delivered the opinion of the Court.
The question presented is whether a Federal District Court has jurisdiction under 28 U. S. C. § 1346 (a)(1) of a suit by a taxpayer for the refund of income tax payments which did not discharge the entire amount of his assessment.
This is our second consideration of the case. In the 1957 Term, we decided that full payment of the assessment is a jurisdictional prerequisite to suit, 357 U. S. 63. Subsequently the Court granted a petition for rehearing. 360 U. S. 922. The case has been exhaustively briefed and ably argued. After giving the problem our most careful attention, we have concluded that our original disposition of the case was correct.
Under such circumstances, normally a brief epilogue to the prior opinion would be sufficient to account for our decision. However, because petitioner in reargument has placed somewhat greater emphasis upon certain contentions than he had previously, and because our dissenting colleagues have elaborated upon the reasons for their *147disagreement, we deem it advisable to set forth our reasoning in some detail, even though this necessitates repeating much of what we have already said.
The Facts.
The relevant facts are undisputed and uncomplicated. This litigation had its source in a dispute between petitioner and the Commissioner of Internal Revenue concerning the proper characterization of certain losses which petitioner suffered during 1950. Petitioner reported them as ordinary losses, but the Commissioner treated them as capital losses and levied a deficiency assessment in the amount of $28,908.60, including interest. Petitioner paid $5,058.54 and then filed with the Commissioner a claim for refund of that amount. After the claim was disallowed, petitioner sued for refund in a District Court. The Government moved to dismiss, and the judge decided that the petitioner “should not maintain” the action because he had not paid the full amount of the assessment. But since there was a conflict among the Courts of Appeals on this jurisdictional question, and since the Tenth Circuit had not yet passed upon it, the judge believed it desirable to determine the merits of the claim. He thereupon concluded that the losses were capital in nature and entered judgment in favor of the Government. 142 F. Supp. 602. The Court of Appeals for the Tenth Circuit agreed with the district judge upon the jurisdictional issue, and consequently remanded with directions to vacate the judgment and dismiss the complaint. 246 F. 2d 929. We granted certiorari because the Courts of Appeals were in conflict with respect to a question which is of considerable importance in the administration of the tax laws.1
*148The Statute.
The question raised in this case has not only raised a conflict in the federal decisions, but has also in recent years provoked controversy among legal commentators.2 In view of this divergence of expert opinion, it would be surprising if the words of the statute inexorably dictated but a single reasonable conclusion. Nevertheless, one of the arguments which has been most strenuously urged is that the plain language of the statute precludes, or at the very least strongly militates against, a decision that full payment of the .income tax assessment is a jurisdictional condition precedent to maintenance of a refund suit in a District Court. If this were true, presumably we could but recite the statute and enter judgment for petitioner— though we might be pardoned some perplexity as to how such a simple matter could have caused so much confusion. Regrettably, this facile an approach will not serve.
Section 1346 (a)(1) provides that the District Courts shall have jurisdiction, concurrent with the Court of Claims, of
“(1) Any civil action against the United States for the recovery of any internal-revenue tax alleged to have been erroneously or illegally assessed or collected, or any penalty claimed to have been collected *149without authority or any sum alleged to have been excessive or in any manner wrongfully collected under the internal-revenue laws . . . (Emphasis added.)
It is clear enough that the phrase “any internal-revenue tax” can readily be construed to refer to payment of the entire amount of an assessment. Such an interpretation is suggested by the nature of the income tax, which is “A tax . . . imposed for each taxable year,” with the “amount of the tax” determined in accordance with prescribed schedules.3 (Emphasis added.) But it is argued that this reading of the statute is foreclosed by the presence in § 1346 (a)(1) of the phrase “any sum.” This contention appears to be based upon the notion that “any sum” is a catchall which confers jurisdiction to adjudicate suits for refund of part of a tax. A catchall the phrase surely is; but to say this is not to define what it catches. The sweeping role which petitioner assigns these words is based upon a conjunctive reading of “any internal-revenue tax,” “any penalty,” and “any sum.” But we believe that the statute more readily lends itself to the disjunctive reading which is suggested by the connective “or.” That is, “any sum,” instead of being related to “any internal-revenue tax” and “any penalty,” may refer to amounts which are neither taxes nor penalties. Under this interpretation, the function of the phrase is to permit suit for recovery of items which might not be designated as either “taxes” or “penalties” by Congress or the courts. One obvious example of such a “sum” is interest. And it is significant that many old tax statutes described the amount which was to be assessed under certain circumstances as a “sum” to be added to the tax, simply as a *150“sum,” as a “percentum,” or as “costs.” 4 Such a rendition of the statute, which is supported by precedent,5 frees the phrase “any internal-revenue tax” from the qualifications imposed upon it by petitioner and permits it to be given what we regard as its more natural reading — the full tax. Moreover, this construction, under which each phrase is assigned a distinct meaning, imputes to Congress a surer grammatical touch than does the alternative interpretation, under which the “any sum” phrase completely assimilates the other two. Surely a much clearer statute could have been written to authorize suits for refund of any part of a tax merely by use of the phrase “a tax or any portion thereof,” or simply “any sum paid under the internal revenue laws.” This Court naturally does not review congressional enactments as a panel of grammarians; but neither do we regard ordinary principles of English prose as irrelevant to a construction of those enactments. Cf. Commissioner v. Acker, 361 U. S. 87.
We conclude that the language of § 1346 (a)(1) can be more readily construed to require payment of the full tax before suit than to permit suit for recovery of a part *151payment. But, as we recognized in the prior opinion, the statutory language is not absolutely controlling, and consequently resort must be had to whatever other materials might be relevant.6
Legislative History and Historical Background.
Although frequently the legislative history of a statute is the most fruitful source of instruction as to its proper interpretation, in this case that history is barren of any clue to congressional intent.
The precursor of § 1346 (a)(1) was § 1310 (c) of the Revenue Act of 1921,7 in which the language with which we are here concerned appeared for the first time in a jurisdictional statute. Section 1310 (c) had an overt purpose unrelated to the question whether full payment of an assessed tax was a jurisdictional prerequisite to a suit for refund. Prior to 1921, tax refund suits against the United States could be maintained in the District Courts under the authority of the Tucker Act, which had been passed in 1887.8 Where the claim exceeded $10,000, however, such a suit could not be brought, and in such a situation the taxpayer’s remedy in District Court was against the Collector. *152But because the Collector had to be sued personally, no District Court action was available if he was deceased.9 The 1921 provision, which was an amendment to the Tucker Act, was explicitly designed to permit taxpayers to sue the United States in the District Courts for sums exceeding $10,000 where the Collector had died.10
The ancestry of the language of § 1346 (a)(1) is no more enlightening than is the legislative history of the 1921 provision. This language, which, as we have stated, appeared in substantially its present form in the 1921 amendment, was apparently taken from R. S. § 3226 (1878). But § 3226 was not a jurisdictional statute at all; it simply specified that suits for recovery of taxes, penalties, or sums could not be maintained until after a claim for refund had been submitted to the Commissioner.11
Thus there is presented a vexing situation — statutory language which is inconclusive and legislative history which is irrelevant. This, of course, does not necessarily mean that § 1346 (a) (1) expresses no congressional intent with respect to the issue before the Court; but it does make that intent uncommonly difficult to divine.
