FRANK, Circuit Judge.
1. Our starting point is this: Without the prohibition of home-work contained in the order of the Administrator the Act, in its application to this industry, will be unenforcible and will become virtually a dead letter. For so it was found as a fact by the Administrator, to whom the Act assigns its enforcement.2 And the truth of his findings petitioners cannot here dispute, since they do not assert that those findings are not supported by substantial evidence heard at the Administrator’s hearing. At most, some of the petitioners cite a part of the evidence which is at variance with the findings but make no effort to show that there was not other contrary evidence of a substantial character. Moreover, as the, printed supplements to their briefs, filed under our Rule 22, do not contain all the evidence, we must assume that, if we were, to read all of it, the findings would be amply justified.3 We must, too, take those findings “at their face value,” although the Act did not require the Administrator to make them.4 Indeed, assuming for the moment that, if necessary to make the statute effective, the Act conferred on him the power to issue such a regulation, there is a “presumption of the existence of facts justifying its * * * exercise.”5
2. Notwithstanding that, on this record, petitioners are obliged to confess that the wage order will fail without the home work prohibition, they assert that the Administrator had no power to issue it. Faced with the provisions of § 8(f) — which authorize him to insert in wage orders issued pursuant to § 8 “such terms and conditions as” he “finds necessary to carry out the purposes of such orders, to prevent the circumvention * * *, and to safeguard minimum wage rates established therein”— petitioners say that, although on the facts here the elimination of home work is perhaps within that language, the regulation is so sweeping in its consequences that, had Congress intended to authorize it, the statute would have dealt with that subject specifically as it did with child labor in § 12. But in § 12 Congress dealt with child labor as an independent matter, completely eliminating the employment of minors in the affected industries because of the socially and economically undesirable character of such employment and without regard to the effect on the wage rates and hours of adults. Here the Administrator has prohibited home work not at all on the ground of its inherent 'undesirability but solely as a means of preventing the circumvention or evasion of an order prescribing adult wage rates. Moreover, doubtless having in mind the provision of § 8(b) that a wage order must “not substantially curtail employment in the industry,” the [613]*613Administrator has made a finding (which we must accept as true) that the home work regulation will have no such effect; the findings show that its consequences to the employers and employees is not as drastic as petitioners assert in their briefs. This is not a case, then, where an effort is being made to utilize § 8(f) as a subterfuge to achieve an independent end outside the scope of the Act; the regulation here is a means of accomplishing the purpose of an authorized wage order by stopping evasions of that order, and the Administrator in § 8(f) was expressly empowered to use means of that kind.6 Nor, in view of the Administrator’s findings, can it be said that this is a case where the means are so disproportionate to the authorized end that they cease to be means except in form and in truth become an independent end not contemplated in the Act.
Addison v. Holly Hill Fruit Products Co., Inc., 64 S.Ct. 1215, 1221, is not contrary to our conclusion. There the Court, interpreting one of the several specific exemptions from the Act, noted that those exemptions were “catalogued with particularity,” and said: “Exemptions made in such detail preclude their enlargement by implication.”
3. Petitioners, however, maintain that the amendment to the Act, in 1940, which added § 6(a)(5), with its specific reference to homework in Puerto Rico and the Virgin Islands, shows that Congress denied power elsewhere with respect to that subject. That argument cannot stand up; for the legislative history of § 6(a)(5) discloses that it was added to meet the peculiar economic conditions existing in Puerto Rico and the Virgin Islands; it might better be argued, indeed, that Congress found it necessary to amend the Act by adding that subsection precisely in order to limit the exercise of that power theretofore existing, before that amendment, with respect to those and all other areas covered by the Act.
