concurring in the result:
I agree with and join the majority of the court in holding that the plaintiffs are not entitled to judgment and the petitions must be dismissed. But I depart so widely in my reasons as to Count I that I think it better to set them forth separately. This I will do in summary form, since no one is joining me in this statement and nothing I say will be "the law.” I divide what follows into two parts, I Jurisdiction, and II Merits, Count I. As to Count II, I join in the court’s per curiam opinion.
I. Jurisdiction
There are two jurisdictional barriers the plaintiffs must surmount. The first is imposed by 28 U.S.C. § 455, as amended, (Supp. V, 1975), which prohibits our participation in cases wherein our impartiality might be questioned, with detailed treatment of different classes of disqualification. The second rises from the Testan doctrine (United States v. Testan, 424 U.S. 392 (1976)) which will have to be carefully considered hereafter in connection with any claim which is at all outside the usual bread and butter run of our cases.
A. As to 28 U.S.C. § 455. I will be frank to say that anyone might reasonably question my impartiality. However, the disqualification is not jurisdictional, may be waived, and has been waived by both parties, unless it falls under § 455(b). If it does, it cannot be waived. Section 455(b) bars a judge who has a "financial interest” in a party. If that were all there would be no problem because we have only an interest in the issues to be litigated, or some of them, as abstract propositions, not in the parties, and therefore our interest does not come under the definition of "financial interest” in section 455(d)(4). But the statute also bars us if we have "any other interest that could be materially affected by the outcome of the proceeding,” that is, are interested otherwise than in a pecuniary way. This *265language seems to me hopelessly ambiguous. We are not interested in the outcome of this case, only in the reasons given for that outcome. The reasons just might possibly support a back pay claim by ourselves. If this kind of interest is an interest in the outcome of a case, it trails off into all sorts of remote connections. Thus, a judge sitting on a dispute between a labor union and a manufacturing corporation might have section 455(b) issue interest if he held stock in any other company that employed union labor, members of any other union. I think, rather than leaving issue interest in such uncertainty, the Congress may well have intended that any such issue interest should be dealt with under section 455(a) and thus that the disqualification should be waivable. This may seem like hair splitting and probably is, but hair splitting often is unavoidable in any attempt to determine the boundaries of Court of Claims jurisdiction.
If section 455(b) is fatal to my jurisdiction, it is fatal to that of all other active and senior judges of this court and that of any other Federal judge who might sit here by designation. Thus, as plaintiffs have presently stated their claim, they would have no tribunal in which they could enforce it if valid, though it is largely founded on the Constitution. It is a common canon of interpretation to read a jurisdictional statute as not barring litigation of a claim of that kind unless the language affords "clear and convincing evidence” that Congress intended that result. Califano v. Sanders, 430 U.S. 99 (1977). To the same effect, in a case closer to us see Russian Volunteer Fleet v. United States, 282 U.S. 481, 491 (1931). Thus, for purposes of this case, section 455(b) does not bar jurisdiction, even if, which I do not admit, in some other context, it might do so, for a similar interest. I find it ambiguous in any context. Ordinarily, of course, the matter would be resolved under section 455(a) and the ambiguities of section 455(b) need not cause difficulty.
As I read my colleagues, they find an unambiguous bar in section 455 and find in a supposed "doctrine of necessity” the means of leaping over it. The differences from my approach are more than semantic.
*266The cold truth is that the plaintiffs are suing in the United States Court of Claims under the Tucker Act, 28 U.S.C. § 1491, as amended. They invoke only the powers of this court, such as they are. This court is not an ombudsman. It is a court of severely limited jurisdiction. It can take cases only against the United States, only if they seek money judgments, and only to the extent Congress has waived sovereign immunity. Where money damages are not claimed, Congress has recently enacted a most sweeping waiver of immunity, Act of October 21, 1976, Pub. L. 94-574, 90 Stat. 2721, but where money is claimed, existing waivers are limited and narrowly construed, leaving large categories of possible claims not provided for. The decision whether a claim is consented or unconsented is frequently a matter of legal hair splitting courts would be ashamed of in any other context. For example, by recent authoritative decision, United States v. Hopkins, 427 U.S. 123 (1976), a contract employee of an unappropriated fund agency can sue for back pay, if deprived of it by legal wrong, but one appointed and enrolled in the ordinary non-contractual way cannot. There is no reason for this distinction. It is simply a matter of defective statutory draftsmanship. Someday, probably, it will be corrected.
To me it seems absurd for a court having such limited jurisdiction to invoke a "doctrine of necessity,” saying we must override an unambiguous statutory bar because if we do not, the poor fellows have no remedy. If they have none, they only join a club that has many other members.
The "doctrine of necessity” to my mind exists only for courts of general jurisdiction and cannot be availed of to override unambiguous jurisdictional limits. If, however, section 455(b) is conceded to be ambiguous, the "doctrine of necessity” becomes only a minor aspect of the general rule stated in Califano and in other cases cited therein.
B. The Testan Doctrine. The next problem, as I see it, is also jurisdictional. It is a venerable maxim of our jurisprudence that no one can sue for the salary of a position, except a salary actually fixed by law as the salary of a position he actually holds, or held until unlawfully removed. The reasons given for this rule are various. In United States v. Testan, supra, it is imputed to the absence *267of any expression of intent by Congress that the United States shall be liable to pay the money when that is not expressed in legislation, outside the Tucker Act, such as the Back Pay Act, 5 U.S.C. § 5596(b). This must exist as a jurisdictional prerequisite to suit in the Court of Claims, entirely apart from whether legal rights have been breached, or wrongs committed, that might be remedied in some other way. The problem commonly arises, as in Testan, in promotion controversies. A claimant to the emoluments of a promotion almost always fails in this court on jurisdictional grounds.
In United States v. Lovett, 328 U.S. 303 (1946), it would not have been enough for jurisdiction here if Congress had passed a bill of attainder against Mr. Lovett to bar his appointment or promotion to a position he did not hold. That it sought to deny him a position he already had, or rather the pay of it, would have met the Testan test, had it then been articulated.
Though not mentioned in Testan, the power of the purse is older than the judicial independence and like it is a safeguard of freedom. Parliaments exercised power of the purse to clip the talons of the Stuart kings when judges were officers of the Royal household, toadies who could be appointed or removed at the Royal pleasure. The ever infamous Judge Jeffries was one of them. To sue the Government for back pay and obtain a judgment involves an infringement on this ancient power unless expressly consented to; hence the Testan decision and others before it, where the Attorney General complained on behalf of the Crown (President and Congress) that we had taken a jurisdiction not properly ours. Differences of opinion are possible as to the scope of consent, but none as to the legal impossibility of exceeding that scope once it is determined. And it is wholly an independent question from whether the Congress has violated anybody’s constitutional rights. Lynch v. United States, 292 U.S. 571 (1934).
Applying a Testan analysis to the claims before us, the plaintiffs’ most plausible theory, under Count I, is that Congress has under Article III a duty to adjust the pay of judges from time to time, to compensate for decline in the buying power of the dollar. This would appear to be very *268similar to the duty of the Civil Service Commission, involved in Testan, to equalize the job classifications of Government employees. Our theory was that on the Commission’s performing this duty, plaintiffs would then have a money claim to satisfy jurisdictional requirements. According to the Supreme Court, reversing, a back pay claim here cannot be postulated on the supposition that another body will perform its duty, if Congress has not indicated its intent to found a pecuniary liability on such a supposition. Similarly plaintiffs here do not have a money claim on the mere assertion that Congress should have adjusted their salaries, when Congress has never indicated that any employees, judicial or other, should have a money claim above and beyond the salary fixed by statute. This is a matter wholly independent of whether the legal duty sought to be enforced originates in the Constitution. We cannot bootstrap ourselves into jurisdiction by the equitable doctrine of viewing that as done which ought to have been done.
Plaintiffs, however, also say that their Article III right is "self-executing.” This is a very important contention, because the court has apparently adopted it, and if correct it gets us around the Testan barrier and opens up the merits for our consideration. In my view, however, the Article III salary right is to an exercise of sound discretion by the Congress, taking into account the decline in the buying power of the dollar. I do not at all agree Congress is bound, in adjusting salaries, to conform to the changes in the Consumer Price Index. Other options are open. I discuss the nature of the right, as I see it, more fully below. If such a right can be described as self-executing, that of the Testan plaintiffs was self-executing too, and the barrier the Supreme Court held to be fatal to our jurisdiction was imaginary. I may say I find an inherent contradiction in the majority designation of the right as self-executing, and the other incidents of the right as attributed to it by the court.
As regards Count II of the petitions, I do not think the Testan analysis is fatal to jurisdiction, for the reasons stated in Gentry v. United States, 212 Ct. Cl. 1, 546 F.2d 343 (1976). The claim in Count II is founded on a statute, as *269amended, by deletion of an allegedly unconstitutional proviso. The claim did not require action by anyone other than this court. Hence I agree that the court may properly go into the merits under Count II.