It is argued, however, that the puzzle may be solved through consideration of the historical basis of a suit to recover a tax illegally assessed. The argument proceeds as follows: A suit to recover taxes could, before the Tucker *153Act, be brought only against the Collector. Such a suit was based upon the common-law count of assumpsit for money had and received, and the nature of that count requires the inference that a suit for recovery of part payment of a tax could have been maintained. Neither the Tucker Act nor the 1921 amendment indicates an intent to change the nature of the refund action in any pertinent respect. Consequently, there is no warrant for importing into § 1346 (a)(1) a full-payment requirement.
For reasons which will appear later, we believe that the conclusion would not follow even if the premises were clearly sound. But in addition we have substantial doubt about the validity of the premises. As we have already indicated, the language of the 1921 amendment does in fact tend to indicate a congressional purpose to require full payment as a jurisdictional prerequisite to suit for refund. Moreover, we are not satisfied that the suit against the Collector was identical to the common-law action of assumpsit for money had and received. One difficulty is that, because of the Act of February 26, 1845, c. 22, 5 Stat. 727, which restored the right of action against the Collector after this Court had held that it had been implicitly eliminated by other legislation,12 the Court no longer regarded the suit as a common-law action, but rather as a statutory remedy which “in its nature [was] a remedy against the Government.” Curtis’s Administratrix v. Fiedler, 2 Black 461, 479. On the other hand, it is true that none of the statutes relating to this type of suit clearly indicate a congressional intention to require full payment of the assessed tax before suit.13 Nevertheless, the opinion of this Court in Cheatham v. United States, 92 U. S. 85, prevents us from accepting the *154analogy between the statutory action against the Collector and the common-law count. In this 1875 opinion, the Court described the remedies available to taxpayers as follows:
“So also, in the internal-revenue department, the statute which we have copied allows appeals from the assessor to the commissioner of internal revenue ; and, if dissatisfied with his decision, on paying the tax the party can sue the collector; and, if the money was wrongfully exacted, the courts will give him relief by a judgment, which the United States pledges herself to pay.
“. . . While a free course of remonstrance and appeal is allowed within the departments before the money is finally exacted, the general government has wisely made the payment of the tax claimed, whether of customs or of internal revenue, a condition precedent to a resort to the courts by the party against whom the tax is assessed. ... If the compliance with this condition [that appeal must be made to the Commissioner and suit brought within six months of his decision] requires the party aggrieved to pay the money, he must do it. He cannot, after the decision is rendered against him, protract the time within which he can contest that decision in the courts by his own delay in paying the money. It is essential to the honor and orderly conduct of the government that its taxes should be promptly paid, and drawbacks speedily adjusted; and the rule prescribed in this class of cases is neither arbitrary nor unreasonable. . . .
“The objecting party can take his appeal. He can, if the decision is delayed beyond twelve months, *155rest his case on that decision; or he can pay the amount claimed, and commence his suit at any time within that period. So, after the decision, he can pay at once, and commence suit within the six months . . . .” 92 U. S., at 88-89. (Emphasis added.)
Reargument has not changed our view that this language reflects an understanding that full payment of the tax was a prerequisite to suit. Of course, as stated in our prior opinion, the Cheatham statement is dictum; but we reiterate that it appears to us to be “carefully considered dictum.” 357 U. S., at 68. Equally important is the fact that the Court was construing the claim-for-refund statute from which, as amended, the language of § 1346 (a)(1) was presumably taken.14 Thus it seems that in Cheatham the Supreme Court interpreted this language not only to specify which claims for refund must first be presented for administrative reconsideration, but also to constitute an additional qualification upon the statutory right to sue the Collector. It is true that the version of the provision involved in Cheatham contained only the phrase “any tax.” But the phrases “any penalty” and “any sum” were added well before the decision in Cheatham;15 the history of these amendments makes it quite clear that they were not designed to effect any change relevant to the Cheatham rule;16 language in *156opinions of this Court after Cheatham is consistent with the Cheatham statement;17 and in any event, as we have indicated, we can see nothing in these additional words which would negate the full-payment requirement.
*157If this were all the material relevant to a construction of § 1346 (a)(1), determination of the issue at bar would be inordinately difficult. Favoring petitioner would be the theory that, in the early nineteenth century, a suit for recovery of part payment of an assessment could be maintained against the Collector, together with the absence of any conclusive evidence that Congress has ever intended to inaugurate a new rule; favoring respondent would be the Cheatham statement and the language of the 1921 statute. There are, however, additional factors which are dispositive.
We are not here concerned with a single sentence in an isolated statute, but rather with a jurisdictional provision which is a keystone in a carefully articulated and quite complicated structure of tax laws. From these related statutes, all of which were passed after 1921, it is apparent that Congress has several times acted upon the assumption that § 1346 (a)(1) requires full payment before suit. Of course, if the clear purpose of Congress at any time had been to permit suit to recover a part payment, this subsequent legislation would have to be disregarded. But, as we have stated, the evidence pertaining to this intent *158is extremely weak, and we are convinced that it is entirely too insubstantial to justify destroying the existing harmony of the tax statutes. The laws which we consider especially pertinent are the statute establishing the Board of Tax Appeals (now the Tax Court), the Declaratory Judgment Act, and § 7422 (e) of the Internal Revenue Code of 1954.
The Board of Tax Appeals.
The Board of Tax Appeals was established by Congress in 1924 to permit taxpayers to secure a determination of tax liability before payment of the deficiency.18 The Government argues that the Congress which passed this 1924 legislation thought full payment of the tax assessed was a condition for bringing suit in a District Court; that Congress believed this sometimes caused hardship; and that Congress set up the Board to alleviate that hardship. Petitioner denies this, and contends that Congress’ sole purpose was to enable taxpayers to prevent the Government from collecting taxes by exercise of its power of distraint.19
We believe that the legislative history surrounding both the creation of the Board and the subsequent revisions of the basic statute supports the Government. The House Committee Report, for example, explained the purpose of the bill as follows:
“The committee recommends the establishment of a Board of Tax Appeals to which a taxpayer may appeal prior to the payment of an additional assessment of income, excess-profits, war-profits, or estate taxes. Although a taxpayer may, after payment of *159his tax, bring suit for the recovery thereof and thus secure a judicial determination on the questions involved, he can not, in view of section 3224 of the Revised Statutes, which prohibits suits to enjoin the collection of taxes, secure such a determination prior to the payment of the tax. The right of appeal after payment of the tax is an incomplete remedy, and does little to remove the hardship occasioned by an incorrect assessment. The payment of a large additional tax on income received several years previous and which may have, since its receipt, been either wiped out by subsequent losses, invested in nonliquid assets, or spent, sometimes forces taxpayers into bankruptcy, and often causes great financial hardship and sacrifice. These results are not remedied by permitting the taxpayer to sue for the recovery of the tax after this payment. He is entitled to an appeal and to a determination of his liability for the tax prior to its payment.” 20 (Emphasis added.)