4. Petitioners further contend that the legislative history of § 8(f) demonstrates that Congress did not intend thereby to delegate any authority concerning hornework. That history, briefly told, is as follows : The Senate bill, as reported by the Committee in charge, provided that all minimum wage rates and wage differentials should be established by a Board through the issuance of labor standard orders. With-respect to such orders the Board was given powers in a provision substantially the same as § 8(f) of the Act except that after the word “conditions” there was a parenthetical clause “(including the restriction or prohibition of such acts or practices).” On the floor, an amendment was adopted, without comment or objection, inserting in the parentheses the words “industrial homework.” The original House bill, which was much the same as the Senate bill, included this same provision containing the matter in parentheses. This bill, however, was recommitted. The House Committee then reported a new bill which contained no provision for wage orders but established fixed minimum wages, and included no provision resembling § 8 (f), i. e., for the prevention of circumvention or evasion. This substitute bill (with modifications not relevant here) passed the House. In the Conference Committee a compromise was made between the Senate and House bills which resulted in the present Act, with § 6 containing fixed wage rates subject to acceleration as provided in § 8. Neither the Conference Report nor the subsequent debates discussed any reasons for omitting the matter in the parentheses from the provision which now appears as § 8(f). We see nothing in that ambiguous history disclosing an intention to eliminate from § 8(f) the power to prohibit home work if that prohibition is necessary to prevent circumvention or evasions.
Cudahy Packing Company v. Holland, 315 U.S. 357, 62 S.Ct. 651, 86 L.Ed. 895, is not in point. True, there, the Court referred to the fact that authority to delegate the subpoena power, expressly granted in the Senate bill, had been rejected by the Conference Committee; but, as the Court pointed out, the significance of that fact was - that the Conference Committee substituted a provision giving the Administrator the subpoena power conferred upon the [614]*614Federal Trade Commission, and that agency, and other agencies upon which like power had been conferred, had never theretofore construed it to include the right to delegate the issuance of subpoenas; the Court also said that, if the right to delegate the subpoena power were implied, then necessarily there would be a similar implication as to all the functions assigned by the statute to the Administrator, a conclusion which the important nature of several of those functions precluded.
5. But petitioners assert that, even granting that § 8(f), taken alone, would include the power to issue the home work order, other provisions of the Act show that Congress could not have intended to authorize so extensive a regulation. The argument runs thus: § 8(f), by its terms, restricts the Administrator’s authority to that of annexing “terms and conditions’ to “orders” issued under § 8; no similar power is given him as to wage rates automatically established under § 6 when no § 8 order is operative; by § 8(e), all orders (except in unusual circumstances) expire in October, 1945. If, then, say petitioners, § 8(f) were construed to authorize the homework prohibition here, that prohibition would expire in 1945. It is unreasonable to believe, argue petitioners, that Congress intended that so extensive a prohibition should be in effect for a period of at most seven years (in this case a little more than a year), that home work could be banned during but a small span and not for the long future. Accordingly, petitioners urge, as § 8(f) applies only to orders, it must, for the sake of consistency, be construed not to include so extensive a power.
That argument proves too much. It cannot stop with eliminating from § 8(f) the authority to forbid home work. Pushed to its logical conclusion, this contention says that any regulation under that subsection lacks validity unless the statute expressly authorizes a similar regulation concerning all wage rates; petitioners would thus have us read § 8(f) out of the Act. As, of course, petitioners do not venture to go that far, their “consistency” contention comes to this: § 8(f) must be narrowly interpreted so as not to confer authority of any importance; in other words, the Administrator may make a regulation to prevent minor evasions of a wage order, but he is powerless to prevent major evasions which, as here, will gut the order. Such an interpretation — which flies in the face of the wording of § 8(f), rendering it virtually meaningless, making practically useless many a wage order, thus all but destroying § 8 — ascribes to Congress an unreasonable intention.