II. Merits, Count I
The court, having determined that Article III gives a self-executing right, goes on to discuss what the duties of Congress were and are in the premises, and concludes that the neglect of Congress to make cost of living adjustments in judges’ pay, if reprehensible, does not rise to the dignity of a constitutional violation. I seem to read between the lines a reluctance to hold that there can be such a thing as a constitutional violation not remediable in any court. The popular view of such matters no doubt is that if you can’t sue on it, the Constitution doesn’t cover it, much as serious thinkers used in the twenties to say that it was impossible for the Prince of Wales (later King Edward VIII) to be unfashionably dressed. As people see it, a constitutional right is nothing if it is not a successful lawsuit. My view, however, is that constitutional rights can and do exist, and can be breached, even though no court is open for their enforcement. This certainly has occurred historically. E.g., in United States v. Sioux Nation, 207 Ct. Cl. 234, 518 F.2d 1298, cert. denied, 423 U.S. 1016 (1975), we recognized at least that this court formerly had taken such a view. The Congress has the duty to rectify constitutional breaches not remediable in the courts, or widen court jurisdiction to cover the cases, but cannot be ordered to do either. My colleagues view the possibility of, e.g., a discriminatory failure to adjust for inflation, as to judges only, or a Weimar Republic type of inflation of the money to zero value, as horrors they do not wish to approve as constitutional in advance, and therefore they assert their jurisdiction, get into the merits, and exonerate the Congress of any constitutional violation up to now.
Since in my view we do not have jurisdiction, the added heinousness of any constitutional violation in the premises would not confer it. It is therefore inappropriate for me to hold whether or not there have been violations of Article *270III rights up to now. However, I think it is at least arguable that Article III confers a right not just to exercise of sound discretion and to nondiscrimination, but also to effective ways and means to implement regular adjustments, for cost of living, in face of the inflation which we have experienced continuously over the entire lifetime of the majority of persons now living, and must expect to continue, checked more or less, perhaps, but never stopped, until all of us are dead. Such a right, if it exists, is frustrated by the unfortunate statutory arrangements (Federal Salary Act, 2 U.S.C. § 351 & ff.) which link such adjustments to those proposed for others than judges, and expose them to defeat because of differences about entirely unrelated issues. Also, ordinary legislation, if defeated, can be redrafted to meet the objections of those who cast adverse votes, but under the Act under consideration, there is no provision for this, and an adverse vote halts action for an indefinite period, in 1974 as long as 3 years. I could not, therefore, jurisdiction questions aside, have joined in holding Article III is unviolated up to now. It does not follow, the Article not being self-executing, that ascertaining the damages would have been easy, and it could be they would have been nominal, particularly with respect to plaintiffs whose pay when they first entered the Federal judiciary did not materially exceed in real buying power the pay actually received during the period sued upon. A further caveat is this: the Federal Salary Act, being in my view an inadequate performance of a duty imposed on Congress by Article III, but some performance as far as it goes, it does not follow that the Act should be struck down, in whole or in part. Such a deletion from the statute book would make it even more out of harmony with' the constitution than it is now. A deletion, on constitutional grounds, such as in Gentry, supra, makes the remainder of the statute constitutional and thus is a kind of judicial amendment that thrusts in an entirely different direction.
The majority’s holding that it has jurisdiction and its review of the statute and performance thereunder I think *271warrant me in going as far as I have done in my comment on Count I of the petitions.
Skelton, Senior Judge, with whom Kashiwa and Kun-zig, Judges, join, concurring in part and dissenting in part:*CLAUSE B PROVIDING FOR A ONE-HOUSE VETO IS UNCONSTITUTIONAL
I concur with and join in all of the majority opinion except that part dealing with Count II of plaintiffs’ petitions wherein the majority holds that subdivision (B) of Title 2, Section 359(1) (hereinafter Clause B) which provides for a "one-House veto” is constitutional.1 I cannot agree that Clause B is constitutional, because in my opinion, the one-House veto which it provides for violates and is contrary to the Constitution and for that reason Clause B is invalid and constitutionally impermissive. The one-House veto violates the following articles of the Constitution: (1) Article I, section I, which vests the legislative power of the United States, not in one House, but "in a Congress * * * which shall consist of a Senate and a House of Representatives;” (2) article I, section 7, clause 3, which reserves to the President the power to veto *272"every order, Resolution, or vote to which the concurrence of the Senate and House of Representatives may be necessary;” and (3) article II, section I, which vests all Executive power in the President.2
Furthermore, the one-House veto violates the principles of the doctrine of Separation of Powers on which our government was founded. It is contrary to the intent of the Framers of the Constitution. They intended that "the powers of the three great branches of the National Government be largely separate from one another.” Buckley v. Valeo, 424 U.S. 1, 120 (1976).
There is much authority for the proposition that the one-House veto is unconstitutional. As early as 1897 the Senate Committee on the Judiciary stated that "every exercise of 'legislative powers’ involves the concurrence of the two Houses.3 This principle was followed consistently until 1939 when Congress began to enact legislation that has allowed one or both Houses of Congress (or committees) to repeal, invalidate, or modify previously delegated authority.4 President Woodrow Wilson was definitely of the view that such statutes were unconstitutional and openly stated his views to that effect.5 Every President, beginning with Hoover and including Roosevelt (who wrote an opinion about it), Truman, Eisenhower, Kennedy, Johnson, Nixon, *273and Ford, each supported by his Attorney General, has contended that the one-House veto was unconstitutional.6
When President Ford signed the Federal Election Campaign Act, 2 U.S.C. § 431, et seq. on May 11,1976, he stated unequivocally that the one-House veto in the Act violated the Constitution. In this regard he stated:
A more fundamental concern is that these amendments jeopardize the independence of the Federal Election Commission by permitting either House of Congress to veto regulations which the Commission as an Executive agency, issues. This provision not only circumvents their original intent of campaign reform but, in my opinion, violates the Constitution. I have therefore directed the Attorney General to challenge the constitutionality of this provision at the earliest possible opportunity. [Emphasis supplied.]
Many Representatives and Senators have expressed the view that the one-House veto was unconstitutional.7 Senator Dirksen in speaking against a committee veto (which to all intents and purposes is the same as a one-House veto) stated:
Those provisions were unconstitutional then and they are unconstitutional now. It is true that there has been no pronouncement to that effect by the Supreme Court of *274the United States or by any other competent Federal court. The reason that it is difficult to test the legality of such statutes is found in the fact that, in order for the matter to be presented to the courts, somebody must have standing to challenge its validity.
However, that fact does not make it constitutional. * * * [100 Cong. Rec. 5095 (1954)]
It has been the more or less consistent view of the Attorneys General and the Department of Justice that the one-House veto is unconstitutional. Assistant Attorney General Antonin Scalia, Office of Legal Counsel, Department of Justice, testified to that effect on May 15, 1975, before the Subcommittee on Separation of Powers of the Senate Committee on the Judiciary.
It is very significant and most persuasive that the 140 able, experienced, and knowledgeable federal judges who filed these suits, together with their distinguished attorney former Supreme Court Justice Arthur Goldberg, all of whom are learned in constitutional law, believe the one-House veto is unconstitutional as shown by the allegations in their petitions and the arguments in their briefs.
The Department of Justice presented arguments attacking the constitutionality of the one-House veto in the recent case of Clark v. Valeo, 559 F. 2d 642, decided January 21, 1977, by the Circuit Court of the District of Columbia. In its brief in that case the Assistant Attorney General stated:
The fundamental vice of the one-House veto is that it purports to permit lawmaking without participation in the relevant policy choices by both houses of Congress or by the President. [Id. at 19-20.]
On one occasion, the Attorney General, in commenting on the interference with the executive power of the President by the use of the one-House veto by either House of Congress, stated:
It must be assumed that the functions of the President under this Act were executive in their nature or they could not have been constitutionally conferred upon him, and so there was set up a method by which one House of Congress might disapprove Executive action. No one would question the power of Congress to provide for *275delay in the execution of such an administrative order, or its power to withdraw the authority to make the order, provided the withdrawal takes the form of legislation. The attempt to give to either House of Congress, by action which is not legislation, power to disapprove administrative acts, raises a grave question as to the validity of the entire provision in the Act. * * * [37 Ops. Att’y Gen. 56 (1933).] [Emphasis supplied.]
Attorney General Brownell pointed out that the attaching of invalid conditions to legislation enacted by Congress would gravely jeopardize the constitutional system of the separability of the branches of government. He stated:
The present proviso cannot be sustained on the theory that it is a proper condition attached to an appropriation. It is recognized that the Congress may grant or withhold appropriations as it chooses, and when making an appropriation may direct the purposes to which the appropriation shall be devoted. It may also impose conditions with respect to the use of the appropriation, provided always that the conditions do not require operation of the Government in a way forbidden by the Constitution. If the practice of attaching invalid conditions to legislative enactments were permissible, it is evident that the constitutional system of the separability of the branches of Government would be placed in the gravest jeopardy. [41 Ops. Att’y Gen. 23.] [Emphasis supplied.]