Moreover, throughout the congressional debates are to be found frequent expressions of the principle that payment of the full tax was a precondition to suit: “pay his tax . . . then . . . file a claim for refund”; “pay the tax and then sue”; “a review in the courts after payment of the tax”; “he may still seek court review, but he must first pay the tax assessed”; “in order to go to court he must pay his assessment”; “he must pay it [his assess*160ment] before he can have a trial in court”; “pay the taxes adjudicated against him, and then commence a suit in a court”; “pay the tax . . . [t]hen . . . sue to get it back”; “paying his tax and bringing his suit”; “first pay his tax and then sue to get it back”; “take his case to the district court — conditioned, of course, upon his paying the assessment.” 21
Petitioner’s argument falls under the weight of this evidence. It is true, of course, that the Board of Tax Appeals procedure has the effect of staying collection,22 and it may well be that Congress so provided in order to alleviate hardships caused by the long-standing bar against suits to enjoin the collection of taxes. But it is a considerable leap to the further conclusion that amelioration of the hardship of prelitigation payment as a jurisdictional requirement was not another important *161motivation for Congress' action.23 To reconcile the legislative history with this conclusion seems to require the presumption that all the Congressmen who spoke of payment of the assessment before suit as a hardship understood — without saying — that suit could be brought for whatever part of the assessment had been paid, but believed that, as a practical matter, hardship would nonetheless arise because the Government would require payment of the balance of the tax by exercising its power of distraint. But if this was in fact the view of these legislators, it is indeed extraordinary that they did not say so.24 *162Moreover, if Congress’ only concern was to prevent dis-traint, it is somewhat difficult to understand why Congress did not simply authorize injunction suits. It is interesting to note in this connection that bills to permit the same type of prepayment litigation in the District Courts as is *163possible in the Tax Court have been introduced several times, but none has ever been adopted.25
In sum, even assuming that one purpose of Congress in establishing the Board was to permit taxpayers to avoid distraint, it seems evident that another purpose was to furnish a forum where full payment of the assessment would not be a condition precedent to suit. The result is a system in which there is one tribunal for prepayment litigation and another for post-payment litigation, with no room contemplated for a hybrid of the type proposed by petitioner.
*164The Declaratory Judgment Act.
The Federal Declaratory Judgment Act of 193426 was amended by § 405 of the Revenue Act of 1935 expressly to except disputes “with respect to Federal taxes.” 27 The Senate Report explained the purpose of the amendment as follows :
“Your committee has added an amendment making it clear that the Federal Declaratory Judgments Act of June 14, 1934, has no application to Federal taxes. The application of the Declaratory Judgments Act to taxes would constitute a radical departure from the long-continued policy of Congress (as expressed in Rev. Stat. 3224 and other provisions) with respect to the determination, assessment, and collection of Federal taxes. Your committee believes that the orderly and prompt determination and collection of Federal taxes should not be interfered with by a procedure designed to facilitate the settlement of private controversies, and that existing procedure both in the Board of Tax Appeals and the courts affords ample remedies for the correction of tax errors.” 28 (Emphasis added.)
It is clear enough that one “radical departure” which was averted by the amendment was the potential circumvention of the “pay first and litigate later” rule by way of suits for declaratory judgments in tax cases.29 Peti*165tioner would have us give this Court’s imprimatur to precisely the same type of “radical departure,” since a suit for recovery of but a part of an assessment would determine the legality of the balance by operation of the principle of collateral estoppel. With respect to this unpaid portion, the taxpayer would be securing what is in effect — even though not technically — a declaratory judgment. The frustration of congressional intent which petitioner asks us to endorse could hardly be more glaring, for he has conceded that his argument leads logically to the conclusion that payment of even $1 on a large assessment entitles the taxpayer to sue — a concession amply warranted by the obvious impracticality of any judicially created jurisdictional standard midway between full payment and any payment.
Section 7422 (e) of the 1954 Code.
One distinct possibility which would emerge from a decision in favor of petitioner would be that a taxpayer might be able to split his cause of action, bringing suit for refund of part of the tax in a Federal District Court and litigating in the Tax Court with respect to the remainder. In such a situation the first decision would, of course, control. Thus if for any reason a litigant would prefer a District Court adjudication,30 he might sue for a small portion of the tax in that tribunal while at the same time protecting the balance from distraint by invoking the protection of the Tax Court procedure. On the other hand, different questions would arise if this device were not employed. For example, would the Government be required to file a compulsory counterclaim for the unpaid *166balance in District Court under Rule 13 of the Federal Rules of Civil Procedure? If so, which party would have the burden of proof? 31
Section 7422 (e) of the 1954 Internal Revenue Code makes it apparent that Congress has assumed these problems are nonexistent except in the rare case where the taxpayer brings suit in a District Court and the Commissioner then notifies him of an additional deficiency. Under § 7422 (e) such a claimant is given the option of pursuing his suit in the District Court or in the Tax Court, but he cannot litigate in both. Moreover, if he decides to remain in the District Court, the Government may — but seemingly is not required to — bring a counterclaim; and if it does, the taxpayer has the burden of proof.32 If we *167were to overturn the assumption upon which Congress has acted, we would generate upon a broad scale the very problems Congress believed it had solved.33
These, then, are the basic reasons for our decision, and our views would be unaffected by the constancy or inconstancy of administrative practice. However, because the petition for rehearing in this case focused almost exclusively upon a single clause in the prior opinion — “there does not appear to be a single case before 1940 in which a taxpayer attempted a suit for refund of income taxes without paying the full amount the Government alleged to be due,” 357 U. S., at 69 — we feel obliged to comment upon the material introduced upon reargument. The *168reargument has, if anything, strengthened, rather than weakened, the substance of this statement, which was directed to the question whether there has been a consistent understanding of the “pay first and litigate later” principle by the interested government agencies and by the bar.
So far as appears, Suhr v. United States, 18 F. 2d 81, decided by the Third Circuit in 1927, is the earliest case in which a taxpayer in a refund action sought to contest an assessment without having paid the full amount then due.34 In holding that the District Court had no jurisdiction of the action, the Court of Appeals said:
“None of the various tax acts provide for recourse to the courts by a taxpayer until he has failed to get relief from the proper administrative body or has paid all the taxes assessed against him. The payment of a part does not confer jurisdiction upon the courts. . . . There is no provision for refund to the taxpayer of any excess payment of any installment or part of his tax, if the whole tax for the year has not been paid.” Id., at 83.