Were it necessary for us here to pass on the matter, we would be obliged to consider whether consistency and reasonableness require that § 8(f) be interpreted so as to apply to all wage rates or whether, quite aside from § 8(f), the Administrator has the implied power to issue regulations necessary to protect all wage rates from evasions.7 But the power of the Administrator to safeguard wage rates when no wage order is in effect is an issue not now before us. Since in the instant case there is such a wage order, it is not our present concern whether or not the act is deficient in its protection of wage rates not established by an order. If there is such a flaw, it is the function of Congress to deal with it. The legislative process is inherently such that, on occasions, the applications of a statute in practice disclose inconsistencies. While the literal meaning of a statute must yield to its evident purpose or policy, yet where a statutory provision accords with that purpose, the courts should seldom enlarge that provision, in the interest of symmetry or uniformity, in order to supply an omission.8 In interpreting an[615]*615other section of this very Act, the Supreme Court said the other day: “Legislation introducing a new system is at best empirical, and not infrequently administration reveals gaps or inadequacies of one sort or another that may call for amendatory legislation. But it is no warrant for extending a statute that experience may disclose that it should have been made more comprehensive.”9 If, then, it were true that, in working out the compromise between the House and Senate bills, Congress, by literally restricting the provisions of § 8(f) to orders, inconsistently left other wage rates subject to evasion, that would be no reason for holding that, as to orders, Congress did not mean what it said.
6. We cannot agree with the suggestion that Congress, if it had intended the Administrator to regulate home work, would have required him first to consult the industry committees or to hold hearings. For §§ 7(c) and 13 empower him to take action having more extensive consequences without such consultation or hearings.
We also consider untenable the suggestion that the home work regulation is invalid because the statute did not expressly require a hearing as a condition precedent to its issuance. Aside from the fact that here such a hearing was in fact held, the short answer is that the Constitution does not require a hearing before the promulgation of such a regulation. Bowles v. Willingham, 321 U.S. 503, 64 S.Ct. 641; Phillips v. Commissioner, 283 U.S. 589, 596, 597, 51 S.Ct. 608, 75 L.Ed. 1289; Bi-Metallic Investment Co. v. State Board, 239 U.S. 441, 36 S.Ct. 141, 60 L.Ed. 372.10
7. We reject the argument that stricter enforcement or some other measure would meet the problem without the need for prohibiting home work, for the Administrator has made express findings to the contrary.11
8. Nor is there, we think, anything to the point that the Administrator has made an unreasonable discriminatory classification by his exemptions from the prohibition. The Fifth Amendment contains no “equal protection” clause.12 Moreover, it is by no means clear that the exemption would be an invalid classification even under the Fourteenth Amendment.13
9. Equally unsound is the argument that the prohibition of home work violates due process. It is perhaps sufficient to nc“-e that, to support this argument, petitioners rely heavily on the remarks of Field, J., concerning liberty of contract in his concurring opinion in Butchers Union Co. v. Crescent City Co., 111 U.S. 746, 757, 4 S.Ct. 652, 28 L.Ed. 585. Surely the extreme views there expressed are no longer authoritative.14
10. Finally we come to the contention that, if the statute confers the asserted authority on the Administrator, then it unconstitutionally delegates legislative power. The question raised by that contention is not new. More than two thousand years ago, a profound student of government, from whom we derive the concept of a “government of laws, and not of men.”15 explained the inescapability of some delegation by legislators. The “rule of law,” he said, “is preferable to that of any indi[616]*616vidual” and “he who bids the law rule maybe deemed to hid God and Reason alone rule, but he who bids man rule adds an element of the beast; for desire is a wild beast, and passion perverts the minds of rulers, even when they are the best of men,” whereas “the law is reason unaffected by desire.” But, sagely, he