In the case of Clark v. Valeo, supra, the majority of the court held that the plaintiff did not have standing to maintain the suit and for that reason did not reach the question of whether or not the one-House veto in the statute involved in the case was constitutional. However, in a very able opinion, Judge MacKinnon held that the plaintiff had standing as a voter to maintain the suit, and stated emphatically that the one-House veto provision in the Federal Election Campaign Act was unconstitutional. He said:
* * * This case has a lay-over and a one-house veto. There is nothing wrong with a lay-over provision, be it 30 days or six months. This can allow Congress to act in a constitutional manner through both houses and the President and that is permissible. But the one-house veto is not a ”difference. . . of degree” — it is a completely *276different method of accomplishing a legislative result by a congressional procedure not authorized by the Constitution; i.e., by one house instead of by two houses and the President.
That the unconstitutional procedure (the one-house veto, not the lay-over) may also influence legislation is not to be equated with the influence or action that Congress may exercise or resort to during a simple layover provision. There may be a difference only in degree between the influence of a 30 day lay-over and a six months lay-over, but there is a radical difference in kind between the influence of a one-house veto and any simple lay-over provision. The former is invalid, the latter is valid, for the reasons herein outlined. [Emphasis supplied.] [559 F. 2d at 681.]
Judge MacKinnon pointed out that under the one-House veto provision, a bare majority of a quorum in the Senate (26 Senators) or a bare majority of a quorum in the House (110 Representatives) could defeat the proposed regulations of the election commission. In this connection, he stated:
* * * [T]he scheme of the Act makes it possible for a bare majority of a quorum (which frequently occurs) in either house of Congress to influence regulations constituting "mies of law,” while completely depriving the President, possibly one house of Congress, and one-third plus one of the members of each house, from exercising legislative power supposedly vested in them by the Constitution. Short reflection upon the enormity of these constitutional violations will convince anyone of the tremendous harm they cause to the basic procedures that the Constitution provides for the enactment of legislation to govern the Nation. [Emphasis in original] [Id at 683.]
* * * Such a bare majority is given the power to accomplish what the Constitution otherwise requires of a two-thirds majority in both houses. This enhanced legislative authority for a bare majority of a quorum of one house to take positive legislative action clearly violates the constitutional requirement that legislation should be passed by both houses and be signed by the President. Art. I, sections 1, 7. The Constitution confers "AZZ legislative powers” on both houses and does not permit a single house to usurp the constitutional legislative power of "Congress.” [Emphasis in original] [Footnotes omitted] [Id at 683-84.]
*277Although Judge MacKinnon’s opinion is labeled a "dissenting” opinion, such is not really the case on the one-House veto question, because the majority did not reach nor decide that issue. Justice White of the Supreme Court, in a one paragraph comment in his concurring opinion in Buckley v. Valeo, 424 U.S. 1, 284-85 (1976), discussed certain aspects of the one-House veto as applied to the peculiar facts in that case, but the opinion of Judge MacKinnon aforesaid is the only decision that has ever been written by an appellate judge that has considered the question in depth on the merits. In my opinion, his decision was correct and should be followed.
Furthermore, the vice of the one-House veto created by Clause B of the Act is compounded by the fact that in exercising such veto by a bare majority of a quorum of either House a small group of Senators or Representatives, or even a larger group, can do so selectively, that is, they can veto "all or part” of the salary adjustments made by the President and there is no recourse or appeal from their decision. In this manner they can dictate, change, or rewrite the salary adjustments made by the President. Not even the President has such an "item veto” nor the security that his veto will not be overridden. Such a system ignores and is contrary to the Constitution and the doctrine of Separation of Powers of our Government.
It is significant and very important to note that Assistant Attorney General Lee, who argued the instant case for the Government, admitted and conceded in open court that Clause B of the Act in question was unconstitutional.8 He even stated that if this was a proper case otherwise, nothing would please him more than to have this court hold that Clause B was unconstitutional. In making these statements, admissions, and concessions, he was speaking for the Government. It would seem that as far as this case is concerned, the Government should be virtually bound by its admission that the one-House veto is indeed unconstitutional. In 28 U.S.C. § 516 (1970), it is provided as follows:
*278Except as otherwise authorized by law, the conduct of litigation in which the United States, an agency, or officer thereof is a party, or is interested, and securing evidence therefor, is reserved to officers of the Department of Justice, under the direction of the Attorney General.
In United States v. California, 332 U.S. 19, 26-27 (1947), the Court recognized the powers of the Attorney General in the following language:
* * * It is not denied that Congress has given a very broad authority to the Attorney General to institute and conduct litigation in order to establish and safeguard government rights and properties. * * * [Footnote omitted.]
See also United States v. San Jacinto Tin Co., 125 U.S. 273, 279, 284 (1888); Kern River Co. v. United States, 257 U.S. 147, 154-55 (1921); Sanitary District v. United States, 266 U.S. 405, 425-26 (1925). Furthermore, it is significant that the position of the Attorney General in a suit against the United States in the Court of Claims is specially dealt with in 28 U.S.C. § 518 (1970) as follows:
(a) Except when the Attorney General in a particular case directs otherwise, the Attorney General and the Solicitor General shall conduct and argue suits and appeals in the Supreme Court and suits in the Court of Claims in which the United States is interested.
(b) When the Attorney General considers it in the interests of the United States, he may personally conduct and argue any case in a court of the United States in which the United States is interested, or he may direct the Solicitor General or any officer of the Department of Justice to do so.
Although we have received the views of Congress regarding the unconstitutionality of Clause B, we refer to the Court’s opinion in The Gray Jacket, 72 U.S. (5 Wall.) 370, 371 (1866);
The court has considered the question whether counsel shall be heard in this cause on behalf of the Treasury Department, and has instructed me to say that in causes where the United States is a party, and is represented by the Attorney-General or the Assistant Attorney-General, or special counsel employed by the Attorney-General, no *279counsel can be heard in opposition on behalf of any other of the departments of the government.
We have, therefore, given the views of Assistant Attorney General Lee great weight in deciding the question of the unconstitutionality of Clause B.
There is no question that the Constitution authorizes and empowers Congress to fix the salaries of federal judges.9 However, this is power and authority that can be delegated to the President or an executive agency. In the instant case, Congress delegated to the President the authority and power to fix the salaries of federal judges by the enactment of the Postal Revenue and Federal Salary Act of 1967 (Act), 2 U.S.C. § 351, et seq. This was a valid exercise of legislative power. Once made, such a delegation of power to adjust salaries becomes an executive power, and it remains such until and unless legislation is enacted (by both houses) withdrawing it. See Matter of Moran v. LaGuardia, 270 N.E. 450, 1 N.Y.2d 961 (1936). No such legislation was ever enacted.
The majority admits that the Act was a delegation of authority by Congress to the President with reference to the adjustment of salaries, but contend that he only had authority under such delegation to make recommendations to Congress. While it is true the statute speaks in terms of "recommendations,” the meaning and intent of the statute was to delegate to the President the power and authority to set and adjust the salaries of judges and not to merely make recommendations. This is clearly shown by the facts surrounding the application of the statute and the fixing of salaries by the President on two different occasions when the salaries set by the President became law automatically without any action whatever by the Congress. In 1967 the salary of district judges was $30,000 and that of judges of courts of appeals was $33,000. In that year the Postal Revenue and Salary Act, 2 U.S.C. §§ 351-61, was passed by Congress. Pursuant to the authority delegated to him in the Act by Congress, the President fixed the salary of district judges at $40,000 and that of courts of appeals *280judges at $42,500 and submitted these adjustments to Congress. The Congress did nothing whatever regarding these salary adjustments, and, accordingly, they became law on March 15, 1969.10
Again, on January 17, 1977, the President, acting under the authority delegated to him in the Act by Congress, fixed and adjusted the salary of district judges at $54,500 and that of courts of appeals judges at $57,500, and on that day submitted such adjustments to Congress. Again, the Congress did nothing and these salary adjustments became law on March 1, 1977.11
The foregoing 1969 and 1977 adjustment of the salaries of judges by the President, which became law without Congress doing anything about them, is convincing proof that these salary adjustments were not mere "recommendations” by the President to Congress, but were the exercise of executive functions pursuant to the power and authority delegated to him and conferred upon him by Congress. Had these salary adjustments been mere recommendations, they would have died when Congress did nothing about them, and could never have become law automatically. Actually, these salary adjustments could be said to have the status of regulations issued by the heads of the various departments and other executive agencies of the government pursuant to statutes enacted by Congress. Such regulations have the force and effect of law without any action being taken regarding them by Congress. No one would contend that one House of the Congress could invalidate any of such regulations by a simple resolution.