*169Although the statement by the court might have been dictum,35 it was in accord with substantially contemporaneous statements by Secretary of the Treasury A. W. Mellon, by Under Secretary of the Treasury Garrard B. Winston, by the first Chairman of the Board of Tax Appeals, Charles D. Hamel, and by legal commentators.36
*170There is strong circumstantial evidence that this view of the jurisdiction of the courts was shared by the bar at least until 1940, when the Second Circuit Court of Appeals rejected the Government’s position in Coates v. United States, 111 F. 2d 609. Out of the many thousands of refund cases litigated in the pre-1940 period — the Govern*171ment reports that there have been approximately 40,000 such suits in the past 40 years — exhaustive research has uncovered only nine suits in which the issue was present, in six of which the Government contested jurisdiction on part-payment grounds.37 The Government’s failure to *175raise the issue in the other three is obviously entirely without significance. Considerations of litigation strategy may have been thought to militate against resting upon such a defense in those cases. Moreover, where only nine lawsuits involving a particular issue arise over a period of many decades, the policy of the Executive Department on that issue can hardly be expected to become familiar to every government attorney. But most important, the number of cases before 1940 in which the issue was present is simply so inconsequential that it reinforces the conclusion of the prior opinion with respect to the uniformity of the pre-1940 belief that full payment had to precede suit.
A word should also be said about the argument that requiring taxpayers to pay the full assessments before bringing suits will subject some of them to great hardship. This contention seems to ignore entirely the right of the taxpayer to appeal the deficiency to the Tax Court without paying a cent.38 If he permits his time for filing such an appeal to expire, he can hardly complain that he has been unjustly treated, for he is in precisely the same position as any other person who is barred by a statute of limitations. On the other hand, the Government has a substantial interest in protecting the public purse, an interest which would be substantially impaired if a taxpayer could sue in a District Court without paying his tax in full. It is instructive to note that, as of June 30, 1959, tax cases pending in the Tax Court involved $920,046,748, and refund suits in other courts involved $446,673,640.39 *176It is quite true that the filing of an appeal to the Tax Court normally precludes the Government from requiring payment of the tax,40 but a decision in petitioner’s favor could be expected to throw a great portion of the Tax Court litigation into the District Courts.41 Of course, the Government can collect the tax from a District Court suitor by exercising its power of distraint — if he does not split his cause of action — but we cannot believe that compelling resort to this extraordinary procedure is either wise or in accord with congressional intent. Our system of taxation is based upon voluntary assessment and payment, not upon distraint.42 A full-payment requirement will promote the smooth functioning of this system; a part-payment rule would work at cross-purposes with it.43
In sum, if we were to accept petitioner’s argument, we would sacrifice the harmony of our carefully structured twentieth century system of tax litigation, and all that *177would be achieved would be a supposed harmony of § 1346 (a)(1) with what might have been the nineteenth century law had the issue ever been' raised. Reargument has but fortified our view that § 1346 (a)(1), correctly construed, requires full payment of the assessment before an income tax refund suit can be maintained in a Federal District Court.
Affirmed.
Mr. Justice Frankfurter.I should like to append a word to my Brother Whit-taker's opinion, with which I entirely agree.
While Dobson v. Commissioner, 320 U. S. 489, is no longer law, the opinion of the much lamented Mr. Justice Jackson, based as it was on his great experience in tax litigation, has not lost its force insofar as it laid bare the complexities and perplexities for judicial construction of tax legislation. For one not a specialist in this field to examine every tax question that comes before the Court independently would involve in most cases an inquiry into the course of tax legislation and litigation far beyond the facts of. the immediate case. Such an inquiry entails weeks of study and reflection. Therefore, in construing a tax law it has been my rule to follow almost blindly accepted understanding of the meaning of tax legislation, when that is manifested by long-continued, uniform *178practice, unless a statute leaves no admissible opening for administrative construction.
Therefore, when advised in connection with the disposition of this case after its first argument that “there does not appear to be a single case before 1940 in which a taxpayer attempted a suit for refund of income taxes without paying the full amount the Government alleged to be due,” (357 U. S. 63, at 69), I deemed such a long-continued, unbroken practical construction of the statute controlling as to the meaning of the Revenue Act of 1921, now 28 U. S. C. § 1346 (a)(1). Once the basis which for me governed the disposition of the case was no longer available, I was thrown back to an independent inquiry of the course of tax legislation and litigation for more than a hundred years, for all of that was relevant to a true understanding of the problem presented by this case. This involved many weeks of study during what is called the summer vacation. Such a study led to the conclusion set forth in detail in the opinion of my Brother Whittaker.
The decision of the Court of Appeals in Flora conflicted with Bushmiaer v. United States, 230 F. 2d 146 (C. A. 8th Cir.). Cf. Coates v. United States, 111 F. 2d 609 (C. A. 2d Cir.); Sirian Lamp *148Co. v. Manning, 123 F. 2d 776 (C. A. 3d Cir.); Suhr v. United States, 18 F. 2d 81 (C. A. 3d Cir.), semble.
As will appear later, prior to 1940 the general view was that full payment was a jurisdictional prerequisite. But a substantial difference of opinion arose after 1940, when the Court of Appeals for the Second Circuit decided Coates v. United States, 111 F. 2d 609, against the Government. See Riordan, Must You Pay Full Tax Assessment Before Suing in the District Court? 8 J. Tax. 179; Beaman, When Not to Go to the Tax Court: Advantages and Procedures in Going to the District Court, 7 J. Tax. 356; Rudick and Wender, Federal Income Taxation, 32 N. Y. U. L. Rev. 751, 777-778; Note, 44 Calif. L. Rev. 956; Note, 2 How. L. J. 290.
See I. R. C. (1954), §§ 1 (a), 1 (b) (1), 68A Stat. 5, 6. The same general pattern has existed for many years. See, e. g., §§ 116, 117, of the Act of June 30, 1864, c. 173, 13 Stat. 281-282.
Revenue Act of 1924, c. 234, §275 (a), 43 Stat. 298; Revenue Act of 1918, c. 18, §250 (e), 40 Stat. 1084; Act of June 6, 1872, c. 315, §21, 17 Stat. 246; Act of June 30, 1864, e. 173, §119, 13 Stat. 283. See also Helvering v. Mitchell, 303 U. S. 391, 405.
Lower courts have given this construction to the same three phrases in certain claim-for-refund and limitations provisions in prior tax statutes. United States v. Magoon, 77 F. 2d 804; Union Trust Co. v. United States, 5 F. Supp. 259, 261 (“The natural definition of 'tax' comprehends one 'assessment' or one tax in the entire amount of liability”), aff'd, 70 F. 2d 629, 630 (“We agree with the District Court that ‘tax,’ 'penalty,' and ‘sum’ refer to distinct categories of illegal collections and ‘tax’ includes the entire tax liability as assessed by the Commissioner”); United States v. Clarke, 69 F. 2d 748; Hills v. United States, 50 F. 2d 302, 55 F. 2d 1001 (Ct. Cl.); cf. Blair v. Birkenstock, 271 U. S. 348.