noted that “there may indeed be cases in which the law seems unable to determine,” and asked “but in such cases can a man?” He answered that “the law trains officers for this express purpose, and appoints them to determine matters which are left undecided by it, to the best of their judgment. Further it permits them to make any amendment of the existing laws which experience suggests * * * And at this day there are magistrates, for example judges, who have authority to decide matters which the law is unable to determine * * *” He added that “no one doubts that the law would command and decide in the best manner whenever it could. But some things can, and others cannot, be comprehended under the law, and this is the origin of the vexed question whether the best law or.the best man should rule. For matters of detail about which men deliberate cannot be included in legislation. Nor does anyone deny that the decisions of such matters must be left to man * * *”
Without mentioning that author, our Supreme Court has often echoed his words. In 1904, it said,16 “Congress legislated on the subject as far as was reasonably practicable, and from the necessities of the case was compelled to leave to executive officials the duty of bringing about the result pointed out by the statute. To deny the power of Congress to delegate such a .duty would, in effect, amount but to declaring that the plenary power vested in Congress to regulate foreign commerce could not be efficaciously exerted.” In 1934, it said, “Undoubtedly legislation must often be adapted to complex conditions involving a host of details with which the national Legislature cannot deal directly. [617]*617The Constitution has never been regarded as denying to the Congress necessary resources of flexibility and practicability * * *. Without capacity to give authorizations of that sort we should have the anomaly of a legislative power which in many circumstances calling for its exertion would be but a futility.”17
True, in the case last quoted and in another decided about the same time,18 it was held that the delegations there involved were so lacking in adequate standards, so unrestrained, as to be unconstitutional. But, in the light of many subsequent decisions, those two cases must now be considered exceptional, restricted to their particular or very similar facts. The standard in § 8(f), coupled with the provisions of the other provisions of § 8 and with § 2, amply meet the test of adequacy of standards according to recent Supreme Court decisions.19
Petitioners scarcely try to distinguish those cases. They fall back on a rigid conception of the “separation of powers” doctrine. Such an inflexible conception finds no justification in English or American history,20 and cannot be realized in practice.21 As Holmes, J., said in his dissenting opinion in Springer v. Philippine Islands, [618]*618277 U.S. 189, 211, 48 S.Ct. 480, 485, 72 L.Ed. 404: “It does not seem to need argument to show that however we may disguise it by veiling words we do not and cannot carry out the distinction between legislative and executive action with mathematical precision and divide the branches into watertight compartments, were it ever so desirable to do so, which I am far from believing that it is, or that the Constitution requires.”22
As, in spite of Supreme Court decisions which should put the matter at rest, we still frequently hear arguments which assume an inherent infirmity in delegation of rule-making to administrative officers, it seems worthwhile to analyze the problem somewhat more in detail.
In the history of this country, subordinate legislative powers were delegated at an early day, both by our state legislatures and Congress.23 What is new for us is the name “administrative law.”24 The writings of Goodnow and Freund introduced that name to the American legal profession several decades ago. But only in recent years has it come into extensive use. Even now, it finds no place in many of our conventional “digests.” A new name, a novel label expressive of a new generalization, can have immense consequences. Emerson said, “Generalization is always a new influx of the divinity into the mind. Hence the thrill that attends it.”25 Confronted with disturbing variety, we often feel a tension from which a generalization, an abstraction, relieves us. It serves as a de-problemizer, aiding us to pass from an unstable, problematical, situation to a more stable one. It satisfies a craving, meets what Emerson called “the insatiable