The one-House veto is legislative in character, and, accordingly violates article I, section I of the Constitution which provides that "all legislative powers herein granted shall be vested in a Congress of the United States, which shall consist of a Senate and House of Representatives.” This provision deprives one House of the Congress of the power or authority to enact legislation. All legislation must *281be passed by both Houses of Congress. Otherwise, not only is article I, section I of the Constitution violated, but also article I, section 7, clause 3, which requires that "every Order, Resolution, or Vote to which the Concurrence of the Senate and House of Representatives may be necessary * * * shall be presented to the President of the United States; * * *” who may approve or veto the same. Obviously a one-House veto in the form of a resolution is not presented to the President for his approval or veto, and the Constitution is again violated.
If it could be assumed arguendo that the one-House veto is not legislative in character, then it would have to be a judicial or executive power. Clearly, it is not judicial. If it is neither legislative nor judicial, it is then executive. But neither House nor both Houses nor officials thereof can exercise executive power or functions. See Buckley v. Valeo, 424 U.S. 1 (1976); and Springer v. Philippine Islands, 277 U.S. 189 (1928). Furthermore, article II, section I of the Constitution provides that "the executive power shall be vested in a President of the United States of America.” It is clear that if the one-House veto is the exercise of executive power or function, it violates the above article of the Constitution, and violates the principle of Separation of Powers of our government.
The fixing of judges’ salaries by the President pursuant to a statute (the Act) enacted by both Houses of Congress that delegated this function and authority to him in the instant case was the exercise of an executive function. As Attorney General Brownell stated in a similar case:
It must be assumed that the functions of the President under this Act were executive in their nature or they could not have been constitutionally conferred upon him * * *. [37 Ops. Att’y Gen. 56 (1933)] [Emphasis supplied.]
As stated above, once this executive authority and function to fix judges’ salaries had been conferred upon the President and delegated to him by legislation enacted by both Houses of Congress, it could not be withdrawn nor taken from him except by the passage of a law of equal dignity so providing, and enacted by both Houses of Congress. This has not been done in the instant case, and *282the one-House veto was ineffectual to accomplish this purpose.12
Amici in this case appear to rely principally upon the so-called "necessary and proper clause” of the Constitution, article I, section 8, clause 18.13 The broad interpretation they give to this provision of the Constitution would authorize Congress to enact any and all kinds of legislation that suits its fancy so long as it does not conflict with an express provision of the Constitution. But as the plaintiffs point out, Congress can exclude aliens, but it could not admit aliens on condition that they agree not to exercise free speech. Further, Congress can confer jurisdiction upon the courts, but it cannot confer jurisdiction subject to the condition that judicial decisions must first be approved by the Senate Committee on the Judiciary. In other words, whether Congress can condition a delegation of authority depends upon the validity of that particular condition. This clause of the Constitution does not authorize Congress to ignore the bicameral provision of the Constitution, nor the necessity of submitting to the President every resolution to which the concurrence of both Houses are necessary, nor the provision that all executive authority is vested in the President. These provisions of the Constitution, in effect, prohibit the one-House veto, and the "necessary and proper” clause does not authorize Congress to ignore or transgress these fundamental provisions of our Constitution. No one can contend that the Constitution authorizes the one-House veto. A reasonable interpretation of its provisions points the other way.
SEVERABILITY
I now move to address the question of severability.14 The Government agreed during this case that the one-House *283veto is unconstitutional. However, the Government contends that nonetheless, plaintiffs’ case must fail because the one-House veto clause is inseverable from the rest of Chapter 11, Title 2, U.S.C.
Arguendo, were the majority to agree with the Government that the one-House veto is unconstitutional, (as does the dissent) the question then becomes: Does it follow that the entire Postal Revenue and Federal Salary Act of 1967 must be declared null and void? I believe such a result would be patently ludicrous.
I have argued above that the President’s 1973 recommendations for increases in judicial salaries have already taken effect. They are enforceable, provided Clause B is severable from the remainder of the Postal Revenue and Federal Salary Act.
The classic test of severability is whether (1) the valid provisions of the act are capable of standing alone, and (2) the legislature would have intended them to stand alone without the invalid provisions.15 Electric Bond & Share Co. v. SEC, 303 U.S. 419, 433-39(1938).
After carefully analyzing the arguments of both parties and all amici and having determined that the one-House veto is unconstitutional, I am of the opinion that not only is Clause B a legal nullity, but that it is also severable. It is as if Clause B never existed. The remainder of the Act would then continue in full force and effect, untouched and unchanged by the severance of Clause B.16
*284In determining whether the test of severability is met, courts have applied the rule laid down in Champlin Refining Co. v. Corporation Comm’n of Oklahoma, 286 U.S. 210, 224 (1932):
Unless it is evident that the legislature would not have enacted those provisions which are within its power, independently of that which is not, the invalid part may be dropped if what is left is fully operative as law.
See Buckley v. Valeo, 424 U.S. 1,108-09 (1976); United States v. Jackson, 390 U.S. 570,585 (1975). But see Carter v. Carter Coal Co., 298 U.S. 238,312 (1936). Moreover, the burden of proving inseverability is on the party claiming it. Moore v. Fowinkle, 512 F.2d 629,632 (6th Cir. 1975).
A brief recitation of the parties’ arguments in re severability will set the stage for the discussion of the issue.
Defendant argues that the absence of a severability clause in the Act at bar creates a presumption of nonseverability. Second, defendant contends that the legislative history would indicate Congress would have voted down the entire Postal Revenue and Federal Salary Act had it not contained the one-House veto (Clause B). Third, defendant says that removing Clause B would "end” Congressional control of judicial salaries.
Plaintiffs dispute the existence of a presumption in favor of nonseverability in the absence of a severability clause. Second, they maintain that the legislative history of the 1967 Act proves that the major issue was whether Congress should or should not delegate authority to set Congressional salaries, not judicial. Plaintiffs argue that Congress would not have killed the entire bill if the one-House veto clause had been excised, noting the comparatively minor importance it seemed to be given by the legislators.
Third, plaintiffs contend that Congress was perfectly willing to delegate authority to the President (advised by a commission) to set Government salaries without retaining veto power. Plaintiffs argue that this willingness counters the Government’s assertion that Congress viewed the existence of the one-House veto clause as essential. *285Plaintiffs further note that Congress retained other checks on the exercise of its delegated power — thereby not "ending” Congressional control of judicial salaries.
In sum, plaintiffs’ argument is aimed at convincing the court that the provisions of the Act (all but Clause B) are capable of standing alone, and that Congress would have intended them to stand alone. The Government maintains the opposite proposition.
1. Effect of the Absence of a Severability Clause
Defendant’s principal argument seems to be that the absence of a severability clause here creates a presumption of nonseverability. Prior to the 1920’s, the law on this matter was clear. Throughout the Nineteenth Century, the courts excised unconstitutional clauses from statutes, assuming, in view of the general presumption in favor of constitutionality, that "full effect will be given to such [provisions] as are not repugnant to the Constitution.” Bank of Hamilton v. Dudley, 27 U.S. (2 Pet.) 492,526 (1829). See Sturges v. Crowninshield, 17 U.S. (4 Wheat.) 122,208 (1819); Stern, Separability and Separability Clauses in the Supreme Court, 51 Harv. L. Rev. 76,79 (1937) and cases there cited. Robert Stern (the subject’s leading commentator) wrote of this early period:
[Whether or not a statute contained a severability clause] the courts were very reluctant to invalidate an entire law on grounds of inseparability. Doubts were resolved in favor of the severance of legislation; indeed, in a number of cases, in both state and federal courts, it was apparently assumed that laws were intended to be sustained to the extent they could possibly be held valid. Although the language of presumption was not employed, it is clear that the courts at that time presumed that laws were intended to be severable, rather than the contrary.
Id. at 81.
During the 1920’s and 1930’s, the Court began to suggest that the presence or absence of a severability clause in a statute led to a presumption of severability or of insevera-*286bility.17 These presumptions, and the cases involved, were examined with care by Robert Stern in 1937.
He concluded that first, the Court decisions on severability very much reflected the justices’ views on the merits of the constitutional claim. After examining the cases, Stern wrote of severability that:
[T]he Court is free to decide each case the way it pleases without having its discretion fettered by any restraining doctrine. * * * [T]he Court can easily hold any statute separable or inseparable, as it chooses. Id. at 111.
In this regard, it should be noted that language about a "presumption of inseverability” began to appear at a time when the Supreme Court reached out to strike down many New Deal legislative programs in their entirety as unconstitutional. See, e.g., Carter v. Carter Coal Co., 298 U.S. 238 (1936).