In the prior opinion we stated that, were it not for certain countervailing considerations, the statutory language “might ... be termed a clear authorization” to sue for the refund of part payment of an assessment. 357 U. S., at 65. It is quite obvious that we did not regard the language as clear enough to preclude deciding the case on other grounds. Moreover, it could at that time be assumed that the terms of the statute favored the taxpayer, because eight members of the Court considered the extrinsic evidence alone sufficient to decide the case against him. Although we are still of that opinion, we now state our views with regard to the bare words of the statute because the argument that these words are decisively against the Government has been urged so strenuously.
42 Stat. 311.
24 Stat. 505, as amended, 28 U. S. C. §§ 1346, 1491. See United States v. Emery, Bird, Thayer Realty Co., 237 U. S. 28.
Smietanka v. Indiana Steel Co., 257 U. S. 1.
See H. R. Conf. Rep. No. 486, 67th Cong., 1st Sess. 57; remarks of Senator Jones, 61 Cong. Rec. 7506-7507. Another amendment was added in 1925 giving the right to bring refund suits against the United States where the Collector was out of office. 43 Stat. 972. And in 1954, both the $10,000 limitation and the limitation with respect to the Collector being dead or out of office were eliminated. 68 Stat. 589.
The text of R. S. § 3226 is set forth in note 16, infra, together with a more detailed account of the origin and development of the pertinent statutory language. The successor of R. S. § 3226 is I. R. C. (1954), §7422 (a), 68A Stat. 876.
See Cary v. Curtis, 3 How. 236.
E. g., Act of Feb. 26, 1845, c. 22, 5 Stat. 727; Act of Mar. 3, 1863, c. 74, 12 Stat. 729; Act of June 30, 1864, c. 173, § 44, 13 Stat. 239-240.
See note 16, infra.
Cheatham was decided in O. T. 1875, while the phrases in question were added to the statute on June 6, 1872. See note 16, infra, for a discussion of the statute involved in Cheatham and its amendment.
Section 19 of the Act of July 13, 1866, c. 184, 14 Stat. 152, was involved in Cheatham. That section provided:
“Sec. 19. . . . [N]o suit shall be maintained in any court for the recovery of any tax alleged to have been erroneously or illegally *156assessed or collected, until appeal shall have been duly made to the commissioner of internal revenue . . . ."
The phrases “any penalty” and “any sum” were first introduced into the statute in § 44 of the Act of June 6, 1872, c. 316, 17 Stat. 257-258, which read as follows:
“Sec. 44. That all suits and proceedings for the recovery of any internal tax alleged to have been erroneously assessed or collected, or any penalty claimed to have been collected without authority, or jor any sum which it is alleged was excessive, or in any manner wrongfully collected, shall be brought within two years next after the cause of action accrued and not after; and all claims for the refunding of any internal tax or penalty shall be presented to the commissioner of internal revenue within two years next after the cause of action accrued and not after . . . .” (Emphasis added.)
A careful reading of this statute discloses the absurd result which would flow from construing the addition of the “any sum” language to affect the full-payment rule, which, under this argument, would be based upon the “any tax” phrase in the 1866 statute. That is, since the “any sum” phrase occurs only in the statute of limitations portion of the 1872 statute, and not in the claim-for-refund provision, a person would be able to bring a suit for part payment without filing a claim for refund.
There were no material changes in R. S. § 3226, which provided:
“Sec. 3226. No suit shall be maintained in any court for the recovery of any internal tax alleged to have been erroneously or illegally assessed or collected, or of any penalty claimed to have been collected without authority, or of any sum alleged to have been excessive or in any manner wrongfully collected, until appeal shall have been duly made to the Commissioner of . . . Internal Revenue . . . .”
It is no doubt true, as petitioner says, that these various amendments were designed to require submission of all litigable claims to the Commissioner; but, as we have explained, this indicates no more than an intent to cover taxes, penalties, and sums which might, strictly speaking, be neither taxes nor penalties.
Kings County Savings Institution v. Blair, 116 U. S. 200, 205 (1886) (“No claim for the refunding of taxes can be made according *157to law and the regulations until after the taxes have been paid. . . . [N]o suit can be maintained for taxes illegally collected unless a claim therefor has been made within the time prescribed by the law”); Pollock v. Farmers’ Loan & Trust Co., 157 U. S. 429, 609 (1895) (dissenting opinion) (“The same authorities [including the Cheatham case] have established the rule that the proper course, in a case of illegal taxation, is to pay the tax under protest or with notice of suit, and then bring an action against the officer who collected it”); Bailey v. George, 259 U. S. 16, 20. (1922) (“They might have paid the amount assessed under protest and then brought suit against the Collector . . . .”). This view of Cheatham also corresponds to that of the Court of Appeals in this case. 246 F. 2d, at 930. See also Bushmiaer v. United States, 230 F. 2d 146, 152-155 (dissenting opinion).
43 Stat. 336.
I. R. C. (1954), § 6331, 68A Stat. 783. The Government has possessed the power of distraint for almost 170 years. See Act of Mar. 3, 1791, c. 15, § 23, 1 Stat. 204.
H. R. Rep. No. 179, 68th Cong., 1st Sess. 7. The Senate Committee on Finance filed a similar report. S. Rep. No. 398, 68th Cong., 1st Sess. 8.
The reference to R. S. § 3224 in the House Report clearly was meant simply to demonstrate that a determination prior to payment by way of an injunction suit was not possible because of the statutory bar to such a suit. This anti-injunction provision has been law for many decades. See Act of Mar. 2, 1867, c. 169, § 10, 14 Stat. 475. It is now § 7421 of the Internal Revenue Code of 1954, 68A Stat. 876.
See 65 Cong. Rec. 2621, 2684, 8110; 67 Cong. Rec. 525, 1144, 3529, 3755.
As we have indicated, some of these remarks were made during debates over proposed changes in the Board of Tax Appeals legislation during the middle of the 1920’s, but they all reflect Congress’ understanding of the pre-1924 procedure and of the changes which were made by establishment of the Board. For example, shortly after the Board legislation was passed, Congress considered and rejected a proposal to make appeal to the Board and then to a Circuit Court of Appeals the taxpayer’s sole remedy. In the course of the debate, a number of Senators discussed at length the taxpayer’s right to bring a refund action in court. Some of the cited quotations are taken from that debate. The following remark of Senator Fletcher is also illuminating:
“Mr. FLETCHER. ... I think the most important right that is preserved here ... is the right to go into the district court by the taxpayer upon the payment of the tax. I do not think that we ought to allow him to do that unless he does pay the tax; but when he pays the tax his right to go into the district court is preserved.” 67 Cong. Rec. 3529. (Emphasis added.)
See also the materials quoted in note 24, infra.
See I. R. C. (1954), § 6213 (a), 68A Stat. 771. For the pertinent 1924 legislation, see Revenue Act of 1924, c. 234, § 274, 43 Stat. 297.