demand of harmony in man,” a demand which translates itself into the so-called “law” of “the least effort.”26 But the solution of a problem through the invention of a new generalization is no final solution: The new generalization breeds new problems. Stressing a newly perceived likeness be[619]*619tween many particular happenings which had theretofore seemed unlike, it may blind us to continuing unlikenesses. Hypnotized by a label which emphasizes identities, we may be led to ignore differences. In all fields of thought this evil is encountered. Nowhere can it do more harm than in democratic government — and in democratic courthouse government in particular. For, with its stress on uniformity, an abstraction or generalization tends to become totalitarian in its attitude towards uniquenesses.27 While, then, the concept of “administrative law” is invaluable, because it pulls together, for comparative study and common use, techniques and ideas developed in scattered areas of administrative action,28 there is danger that that concept may yield inelastic uniformities. All administrators should not be treated identically.29 Yet the problem of the delegated powers of a particular administrative of[620]*620ficer is illuminated by the recognition that it is part of a general problem common to what Patterson refers to as the “pluralistic universe of administrative law.” The illumination, however, has brought fear to some. The very word “delegation,” when now coupled with the words “administrative officers,” strikes terror in their hearts. They resemble the child who was horrified when his attention was called to the facts that his tongue was wet and his shoes full of feet. For delegation is a name for something that has always been present in society and always will be.30 Perhaps the fright the label engenders can be reduced by observing that “discretion,” which is familiar, is kin to “delegation.” Almost boundless discretion has traditionally been conferred upon state and federal prosecuting officers to begin or not to begin proceedings of enforcement of some of the huge number of penal statutes (and against some rather than against other violators) and to “settle” such cases by “bargain day” methods.31 Wide, too, is the beneficent discretion of courts sitting in equity to relax stiff legal rules.32 And the power of court’s in general to interpret statutes has been said by some persons to be, in practical' effect, a sort of supplementary legislation which the legislature necessarily leaves to the courts.
Indeed, those who today criticize the transfer of “subsidiary legislation” to administrative officers forget that, inspired by somewhat similar motives, there has been and still is much criticism of the power exercised by judges in construing statutes, that Bentham, Livingston, and their disciples (some even in our time33) have insisted that all “law” must emanate solely from the legislature, and have tried, through codification, to destroy all “judicial legislation.” 34 Repeated attempts on the European continent to exploit that notion have invariably proved disappointing.35 Legal certainty to be attained by eliminating, via codification,, all judicial law making is a fatuous dream.36 Courts in their interpretation of statutes often [621]*621cannot avoid some such legislation. The enactment of many a statute thus, by implication, calls on the courts to engage in supplemental law making. That activity should always, of course, be modest in scope.37 But the necessary generality in the wording of many statutes, and ineptness in the drafting of others, frequently compels the courts, as best they can, to fill in the gaps, an activity which, no matter how one may label it, is in part legislative.38 Sagacious legal scholars of high repute, such as, for instance, John Chip-man Gray, Wigmore, Allen and Radin, have said that courts, in discharging their duty of carrying out the express will of the legislature as faithfully as they can, are frequently unable to escape the responsibility of engaging in supplemental legislation.39 As Chief Justice Hughes said in 1928, “a federal statute finally means what the [Supreme] Court says it means.”40 Thus the courts in their way, as administrators in their way, perform the task of supplementing statutes. In the case of the courts, we call it “interpretation,” or “filling in the gaps”; in the case of administrators we call it “delegation” of au[622]*622thority to “supply the details.” In both instances, the task is unavoidable.