Stern found, secondly, that Court statements about a "presumption of inseverability” were uniformly dicta, saying:
[T]hese presumptions have only been availed of by the Court in construing statutes containing a separability clause .... The converse presumption of indivisibility . . . has not yet been given application.
Id. at 119.
Despite defendant’s arguments, no more recent cases have been found giving effect to the so-called presumption. In all cases that defendant cites, the "presumption” of indivisibility is dicta, for a separability clause was present.
Stern found, thirdly, that it was wrong to jump from the absence of a severability clause in a statute to a "presumption” of inseverability:
What has been said in no way indicates, however, that the presumption of indivisibility is justifiable. This presumption that, in the absence of a separability cláuse, laws are to be presumed incapable of separation can only be defended on the ground that legislatures avail themselves of separability clauses whenever they desire the provisions of a law to be severable, and fail to use the *287device only because they have a contrary intention. Whether legislatures exercise such nice discrimination is very doubtftil. Separability clauses have probably been employed when and because the draftsmen of a statute had doubts as to whether the courts would sustain the law in toto. The absence of such doubts, or careless draftsmanship, are more likely than the existence of an intention that the law be treated as inseparable to be responsible for the failure to incorporate such provisions in a statute.
Id. at 125.
Fourth, the earlier attitude, assuming:
[TJhat statutes were separable and that valid parts should stand unless the contrary was clearly shown, . . . would clearly seem to be the proper one; the courts should attempt to effectuate rather than to thwart legislative policies. The established — if often forgotten— presumption of constitutionality and the related doctrine that a statute should be given a constitutional construction in order to sustain it, provide close analogies.
Id. at 120.
Finally, Stern felt that given the press of events on legislative calendars, it would be unwise and impractical to force a legislature to reenact those portions of a statute that are not constitutionally infirm.
Stern’s article has been quoted at length because it not only indicates the historical antipathy towards presuming inseverability in legislative enactments, but it appears to reflect accurately the present state of the law. Since 1937, the "presumption of inseverability” has not been often used or even mentioned. In United States v. Jackson, 390 U.S. 570 (1968), for example, Jackson, as appellee, cited Electric Bond & Share Co. v. SEC, supra, trying to invoke what he termed the "settled presumption against severa-bility where there is no severability clause.” The Government refuted Jackson’s effort to use such a presumption, arguing that unless the legislature would not have enacted the statute as a whole absent an offending provision, the offending provision may be deleted and the remainder is left, fully operating.
The Court in Jackson (a case without a severability clause) accepted the Solicitor General’s argument and went *288further in undermining any presumption of inseverability. The Court wrote:
The appellees correctly note that Champlin was a case where Congress had included a clause expressly authorizing the severance of any invalid provision, a fact upon which this Court relied in recognizing "a presumption that, eliminating invalid parts, the legislature would have been satisfied with what remained . . . 286 U.S. 210, 235. But whatever relevance such an explicit clause might have in creating a presumption of severability, see Electric Bond & Share Co. v. Comm’n., 303 U.S. 419, 434, the ultimate determination of severability will rarely turn on the presence or absence of such a clause. Thus, for example, the Court in Champlin, after stating the basic test quoted above, cited cases in which invalid statutory provisions had been severed despite the absence of any provision for severability. Pollock v. Farmers’ Loan & Trust Co., 158 U.S. 601,635; Reagan v. Farmers’ Loan & Trust Co., 154 U.S. 362,395-96; Field v. Clark, 143 U.S. 649,695-96.
United States v. Jackson, 390 U.S. 570,585, n. 27.
In the recent case of Tilton v. Richardson, 403 U.S. 672,683-84 (1970), the Court went still further, dealing with the question of severability in the absence of a severability clause as follows:
To this extent the Act therefore trespasses on the Religion Clauses. The restrictive obligations of a recipient institution under § 751(a)(2) cannot, compatibly with the Religion Clauses, expire while the building has substantial value. This circumstance does not require us to invalidate the entire Act, however. 'The cardinal principle of statutory construction is to save and not to destroy. ’ NLRB v. Jones & Laughlin Steel Corp., 301 U.S. 1,30 (1937). In Champlin Rfg. Co. v. Commission, 286 U.S. 210,234 (1932), the Court noted:
'The unconstitutionality of a part of an Act does not necessarily defeat . . . the validity of its remaining provisions. Unless it is evident that the legislature would not have enacted those provisions which are within its power, independently of that which is not, the invalid part may be dropped if what is left is fully operative as a law.’
Nor does the absence of an express severability provision in the Act dictate the demise of the entire statute. E.g., *289United States v. Jackson, 390 U.S. 570,585 n. 27 (1968). (emphasis added)
In the still more recent case of Buckley v. Valeo, 424 U.S. 1,108-09 (1976), the Court excised unconstitutional portions from a statute without apparent regard to whether or not a severability clause was applicable. The Court referred to the above-cited language from Champlin Refining and no more.
' Severability clause or not, Champlin Refining states the basic law: "Unless it is evident that the legislature would not have enacted those provisions which are within its power, independently of that which is not,” the clause is severable. Such a rule makes good sense for the reasons Stern pointed out. Even in the absence of a severability clause, Champlin Refining, at the least, imposes upon the Government the burden of showing that "it is evident” that Congress would not have enacted the entire Postal Revenue and Federal Salary Act of 1967 had it known the one-House veto clause would be held invalid.
Finally, this court’s decision just a few months ago, in Gentry v. United States, supra, provides further authority for discounting the Government’s argument that there exists a presumption against inseverability.
Thus it seems clear, and I would so hold, that there is no automatic presumption of inseverability just because of the absence of a specific severability clause.
2. The Legislative History
The defendant argues further that the legislative history indicates Congress would have voted down the entire Postal Revenue and Federal Salary Act of 1967 had it not contained the one-House veto. To this proposition, plaintiffs offer strong objection.
Plaintiffs maintain that Champlin Refining imposes upon the Government the burden of showing that "it is evident” that Congress would not have enacted the Postal Revenue and Federal Salary Act of 1967 had it known the one-House veto clause would be held invalid. Yet it is far from "evident” that Congress would have voted down the Act — without Clause B.
*290A reading of the entire legislative history18 shows that the one-House veto had little importance in the debate. Rather, the major issue concerned the creation of a salary commission — i.e., whether or not Congress should delegate, if at all, the power to set Congressional salaries. The House strongly favored the idea — the Senate strongly opposed it. At the last minute the Senate gave way (in return for concessions on junk mail rates under other provisions of the same Act.)19 The major Senate concern was that Members of Congress stand up and be counted when their own salaries were at issue. This concern was met adequately in other ways.20
The Government has culled from the debates those comments showing support for Clause B. That such comments appear in the record is hardly surprising; after all, Congress voted for an Act which included this provision. The issue, however, is not whether some Congressmen favored it, or even that it was voted for, but whether it was essential to secure passage of the Act. The comments that the Government quotes fail to make this showing.
First, an examination of the Government’s citations to the legislative history in the House of Representatives is in order. None of the remarks quoted by defendant suggest that Clause B is essential to meet any of the objections raised to § 225, much less necessary to insure passage of the Act. Mr. Udall21 suggests in those remarks that fears of discriminatory behavior by the President, raised by Mr. Gross,22 are implausible (which they are), involve unconstitutional action by the President (which they probably do), and in any event, the Congressional veto weapon is available. Nor do any of the quoted comments suggest that *291Clause A would not give Congress the "final gun.” In fact, neither Mr. Udall nor anyone else distinguished between a one-House veto (Clause B) and a bill that passes both Houses (Clause A). Mr. Udall, for example, states:
The final gun will always be in the hands of Congress. We can always adopt language or a statute inconsistent with what the Commission decides. 113 Cong. Rec. 28643.
Indeed, why should anyone have cared strongly about the difference? The issue was Congressional salaries; it is politically inconceivable that the President would try to force a salary increase upon a Congress that had voted for a bill rejecting it.
The comments in defendant’s brief offer slim grounds for finding that the House of Representatives felt that Clause B was an essential part of § 225, and offer no basis for the proposition that this minor provision was needed to secure passage of Pub.L. 90-206.
Turning now to the legislative history as evidenced by the Senate, it is again apparent that the Government has simply culled those comments from the debates that speak favorably of Clause B. A scrutiny of the entire debate in the Senate, however, shows that Clause B was not essential to meet any major concern expressed during that debate.
The Senate was hostile to any delegation of Congressional salary authority. Its major concern was that legislators would thereby avoid standing up and being counted as for or against increases in their own pay. To delegate the power to set pay might allow legislators to escape the responsibility.