In Old Colony Trust Co. v. Commissioner, 279 U. S. 716, 721, this Court expressed the view thát the Board “was created by Congress to provide taxpayers an opportunity to secure an independent review ... in advance of their paying the tax found by the Commissioner to be due. Before the Act of 1924 the taxpayer could only contest the Commissioner’s determination of the amount of the tax after its payment.”
There are a few interchanges among Senators which might be construed to indicate that they were thinking in terms of preventing distraint; but the same passages demonstrate even more clearly that these Senators also intended to eliminate the necessity of full payment as a prerequisite to suit. For example, the following debate occurred when Senator Reed, who was a member of the Committee on Finance, proposed an amendment which would have permitted a taxpayer to refuse to pay the deficiency even after the Board had ruled against him and which would have required the Government to sue in a District Court.
“Mr. REED of Missouri. . . .
“The practice, as I understand it, has been to require the taxpayer to pay in the amount of the increased assessment, and then to allow him to get it back if he can. In addition to this, distraints frequently have been issued seizing the property of the citizen ....
“Mr. SWANSON. What are the processes by which a citizen who has overpaid can get back his money under the existing law?
“Mr. REED of Missouri. As I understand it, he pays his tax. Then he makes an application for a return of it. That is heard *162through the long, troublesome processes which exist .... When the Treasury is satisfied . . . the taxpayer can go into court at that time. In the meantime, however, he has had to pay his money.
“Mr. SWANSON. Does the Senator mean that if there is a dispute, the tax is not assessed permanently against him until the board reaches its final decision?
“Mr. SMOOT. Until the board of appeals finally passes upon it, and after that if he wants to go to court he can do so, but in order to go to court he must pay his assessment.
“Mr. REED of Missouri. He must pay it before he can have a trial in court.
“Mr. WALSH of Montana. Mr. President, the hardships ... in connection with the collection of these taxes is a very real one. . . . At least two or three instances have come under my notice, and my assistance has been asked in cases where the assessing officers have . . . assessed against the [taxpayer] delinquent taxes of such an amount that he found it impossible to pay in advance and secure redress through the ordinary proceeding in a court of law, simply because it would bankrupt him to endeavor to raise the money. He was therefore obliged to suffer a distraint. . . .
“. . . After the board of review determines the matter, it seems to me, that is as far as the Government ought to be interrupted in the matter of the collection of its revenues. Then the taxpayer would be obliged to pay the tax and take his ordinary action at law to recover whatever he claims was exacted of him illegally.” 65 Cong. Rec. 8109-8114.
A somewhat similar exchange occurred during the 1926 debate over a proposal to prohibit refund suits where an appeal had been taken to the Board.
“Mr. REED of Missouri. . . . Now just one further question:
“Why is it that a taxpayer can not be given his day in court by direct action, without first requiring him to pay the tax that is assessed? I know I shall be met with the statement that it would *163mean interminable delay to the Government; but it frequently happens that the tax that is assessed is ruinous, and that the taxpayer can not raise the money. . . .
“In my own personal experience I have had two clients who were absolutely ruined by assessments that were unjust and that could not have stood up in a court of justice. . . . [A]nd it was no protection to them to say, 'Pay your taxes and then go into court,’ because they did not have the money to pay the taxes and could not raise the money to pay the taxes and be out of the money two or three years.
"... I think the bill needs just one more amendment in this particular, and that is a provision that any citizen can go into court without paying any tax and resist the payment. In the meantime I agree that the Government for its own protection ought to be allowed, perhaps, in such a case as that to issue a distraint. But the idea that a man must first pay his money and then sue to get it back is anomaly in the law.” 67 Cong. Rec. 3530-3533.
Senator Reed later proposed that the appeal from the Board be to the District Court instead of to the Circuit Court of Appeals, and Senator Wadsworth, a member of the Finance Committee, asked:
“Does the Senator not think that other provision in the bill which permits the taxpayer to take his case to the district court — conditioned, of course, upon his paying the assessment — meets the situation?” 67 Cong. Rec. 3755.
S. 1569, 81st Cong., 1st Sess.; S. 384, 82d Cong., 1st Sess.; H. R. 150 and H. R. 246, 83d Cong., 1st Sess.
48 Stat. 955, as amended, 28 U. S. C. §§ 2201, 2202.
49 Stat. 1027.
S. Rep. No. 1240, 74th Cong., 1st Sess. 11.
“Should the Declaratory Judgment Act be held to apply to tax cases it will mean a complete reversal of our present scheme of taxation. The principle of ‘pay first and litigate later’ will be changed to ‘litigate first and pay later.’ This principle has never before been departed from.” Wideman, Application of the Declaratory Judgment Act to Tax Suits, 13 Taxes 539, 540.
For some practitioners’ views on the desirability of litigating tax cases in Federal District Courts, see Dockery, Refund Suits in District Courts, 31 Taxes 523; Yeatman, Tax Controversies, 10 Tex. B. J. 9.
These problems have already occurred to the bar. See Riordan, Must You Pay Full Tax Assessment Before Suing in the District Court? 8 J. Tax. 179, 181.
Ҥ 7422. CIVIL ACTIONS FOR REFUND.
“(e) Stay op PROCEEDINGS. — If the Secretary or his delegate prior to the hearing of a suit brought by a taxpayer in a district court or the Court of Claims for the recovery of any income tax, estate tax, or gift tax (or any penalty relating to such taxes) mails to the taxpayer a notice that a deficiency has been determined in respect of the tax which is the subject matter of taxpayer’s suit, the proceedings in taxpayer’s suit shall be stayed during the period of time in which the taxpayer may file a petition with the Tax Court for a redeter-mination of the’1 asserted deficiency, and for 60 days thereafter. If the taxpayer files a petition with the Tax Court, the district court or the Court of Claims, as the case may be, shall lose jurisdiction of taxpayer’s suit to whatever extent jurisdiction is acquired by the Tax Court of the subject matter of taxpayer’s suit for refund. If the taxpayer does not file a petition with the Tax Court for a rede-termination of the asserted deficiency, the United States may counterclaim in the taxpayer’s suit, or intervene in the event of a suit as described in subsection (c) (relating to suits against officers or employees of the United States), within the period of the stay of proceedings notwithstanding that the time for such pleading may have *167otherwise expired. The taxpayer shall have the burden of proof with respect to the issues raised by such counterclaim or intervention of the United States except as to the issue of whether the taxpayer has been guilty of fraud with intent to evade tax. This subsection shall not apply to a suit by a taxpayer which, prior to the date of enactment of this title, is commenced, instituted, or pending in a district court or the Court of Claims for the recovery of any income tax, estate tax, or gift tax (or any penalty relating to such taxes).” 68A Stat. 877.
