct. 589, 87 L. Ed. 724; cf. American Airlines, Inc., v. Civil Aeronautics Board, D.C.Cir., 1951, 192 F.2d 417, 420; United States v. Detroit Cleveland Navigation Co., 1945, 326 U.S. 236, 241, 66 S.Ct. 75, 90 L.Ed. 38. Such an approach permits the attribution of very great importance to the newness of the use of an ingredient in food and the indecisive character of available scientific data concerning its chemistry and its metabolic fate, as well as the fact that this addition to human diet has been viewed with alarm by representative persons, expert in the relevant scientific fields. This has been no civil action for damages to be decided in accordance with the preponderance of evidence in favor of one or the other of parties involved. One making a rule for the future which in practical effect will determine whether millions of people shall eat something every day may reasonably refuse to subject the general public to even slight risks and small deceptions. In these circumstances, the fact that administrative action has been dominated by great caution but serves to emphasize the reasonableness of the Administrator's conduct.
There has been some suggestion that in our disposition of these cases we should decide whether considerations of deception alone or of safety alone constitute adequate justification for the Administrator's order. But risk of injury to health and risk of deception as to bread age were both relevant matters before the Administrator, and the combined weight of proper findings in both regards led to the exclusion of petitioners' products. If at some future time safety should be adequately established it may be necessary for the Administrator to decide whether the deception here involved is serious enough standing alone to warrant rejection of a proposed ingredient. But it is not for us to express any opinion upon such a situation before it has arisen and resulted in some administrative decision.
Atlas makes a final complaint that the Administrator denied its request, or more accurately its series of requests filed between October 1950 and April 1952, to reopen the hearings which had been closed in September 1949. In this connection, Atlas urges particularly that the Administrator should have received and considered evidence of the result of its most recent experiments and studies concerning the safety of its products as well as recent governmental research conducted under the Administrator's own supervision or available to him. The Administrator found on the basis of his own inquiry that the most recent governmental research is not and does not purport to be decisive as to the safety of the Atlas products and thus did not warrant reopening the record. And, it appeared from the several submissions made by Atlas itself that its own experimentation was in fact incomplete and still in progress. Indeed, Atlas shifted its position several times during the post-hearing history of this proceeding, first asking that the views of the Food Protection Committee of the National Research Council be obtained on certain data; then, disappointed and dissatisfied with the conclusions of that body, asking a stay for two years more of its own research, and finally, asking the Administrator to consider its studies to date.
Reopening of such hearings as these is normally a matter of administrative discretion. See Interstate Commerce Commission v. Jersey City, 1944, 322 U.S. 503, 515-519, 64 S.Ct. 1129, 88 L.Ed. 1420. Nothing appears here to warrant a conclusion that the Administrator has abused his discretion in denying the petition for reopening which was addressed to him. But Atlas, relying on the wording of § 701(f)(2) of the Food, Drug and Cosmetic Act, 52 Stat. 1056, 21 U.S.C.A. § 371(f)(2), and on N.L.R.B. v. Indiana Michigan Electric Co., 1943, 318 U.S. 9, 63 S.Ct. 394, 87 L.Ed. 579, argues that it is within the discretion of this court to order the hearing reopened regardless of whether the Administrator's denial of the request to him was within the proper scope of his discretion. Assuming that we can thus examine the matter de novo, we find nothing in the record or in the subsequent offers of proof by Atlas which makes the result of our independent analysis any different from the Administrator's stated reasoning and conclusion.
In both cases the orders of the Administrator will be affirmed.