This expressed concern was met in two ways. First, an amendment by Senator Byrd23 was passed to the Reorganization Act which thereby allowed the Senate to consider part of any matter before it and to vote on each part separately. This amendment would prevent Senators from hiding a vote increasing their own pay by voting generally on the subject (113 Cong. Rec. 36100,36101). Second, the Senate/House Conference removed from § 225 the power to set "expenses and allowances, and requirements, conditions, and related matters.” This excision was designed to *292prevent a Senator from hiding a vote for a salary increase by voting for a wage/expense package. The basic idea was to single out Congressional wages and force legislators to vote on them directly. This need was met adequately by the provisions for delaying the effectivenéss of the President’s recommendation together with the Byrd Amendment and Clause A. Clause B is not essential.
Of course, it might be argued, from a political point of view, that the Senate strongly opposed delegation and nearly anything increasing the scope of the delegation would have led to the Act’s defeat. Indeed, the Government’s argument is plausible only insofar as it implicitly draws upon such an assumption. Yet, the legislative history does not support this view. Moreover, to use this political speculation to support a conclusion of inseverability (instead of analyzing whether Clause B was essential to the passage of the Act as a whole) would be contrary to established law on this issue. It would make the question of severability turn, not on the relation of a clause to the legislative scheme (i.e., if it is essential to the passage of the Act), but on the closeness of the vote — on political considerations within the legislature, the actual facts of which (quite properly) remain unknown to a reviewing court.
The absurdity of this result (striking down the entire Act due to Clause B’s unconstitutionality and argued insevera-bility) dictates that the basic question to be asked in determining severability is not a political one, but a legal one: Whether Clause B was essential to the basic purposes of the Act’s supporters. The presence of Clause A, the Byrd Amendment, the delay before the President’s recommendations take effect, together with the legislative history (showing that the basic argument was about delegation itself and the need for legislators to reveal publicly their positions on legislative salaries), all suggest that Clause B was not essential.
I believe that the legislative history shows nothing to indicate that subsection (i)(l)(B) of Section 225 was in any respect essential to the passage of the full Act, Pub.L. 90-206. If it shows anything, it is that § 225 (and the commission concept embodied therein) was far more *293critical to the passage of Pub.L. 90-206 than was Clause B to the passage of § 225. In explaining why the Senate/House Conference Committee version of the bill contained § 225, Senator Monroney24 stated to his colleagues:
May I say that from the first day until the last day, and almost every hour on the hour, the conferees on the House side tried to keep the junk mail rate at a lower rate . . . They also insisted on the House passed version of the Presidential Commission. We were just as adamant. In fact, ... I Was reminded of the old law of physics that . . . when an irresistible force meets an immovable object, something has to give. Up to the last five minutes I can say advisedly that the House position was unanimous for the junk mail rate they had put in and the Senate position was unanimous in having this commission taken out. We got to such a trading point on the junk mail that to get a bill or not rested on whether we had to lower junk mail rates . . . The House conferees gave in on that and they gave in on the striking out the objectionable language that . . . would force us to vote for the whole package [the Byrd amendment to the Reorganization Act]. . .
The conference committee was within 5 minutes of a breakup, of picking up our papers and walking out, and there would have been no pay increase for the 3 million Federal workers, including members of our own staffs and for the widows and orphans of those who die between Christmas and 2 months later, when we might resume the conference. It was the only way, as I said before; it was the only way to give when an irresistible force was colliding with an immovable object. (113 Cong. Rec. 36102) (emphasis added)
Thus, if the Government is right in believing that the politics of the situation were such that, without Clause B, there would have been no commission, it is even more plain that without the commission, there would have been no Postal Revenue and Federal Salary Act of 1967 (and, in particular, no increased rates for junk mail).
To hold that Clause B is inseverable is to believe that Congress would rather have had the entire Act struck down than to have it remain without the Clause. Given the *294consequences of each of these alternatives, it is difficult to believe that Congress would have preferred the former. To risk having the entire Act struck down is to risk casting a legal shadow over the paychecks previously sent to many thousands of postal employees (and possibly over postal rates as well).
In view of the known Constitutional doubts about one-House vetoes, Congress could not have preferred the above mentioned risks to a holding of severability. I see little merit in defendant’s argument that legislative history would indicate that Congress would not have intended the valid provisions of the Act to stand alone without the invalid provisions.
3. Removal of Congressional Control
The defendant says that removing Clause B would "end” Congressional control of judicial salaries. The answer is brief and clear. Plaintiffs emphasize that any excising of Clause B from the Act leaves Clause A intact and operative. They are correct. 2 U.S.C. 359(1)(A) reads:
[Recommendations of President . . . effective . . . but only to extent that . . .] (A) There has not been enacted into law a statute which establishes rates of pay other than those proposed by all or part of such recommendations.
The Constitution would not protect the judicial salary recommended by the Commission and the President until after Congress had an opportunity to pass a bill in accordance with the procedure legally provided by Clause A. Thus, even absent Clause B, Congress still maintains its control over judicial salaries. Defendant’s argument cannot be upheld.
4. Congress Willing Delegation
Plaintiffs, in the course of disputing the Government’s contention that the existence of the one-House veto clause was essential to the passage of the Act as a whole, point out that Congress has proved perfectly willing to delegate authority to the President (advised by a commission) to set *295government salaries without retaining veto power. In the Federal Pay Comparability Act of 1970, 84 Stat. 1946 (codified in 5 U.S.C. § 5305 (1970)), Congress provided that if the President’s salary recommendations for general schedule civil servants — the vast majority of all civil servants — are the same as those of the advisory commission, they will take effect without Congressional review.
For the past six years, automatic annual increases in the salary levels of most federal employees have been provided through Executive action alone under the Federal Pay Comparability Act of 1970, supra. In 1975, this benefit was extended to all remaining federal employees, including judges, through enactment of the Executive Salary Cost-of-Living Adjustment Act.25 Except for the power lawfully to enact a statute effecting rates of pay (which Congress may do pursuant to § 359(1)(A)) and the power to refuse to appropriate (which cannot constitutionally be exercised as to federal judges’ salaries in light of Article III, Section 1), Congress has shown a clear willingness to yield its authority regarding federal salaries. Thus, the basis alleged by defendant for holding § 359(1)(B) unseverable is without substance.
The recent 1977 pay raise in the salaries of federal judges (among others) is an example of Congress’s willingness to delegate its authority regarding federal salaries. This pay raise, discussed supra, went into effect unchanged by Congress. The Commission’s recommendations were lowered by the President and his recommendations took effect, per the statutory scheme, without intervening action by Congress. If, as Senate amicus maintains, Congress had not intended to "yield its authority” in this matter, Congress has not manifested such intent.
5. Summary
I feel that this court might well be guided by the way in which other one-House veto provisions have been treated. Although there are no judicial decisions directly on point, *296the actions of the Executive branch provide highly relevant precedent for the following reason:
Many Bills containing one-, or two-House vetoes have been enacted into law over the signature of a President who believed that such a clause was unconstitutional. Before signing such a bill, the President would have had to determine the severability question. If the clause was, in the President’s view, severable, the President would be signing an act that was in itself constitutional, but which contained one unconstitutional clause. It would be reasonable to sign such a bill despite the offending provision. If the clause were not severable, however, the President would have had to consider signing a bill which he would believe unconstitutional in its entirety. It is unlikely that a President would sign such a bill.
One wonders why President Johnson (who, like other Presidents, believed that one-House vetoes were unconstitutional),26 would have signed the Act here at issue into law unless he believed that Clause B was severable, for it seems unlikely that he would have signed this Act believing that the Act was unconstitutional in its entirety.
In summary, I am of the opinion that the valid provisions of the Act are capable of standing alone. The legislature would have intended them to stand alone without the invalid provisions. The burden of proving inseverability is on the party claiming it; the Government has failed to meet its burden.
I would hold that Clause B is severable.
Jurisdiction and Prospectivity
Two subsidiary points remain to be considered if plaintiffs are to recover.
The first concerns defendant’s contention that this court is without jurisdiction due to the absence of a statute that "can fairly be interpreted as mandating compensation by the Federal Government for the damage sustained.” United States v. Testan, 424 U.S. 392,400 (1976), quoting Eastport Steamship Corp. v. United States, 178 Ct.Cl. *297599,607, 372 F.2d 1002, 1009 (1967). Defendant’s position rests principally on the argument that Clause B is not severable from the rest of the Act. If Clause B were not severable, says the Government, upon a further finding that Clause B was unconstitutional, the entire Act would fall, including that part mandating payment, 2 U.S.C. § 359(1)27
However, since I would hold Clause B severable, supra, I find defendant’s lack of jurisdiction argument without merit. Absent Clause B, the remainder of the Act still stands, and the pertinent part, now codified at 2 U.S.C. § 359(1), mandates payment. See United States v. Testan, supra.