The possibility of dual jurisdiction in this type of situation was confirmed by cases such as Camp v. United States, 44 F. 2d 126, and Ohio Steel Foundry Co. v. United States, 69 Ct. Cl. 158, 38 F. 2d 144. See H. R. Rep. No. 1337, 83d Cong., 2d Sess. 109, A431; S. Rep. No. 1662, 83d Cong., 2d Sess. 148, 610.
For additional evidence of recent congressional understanding of the jurisdictional requirement of § 1346 (a)(1), see the House Report which explained the 1954 amendment abolishing the $10,000 limitation on tax suits against the United States, 68 Stat. 589. After explaining the taxpayer’s right to contest a deficiency in the Tax Court, the report states: "The taxpayer may, however, elect to pay his tax and thereafter bring suit to recover the amount claimed to have been illegally exacted.” H. R. Rep. No. 659, 83d Cong., 1st Sess. 2.
Petitioner cites two earlier cases in which the Government failed to raise the jurisdictional issue. Bowers v. Kerbaugh-Empire Co., 271 U. S. 170 (1926); Cook v. Tait, 265 U. S. 47 (1924). The Government distinguishes these cases on the ground that, although the total tax for the year had not been paid, the full amount due at the time of suit had been paid. This situation occurred because under §250 (a) of the Revenue Act of 1921, c. 136, 42 Stat. 264, the tax was paid in four installments, and the plaintiffs in Cook and Bowers apparently had paid the due installments. While we do not suggest that the statute will support this type of distinction, adoption of it by the Government or by the bar would not in any way impair the substantial consistency of the view that full payment has for many decades been a prerequisite to suit in District Court. An error as to the applicability of a principle to a unique factual situation does not mean that the principle itself has been rejected.
The ground for the decision may have been that the District Court had no jurisdiction because the taxpayer was contesting the legality of the balance of the assessment before the Board of Tax Appeals.
In welcoming the members of the Board of Tax Appeals on July 16, 1924, Under Secretary Winston described the difficulties which had arisen in the past.
“. . . Under the law a tax once assessed had to be paid by the taxpayer and then his remedy was to sue for its recovery. He must first find the cash for a liability for which he may not have provided. . . . The first interest of all of the people is, of course, that the Government continue to function, and to do this it must have the means of prompt collection of the necessary supplies to keep it going, that is, taxes. The method was, therefore, the determination by the Commissioner of the amount of tax due, its collection and suit to recover. . . . [T]he tax as assessed had to be paid and the taxpayer was left to his remedy in the courts. The payment of the tax was often a great hardship on the taxpayer, meaning in general that he had to raise the cash for an unexpected liability which might not be lawfully due.” Treas. Dept. Press Release, July 16, 1924. See also remarks by Under Secretary Winston in addressing the Seventeenth Annual Conference of the National Tax Association in September 1924, Proceedings of Seventeenth National Conference 271.
In commenting upon the Board of Tax Appeals legislation, which contemplated leaving the taxpayer to his District Court remedy if the decision of the Board was adverse, Secretary of the Treasury Mellon stated: “The taxpayer, in the event that decision [of the Board] is against him, will have to pay the tax according to the assessment and have recourse to the courts . . . .” 67 Cong. Rec. 552.
On September 17, 1924, the first Chairman of the Board, Charles D. Hamel, read a paper before the Seventeenth Annual Conference *170of the National Tax Association on Taxation which contained the following remark: “Prior to the enactment of the Act of 1924 . . . [i]f the decision on the appeal [to the Commissioner] was in favor of the government, the taxpayer, only after payment of the tax, had the right to protest the correctness of the decision in the courts . . . Proceedings 277-278.
One of the clearest statements of the rule by a commentator is to be found in Bickford, Court Procedure in Federal Tax Cases (Rev. ed. 1929) 3, 7-8, 9, 119.
“There are, however, certain other conditions which must be complied with before a suit is maintainable under this section. Briefly stated, these are as follows:
“1. The tax must have been paid.
“2. After payment, the taxpayer must have filed with the Commissioner ... a sufficient claim for the refund of the taxes sued for.
“The first requirement is obvious. We have, in the preceding portions of this volume, found that a proceeding commenced in the Board of Tax Appeals is the only exception to the rule that no review by the courts is permissible at common law or under the statutes, until the tax has been paid and the Government assured of its revenue.” Id., at 119.
See also Hamel, The United States Board of Tax Appeals (1926), 10; Klein, Federal Income Taxation (1929), 1372, 1642, 1643; Mellon, Taxation: The People’s Business (1924), 62-63; Bailan tine, Federal Income Tax Procedure, Lectures on Taxation, Columbia University Symposium (1932), 179, 192-193; Caspers, Assessment of Additional Income Taxes for Prior Years, 1 Nat. Income Tax Mag. (Oct. 1923), 12; Graupner, The Operation of the Board of Tax Appeals, 3 Nat. Income Tax Mag. (1925), 295. But see Smith, National Taxes, Their Collection, and Rights and Remedies of the Taxpayer, 8 Geo. L. J. 1, 3 (Apr. 1920).
See also Beaman, When Not to Go to the Tax Court: Advantages and Procedures in Going to the District Court, 7 J. Tax. (1957), 356 *171(“[T]he Bushmiaer case [permitting suit for part of the tax] . . . runs counter to a long tradition of administrative practice and interpretation . . . .”); Rudick and Wender, Federal Income Taxation, 32 N. Y. U. L. Rev. (1957), 751, 777-778 (“It is generally said that a taxpayer has two remedies if he disagrees with a determination of the Commissioner. He may pay the deficiency, file a claim for refund, and sue for the tax in the district court .... Alternatively, the taxpayer may petition the Tax Court for review of a deficiency prior to payment. The recent Bushmiaer case is a third alternative. . . . [T]he Bushmiaer case conflicts with more than thirty years of experience in the administration and collection of taxes.”). (Footnote omitted.)
Petitioner cites a number of cases in support of his argument that neither the bar nor the Government has ever assumed that full payment of the tax is a jurisdictional prerequisite to suit for recovery. The following factors rob these cases of the significance attributed to them by the petitioner:
(a) A number of them, although cited by petitioner in his petition for rehearing, were later conceded by him, after his examination of government files, not to be in point.
(b) A number of the cited cases involved excise taxes. The Government suggests — and we agree — that excise tax deficiencies may be divisible into a tax on each transaction or event, and therefore present an entirely different problem with respect to the full-payment rule.
(c) The cases arising after 1940 are insignificant. Once the Second Circuit Court of Appeals had ruled against the Government in Coates, taxpayers would naturally be much more inclined to sue before full payment, and the Government might well decide not to raise the objection in a particular case for reasons relating to litigation strategy.
(d) In some of the cases the only amount remaining unpaid at the time of suit was interest. As we have indicated, the statute lends itself to a construction which would permit suit for the tax after full payment thereof without payment of any part of the interest.
(e) In some of the cases the Government was not legally entitled to collect the unpaid tax at the time of suit, either because the tax *172system at the time permitted installment payment (see note 34, supra), because the unpaid portion had not yet been assessed, or for some other reason. Although the statute may not support any distinction based on facts of this nature, it is quite understandable that a taxpayer might have predicated a suit upon the theory that the distinction was meaningful and that the Government might not have contested it, whether because it agreed or for tactical reasons.