There are those who, while they grudgingly concede the necessity of delegation of subordinate legislative powers to administrative officers, are disturbed because currently it is accompanied by what they consider an unwise breadth of authority in fact-finding given to such officers when deciding particular cases arising under administrative regulations.40a Such persons urge the courts to set narrow limits to the extent to which legislative powers may validly -be assigned to administrators. But authority to find the facts is inseparable from the decision of specific cases.40b Judges trying nonjury cases have at least an equal breadth of authority in fact-finding. And extensive indeed it is, since, in the process of decision-making, the “minor” (fact) premise often plays a part as important as (if not more important than) the “major” (rule) premise.41 (Recognizing that, where the testimony is in conflict, the determination of the facts by a trial judge involves a guess as to the accuracy and honesty of the witnesses, some commentators have, indeed, referred to the “discretion of the judge” in the “estimation of the testimony.”42) Appellate courts sometimes make greater demands of administrators as to precision and detail in fact-finding than they ordinarily do of trial judges sitting without a jury43 — although it is an open secret that some federal trial judges who resent their own obligation in nonjury cases to make findings of fact express antipathy to the fact-finding powers vested in administrative officials. And upper courts always require far more of administrators than they do of juries : few administrators are permitted to return general (factless) verdicts.44 The suggestion that appellate courts should, when reviewing administrative action, inquire more searchingly into the findings of administrators than they now do under the “substantial evidence” rule is difficult to reconcile with the practice of those courts in reviewing decisions of trial courts; moreover it would burden appellate courts impossibly unless the number of appellate judges were increased at least tenfold.45 The truth is that much of the regulation of the affairs of citizens which the complexities of our civilization necessitates calls for a very considerable use of the administrative device,46 and that its use must be accompanied by grants of delegated powers47 both as to the making of rules and the finding of facts. Complexity is our lot and we should not rail against its inevitable concomitants.48 Improvement in the procedures of administrative agencies and in their relations to the courts, when they judicially review administrative conduct, is desirable and pos[623]*623sible.49 The phrases “separation of powers” and “a government of laws, and not of men,” if properly construed, embody principles of the first importance in a democracy; but if so construed as-seriously to cripple effective government, they will lead to democracy’s downfall, for, as the Federalist tells us, an ineffective government paves the way to anarchy and thence to depotism.50 Laws neither execute nor interpret themselves. Men must discharge those functions. Above all what we need is the selection of well-trained, honest, able men, conscientiously obeying the laws, and imbued with the spirit of democracy, to serve as administrators and on the bench.51
Their selection, however, is not a judicial function. And it is surely not our function in this case to thwart the legislative purpose (whether we like it or not) by so interpreting this statute as to leave it, as to the industry here concerned, a mere bit of worthless printing.
Petitions denied.
Aristotle, Politics, Bk. III, Ch. 16, 1287a, 24 et seq. He also spoke of “filling up the gaps which the law is obliged to leave.”
And in his remarks elsewhere on “equity,” he said that “all law is couched in general terms, but there are some cases upon which it is impossible to pronounce correctly in general terms. Accordingly, where a general statement is necessary, but such a statement cannot be correct, the law embraces the majority of cases, although it does not ignore the element of error. Nor is it the less correct on this account; for the error lies not in the law, nor in the legislature, but in the nature of the case. For it is plainly impossible to pronounce with complete accuracy upon such a subject matter as human action. Wherever then the terms of the law are general, but the particular case is an exception to the general law, it is right, where the legislator’s rule is inadequate or erroneous in virtue of its generality, to rectify the defect which the legislator himself, if he were present, and had he known it, would have rectified in legislating * * * This is in fact the nature of the equitable; it is rectification of law where it fails through generality * * * For where the thing to be measured is indefinite the rule must be indefinite * * * ” Nicomachean Ethics, Bk. V, Ch. 10, 1137b, 13-31. “The equitable seems to be just and equity is a kind of justice, but goes beyond the written law. This margin is left by legislators, sometimes voluntarily, sometimes involuntary; involuntarily when the point escapes notice, voluntarily when they are unable to lay down a definition, and yet it is necessary to lay down an absolute rule; also in eases where inexperience makes it hard to define * * *; for life would not be long enough for a person to enumerate the cases.” Rhetoric, Bk. I, Ch. 13.
By a “government of men” Aristotle apparently meant a government in which a specific judgment or decree affecting a specific person is rendered by the legislator or legislative body. Curiously, some of our Congressional “private bill” legislation would come within that criticized category.
Aristotle’s point of view reappeared in a New Fork Times’ editorial of December 15, 1943: “It is the proper function of Congress to frame laws and general policies, to delegate powers where-ever detailed control is necessary, and to see that laws are properly administered. But it is not the function of Congress itself to administer the law. It is not its business to meddle in specific decisions. Once it does so * * * it will find itself overwhelmed with administrative details that it is not remotely organized to attend to * * * Such detailed meddling can only * * * lead towards administrative chaos.”