Further, I find no merit in defendant’s argument that the court cannot grant relief if a portion of the Act must first be declared unconstitutional. Such a proposition is directly contrary to our holding in Gentry v. United States, 212 Ct. Cl. 1, 546 F.2d 343 (1976), and totally foreign to the jurisprudence of this court. In Gentry, plaintiff, an illegitimate minor child, sought benefits under the Civil Service Retirement Act of 1930, as amended. 5 U.S.C. § 8331, et seq. (1970). In deciding for plaintiff, the court held the "live-in” requirement of 5 U.S.C. § 8341(a)(2)(A)(ii) unconstitutional and then stated that the remainder of the "statute must be construed, and payments made, without regard to the [unconstitutional live-in] requirement.” Gentry v. United States, Id at 22, 546 F.2d at 347. See Frontiero v. Richardson, 411 U.S. 677, 680 (1973).
The second point concerns defendant’s contention that, if the court holds the one-House veto unconstitutional and severable (as I would), its decision should be applied prospectively only. The Government thus would deny applicability to these very plaintiffs of a decision favoring recovery in this case. This would be a prospective overrul*298ing foreclosing victory to the very parties who originally brought the suit.
The Government would have us hold the one-House veto validly applied against plaintiffs in this case and, thereby, deny them relief. Defendant, then, is urging a certain species of prospective overruling, one that "does not apply even to the parties before the court.” Linkletter v. Walker, 381 U.S. 618,621-22 (1965); England v. Louisiana State Board of Medical Examiners, 375 U.S. 411,422 (1964). Two grounds are asserted for this proposition.
Initially, defendant argues that Congressional reliance on Clause B should preclude the court from applying a decision of unconstitutionality to the instant case. Reliance is found in Congress’s use of Clause B to deny effect of the President’s 1973 recommended pay raise. Defendant states that Congress’s reliance was justifiable, and that to decide otherwise is inequitable. See Halliday v. United States, 394 U.S. 831 (1969); Tehan v. United States, 382 U.S. 406 (1966); England v. Louisiana State Board of Medical Examiners, supra.
To speak of "Congress’s reliance” is to refer only to the fact that Congress, had it known of Clause B’s unconstitutionality, might (or might not) have passed another statute. Such is always true when a statute is held unconstitutional. To make such reliance a basis for prospective overruling would make such overruling the norm instead of the exception — undermining a basic distinction between court and legislature.
To argue, as defendant does, that Congress would have used Clause A in 1973 had it known that Clause B would later be declared unconstitutional merely states the obvious. I think it would be safe to say Congress would always use a constitutional method of achieving its ends when the alternative is to place reliance on an unconstitutional method. To make the reliance defendant urges the basis for determination of prospectivity would make prospectivity the general rule. But the Supreme Court has stated that neither prospective nor retroactive application is automatic; the effect of a constitutional decision must be determined " 'in each case’ by looking to the particular *299traits of the executive rule in question.” Johnson v. New Jersey, 384 U.S. 719,728 (1966).
Nor do I agree with defendant’s prospectivity argument that it would be inequitable in the instant case to apply a holding that Clause B is unconstitutional. Defendant appears to argue that such an application would be inequitable to Congress. In fact, however, to deny plaintiffs relief through a prospective-only decision would cause the greatest inequity. A prospective-only decision would deny relief to the very plaintiffs who brought this suit, and would have the broader effect of discouraging other plaintiffs from challenging unconstitutional statutes.
Moreover, a holding of unconstitutionality should come as no surprise to Congress. The validity of the one-House veto has been the subject of great controversy, both in Government and in scholarly circles. See Part I of this dissent, supra. See also Berger v. California, 393 U.S. 314 (1969); Roberts v. Russell, 392 U.S. 293 (1968). I would not find Congress’s reliance on the supposed constitutionality of Clause B persuasive.
As a secondary argument for prospectivity, defendant contends that plaintiffs should not be allowed to attack a statute as unconstitutional while simultaneously seeking to retain benefits received under it. In the instant case, defendant claims that plaintiffs should not be permitted to challenge the constitutionality of the Act since they received the benefit of a pay increase in 1969.
The Government relies on Arnett v. Kennedy, 416 U.S. 134,152-54 (1974) (Opinion of Rehnquist, J.). The opinion, in turn, is based on cases which are clearly distinguishable from the instant situation. For example, cases cited in Justice Brandeis’s concurrence in Ashwander v. TVA, 297 U.S. 288,348 (1936) all appear to involve a situation where plaintiffs attacked an entire legislative scheme while seeking to retain benefits conferred by the same legislation. In the case at bar, plaintiffs attack only a single clause of an Act, and the clause has not benefited plaintiffs — to the contrary, it has been used to deny them benefits otherwise available.
Even if defendant’s argument were on point, I can find no recent case which upholds it. In fact, defendant’s chief *300authority, Arnett v. Kennedy, supra, applies more to questions of standing and justiciability than to the issue of prospectivity.28 And, Frontiero v. Richardson, 411 U.S. 677 (1973), presenting an analogous situation to the one at issue here, did not even mention defendant’s proposition.
In summary, defendant’s argument that the court is without jurisdiction and that its decision should be prospective only are both without merit. I would hold that the court has jurisdiction to decide this case and that its decision should grant plaintiffs the relief they seek.
CONCLUSION
In the case before us, we find that on February 4, 1974, the salary of district judges was $40,000 and that of judges of courts of appeals was $42,500. On that date, after receiving the report of the salary commission which recommended a flat increase of 25 percent, the President, acting pursuant to the authority delegated to him by the Act, as he did in 1969 and 1977, fixed the salaries of such judges by increasing them seven and one-half percent as of March 1974; seven and one-half percent as of March 1975; and seven and one-half percent as of March 1976. These adjustments were sent to Congress by the President on February 4, 1974. Such adjustments set the salary of district judges at $43,000 in 1974, $46,200 in 1975, and $49,700 in 1976, and fixed the salary of judges of courts of appeals at $45,700 in 1974, $49,100 in 1975, and $52,800 in 1976. Notwithstanding the exercise of this executive function by the President, the Senate adopted the following resolution on March 6, 1974:
*301Senate Resolution 293
March 6, 1974
Resolved, That the Senate disapproves all the recommendations of the President with respect to rates of pay transmitted to the Congress in the budget for the fiscal year 1975 pursuant to section 225 (h) of the Federal Salary Act of 1967.
Congressional Record
Daily Edition at S2899-S2900
March 6, 1974
This resolution operated as a one-House veto, as the House of Representatives took no action. Consequently, the judges have not received the increased salaries as fixed by the President on February 4, 1974, and the salary of district judges remained at $40,000 and that of judges of courts of appeals at $42,500, except for the five percent cost-of-living increase they received October 1, 1975, which raised the salary of district judges to $42,000 and that of judges of courts of appeals to $44,600. Their salaries remained at these levels until adjusted by the President as aforesaid effective March 1, 1977, when the salary of district judges was raised to $54,500 and that of judges of courts of appeals to $57,500.
The resolution (293) of the Senate of March 6,1974, being a one-House veto of the executive function of the President was a violation of the Constitution and therefore constitutionally impermissive for all of the reasons stated herein, and was totally ineffective in law to invalidate the adjustment of judges salaries by the President. The Supreme Court stated in Norton v. Shelby County, 118 U.S. 425, 442 (1886) in considering an unconstitutional statute:
* * * [I]t confers no rights; it imposes no duties; it affords no protection; it creates no office; it is, in legal contemplation, as inoperative as though it had never been passed.
The one-House veto of the Senate by its adoption of Senate Resolution 293, being totally ineffective to invalidate the President’s adjustments of salaries, such adjustments became law as they did in 1969 and 1977, and by reason thereof 2 U.S.C. § 359 mandates the *302payment to the plaintiffs of the back salaries they seek in this case. They have a clear statutory claim to their claims under Count II of their petitions and judgment should be rendered in their favor for the difference between the salaries they received from March 1974, to March 1977, and what they would have received had it not been for the one-House veto by the Senate. I would enter judgment to this effect.
The plaintiffs seek interest on the amounts due them, but this claim must be denied. We held in an exhaustive opinion on this question in United States v. Mescalero Apache Tribe, 207 Ct. Cl. 369, 518 F.2d 1309 (1975), cert. denied, 425 U.S. 911 (1976), that interest cannot be awarded in a claim against the government in non-eminent domain cases unless a contract, treaty, or statute so provides.
Accordingly, I would enter judgment allowing plaintiffs’ motion for summary judgment as to Count II and denying it as to Count I, and deny defendant’s motion to dismiss as to Count II, but allow it as to Count I, and remand the case to the trial judge to determine the amount due the plaintiffs in proceedings under Rule 131(c).
§ 359. Same; effective date
(1) Except as provided in paragraph (2) of this Section all or part (as the case may be) of the recommendations of the President transmitted to the Congress in the budget under section 358 of this title shall become effective at the beginning of the first pay period which begins after the thirtieth day following the transmittal of such recommendations in the budget; but only to the extent that, between the date of transmittal of such recommendations in the budget and the beginning of such first pay period—
(A) there has not been enacted into law a statute which establishes rates of pay other than those proposed by all or part of such recommendations,
(B) neither House of the Congress has enacted legislation which specifically disapproves all or part of such recommendations, or
(C) both.
(2) Any part of the recommendations of the President may, in accordance with express provisions of such recommendations, be made operative on a date later than the date on which such recommendations otherwise are to take effect. Pub.L. 90-206, Title n, § 225(i), Dec. 16, 1967, 81 Stat. 644.
"Article I, Section 1. Legislative power vested in Congress
All legislative Powers herein granted shall be vested in a Congress of the United States, which shall consist of a Senate and House of Representatives.”
"Article I, Section 7, Clause 3. Approval or veto of orders, resolutions, or votes; repassage over veto
Every Order, Resolution, or Vote to which the Concurrence of the Senate and House of Representatives may be necessary (except on a question of Adjournment) shall be presented to the President of the United States; and before the Same shall take Effect, shall be approved by him, or being disapproved by him, shall be repassed by two thirds of the Senate and House of Representatives, according to the Rules and Limitations prescribed in the Case of a Bill.”
"Article II — The President
Section 1. The executive Power shall be vested in a President of the United States of America.”
S. Rep. No. 1335, 54th Cong., 2d Sess. 8 (1897).
For a collection of those statutes, see Watson, Congress Steps Out: A Look at Congressional Control of the Executive, 63 Calif. L. Rev. 983 (1975).
See messages of President Wilson, 59 Cong. Rec. 7026-27, 8069 (1920), and 58 Cong. Reo. 8074 (1919).
Wilson: 59 CONG. REC. 7026-27, 8609 (1920); Hoover: 76 CONG. REC. 2445 (1933); Roosevelt: H.R. DOC. NO. 252, 75th Cong., 1st Sess. (1937); 83 CONG. REC. 4487 (1938); 90 CONG. REC. 6145 (1944); Jackson, A Presidential Legal Opinion, 66 HARV. L. REV. 1353,1357-58 (1953) (Publishing a confidential opinion by President Roosevelt); Truman: 97 CONG. REC. 5374-75 (1951); 98 CONG. REC. 9756 (1952); Eisenhower: 100 CONG. REC. 7135 (1954); 101 CONG. REC. 10459-60 (1955); Public Papers Of The Presidents Of The United States: Dwight D. Eisenhower at 648-50 (1956); H.R. DOC. NO. 255 Pt. 1, 86th Cong., 2d Sess. at M18 (1960); Kennedy: Public Papers Of The Presidents Of The United States: John F. Kennedy at 6 (1964); Johnson: Public Papers Of The Presidents Of The United States: Lyndon B. Johnson, 1963-64, at 861-62, 1249-51 (1965); 1 Weekly Comp. Pres. Docs. 132, 432-33 (1965); 111 CONG. REC. 12639-40 (1965); Nixon: 8 Weekly Comp. Pres. Docs. 938, 1076 (1972); 9 Weekly Comp. Pres. Docs. 1285-87 (1973); Ford: 10 Weekly Comp. Pres. Docs. 1279 (1974). These statutes have also been the subject of several opinions by the Attorney General. 37 Op. Att/y Gen. 56 (1933); 41 Op. Att’y Gen. 230 (1955); Id. at 300 (1957).
See, e.g., 76 CONG. REC. 3539 (1933) (Senator Byrnes); 84 CONG. REC. 2477-78 (1939) (Representatives Taber and Wolcott); 83 CONG. REC. 3090 (1939) (Senator King); 87 CONG. REC. 1100, 1245, 1269, 1569, 2063-75 (1941) (Senators Clark, Gillette, McCarran, and Murdock); 87 CONG. REC. 6504-07 (1941) (Senators Adams and Taft); 97 CONG. REC. 5443 (1951) (Representative Patman); 100 CONG. REC. 5095 (1954) (Senator Dirksen). See Watson at 988 (my n. 4.)
Assistant Attorney General Lee defended the case on inseverability and other grounds discussed infra.
Article I, section 9, clause 7 of the Constitution; Miles v. Graham, 268 U.S. 501, 508-09 (1925); and O’Malley v. Woodrough, 307 U.S. 277 (1939).
The Executive Salary Cost-of-Living Adjustment Act enacted by Congress on August 9, 1975, increased district judges’ salaries to $42,000 and that of judges of courts of appeals to $44,600, effective October 1, 1975.
While the 1977 adjustments are not in evidence or in the record here, we can take judicial notice of them.
Legal scholars have criticized the one-House veto. See Ginnane, The Control of Federal Administration by Congressional Resolutions and Committees, 66 HARV. L. REV. 569 (1953); and Watson, Congress Steps Out: A Look at Congressional Control of the Executive, 63 CAL. L. REV. 983 (1975).
"The Congress shall have power * * * to make all laws which shall be necessary and proper for carrying into execution the foregoing powers, and all other powers vested by this Constitution in the Government of the United States, or in any Department or Officer thereof.”
In Gentry v. United States, 212 Ct. Cl. 1, 9, 546 F.2d 343, 347 (1976), this court *283adopted the practice of considering severability before constitutionality in cases such as the instant one "because of the nature of this court’s jurisdiction.” In the instant case, since I would hold Clause B severable, and the remaining portion mandates payment, (see infra) no jurisdictional problems are raised. For the convenience of the reader, then, I treat constitutionality and severability in the "normal” order.
When considering the severability of a statutory provision, the issue generally centers on whether the Act, of which that provision is a part, would have been enacted absent the constitutionally infirm provision. In its briefs, the Government argues that the one-House veto provision under consideration is not severable from the remainder of "Chapter 11, Title 2, U.S.C.,” as if "Chapter 11” were the Act of which the veto provision is a part. Quite simply, Chapter 11 is not an Act of Congress. It is merely a codification (2 U.S.C. § 351-361) of a part of an act. Chapter 11 is § 225 of Public Law 90-206: The Postal Revenue and Federal Salary Act of 1967, 81 Stat. 613, et. seq. 2 U.S.C. § 359(1)(B) — the unconstitutional provision under consideration — is § 225(i)(l)(B) of that Act. It is referred to as Clause B for the sake of convenience. What is significant, however, is that § 225 ("Chapter 11”) is not a separate act, but merely a part of Pub. L. No. 90-206.
It is, of course, unnecessary for the majority to reach the question of severability *284since it holds the one-House veto to be constitutional.
Electric Bond & Share Co. v. SEC, 303 U.S. 419,434 (1938); Carter v. Carter Coal Co., 298 U.S. 238,312 (1936); Railroad Retirement Board v. Alton Railroad Co., 295 U.S. 330,362 (1935); Utah Power & Light Co. v. Pfost, 286 U.S. 165,184 (1932).
The relevant reports are contained in the Appendix to Plaintiffs’ Summary Judgment Brief. Excerpts from the floor debates are contained in attachments to that brief and to defendant’s main brief. The floor debates are reported in their entirety at the following pages of 113 Cong. Rec.: 27288, 28406, 28409, 28411, 28612, 28641-4, 28672, 28879, 33590, 33972, 34013, 34014, 34022-3, 34212-5, 34220-2, 34236, 34241, 34255, 34261, 34410, 35524, 35811, 35822, 35839-41, 36088, 36100-8.
113 Cong. Rec. 36102.
In particular, by means of the Byrd Amendment, Clause A, and striking language from the House version. See discussion, infra.
Honorable Morris K. Udall (2d Dist., Arizona).
Honorable H. R. Gross (3d Dist., Iowa).
Honorable Harry F. Byrd, Jr. (Virginia).
Honorable A. S. Mike Monroney (Oklahoma).
89 Stat. 419-423 (codified in various sections of Titles 2, 5,11, 28, 31, 40, and 44, U.S.C.).
See, e.g., Pub. Papers: Lyndon B. Johnson, 1963-64 at 861-62; 1 Weekly Comp, of Pres. Doc. 132,432-33 (1965); 111 Cong. Rec. 12639-40 (1965).
2 U.S.C. § 359(1) (1970) reads:'
Except as provided in paragraph (2) of this section, all or part (as the case may be) of the recommendations of the President transmitted to the Congress in the budget under section 358 of this title shall become effective at the beginning of the pay period which begins after the thirtieth day following the transmittal of such recommendations in the budget;. . .
Defendant’s prospectivity argument appears to be moot in view of the enactment of Pub. L. No. 95-19 by Congress on April 12, 1977, which repealed the "one-House veto” provision (Clause B) of the statute involved here, and which requires all future salary adjustments made by the President to be voted on by both Houses. However, the new statute does not affect the instant cases as they are suits for back pay. With relation to the question of justiciability, we reemphasize that we concur with the analysis of the majority at Part "B” of its discussion of Count I, supra.