In the light of these considerations, we regard the following pre-1941 cases as immaterial: Baldwin v. Higgins, 100 F. 2d 405 (C. A. 2d Cir. 1938) (petitioner concedes); Sampson v. Welch, 23 F. Supp. 271 (D. C. S. D. Cal. 1938) (same); Charleston Lumber Co. v. United States, 20 F, Supp. 83 (D. C. S. D. W. Va. 1937) (same); Sterling v. Ham, 3 F. Supp. 386 (D. C. Me. 1933) (same); Farmers’ Loan & Trust Co. v. Bowers, 15 F. 2d 706 (D. C. S. D. N. Y. 1926), modified, 22 F. 2d 464 (1927), rev’d, 29 F. 2d 14 (C. A. 2d Cir. 1928) (same); Heinemann Chemical Co. v. Heiner, 36-4 CCH Fed. Tax Serv. ¶ 9302 (D. C. W. D. Pa. 1936), rev’d, 92 F. 2d 344 (C. A. 3d Cir. 1937) (only interest unpaid); Welch v. Hassett, 15 F. Supp. 692 (D. C. Mass. 1936), rev’d, 90 F. 2d 833 (C. A. 1st Cir. 1937), aff’d, 303 U. S. 303 (1938) (full assessment paid); Leavitt v. Hendricksen, 37-4 CCH Fed. Tax Serv. ¶9312 (D. C. W. D. Wash. 1937) (no unpaid assessment); Bowers v. Kerbaugh-Empire Co., 271 U. S. 170 (1926) (all due installments paid); Cook v. Tait, 265 U. S. 47 (1924) (same).
Four pre-1941 cases remain. Of these, only two are clearly cases in which the jurisdictional issue was present and not raised by the Government. Tsivoglou v. United States, 31 F. 2d 706 (C. A. 1st Cir. 1929); Thomas v. United States, 85 Ct. Cl. 313, 18 F. Supp. 942 (1937). McFadden v. United States, 20 F. Supp. 625 (D. C. E. D. Pa. 1937); is in the “doubtful” category. There the Commissioner had granted the taxpayer an extension of time for payment of 80% of his assessment and the suit was for the remaining 20%, which had been paid. The relevant facts of the last case, Peerless Paper Box Mfg. Co. v. Routzahn, 22 F. 2d 459 (D. C. N. D. Ohio 1927), are so unclear that the case means nothing. The Government had applied an admitted 1918 overpayment to a 1917 deficiency, but the deficiency was greater than the overpayment. The taxpayer sued to recover this overpayment, and whether there had been full payment at the time of suit depends upon whether the suit is regarded as one for refund of 1917 or 1918 taxes.
*173Nor can we agree entirely with petitioner’s evaluation of a second group of pre-1941 cases — those in which the issue allegedly was present and the Government did raise it but lost. Five of these cases involved primarily the troublesome concurrent jurisdiction problem that arose before passage of § 7422 (e) of the 1954 Code when a taxpayer both appealed to the Tax Court and brought suit in a Federal District Court. Brampton Woolen Co. v. Field, 55 F. 2d 325 (D. C. N. H. 1931), rev’d, 56 F. 2d 23 (C. A. 1st Cir. 1932), cert. denied, 287 U. S. 608; Camp v. United States, 44 F. 2d 126 (C. A. 4th Cir. 1930); Emery v. United States, 27 F. 2d 992 (D. C. W. D. Pa. 1928); Old Colony R. Co. v. United States, 27 F. 2d 994 (D. C. Mass. 1928); Ohio Steel Foundry Co. v. United States, 69 Ct. Cl. 158, 38 F. 2d 144 (1930). In all of these cases except Camp, it appears that the Government did raise the part-payment question. It is true that the contention did not prevail, but this is not very meaningful. In the first place, this question was quite subordinate to the major issue, concurrent jurisdiction. In the second place, the Government won in Brampton on another jurisdictional ground. And finally, in contrast to Flora, in both Camp and Ohio Steel Foundry the full assessment had been paid at the time suit was brought; it was only later that an additional deficiency was asserted by the Commissioner.
To these cases should be added Riverside Hospital v. Larson, 38-4 CCH Fed. Tax Serv. ¶ 9542 (D. C. S. D. Fla. 1938), where the Government raised the full-payment question and won, and Suhr v. United States, 14 F. 2d 227 (D. C. W. D. Pa. 1926), aff’d, 18 F. 2d 81 (C. A. 3d Cir. 1927), another concurrent jurisdiction case where the Government raised the issue and won, although the grounds for the decision are not entirely clear.
This, then, is how we see the pre-1941 situation: Of 14 cases originally cited as being cases in which the jurisdictional issue was present but not raised by the Government, five have been conceded by petitioner not to be in point; six, and possibly seven, are distinguishable for various reasons; and only two, or possibly three, remain. Of five cases cited as being cases in which the jurisdictional issue was raised by the Government, only one, Coates v. United States, 111 F. 2d 609 (C. A. 2d Cir. 1940), or at most three, really involved the Flora question. When to these are added Riverside, where the Government won, Suhr, where it may have won, and Brampton Woolen Co., where it won in the Court of Appeals on another jurisdictional ground, the box score is as follows: two or three cases in which the *174Government failed to raise the issue; one, or possibly three, cases in which the Government argued the question and lost; one case in which it argued the question and won; one case in which it argued the question and may have won; and one case in which it raised the issue and prevailed on another jurisdictional defense — a total of nine cases at most in which the issue was presented, out of which the Government contested jurisdiction in six. Of course, this calculation may not be precise; but, in view of the many thousands of tax refund suits which have been brought during the decades in question, it is an accurate enough approximation to reflect a general understanding of the jurisdictional significance of “pay first, litigate later.”
It would be bootless to consider each of the post-1940 cases cited by petitioner or to list the multitude of cases cited by the Government in which the jurisdictional issue has been raised. As we have stated, we believe these cases have no significance whatsoever. However, perhaps it is worth noting that all but a handful of the cases which petitioner, in the petition for rehearing, asserted to be ones in which the Government failed to raise the jurisdictional issue would be immaterial even if they were pre-Coaies. Thus, for example, petitioner has conceded error with respect to three cases. Dickstein v. McDonald, 149 F. Supp. 580 (D. C. M. D. Pa. 1957), aff’d, 255 F. 2d 640 (C. A. 3d Cir. 1958); O’Connor v. United States, 76 F. Supp. 962 (D. C. S. D. N. Y. 1948); Terrell v. United States, 64 F. Supp. 418 (D. C. E. D. La. 1946). A number of the cases involved excise taxes. E. g., Griffiths Dairy v. Squire, 138 F. 2d 758 (C. A. 9th Cir. 1943); Auricchio v. United States, 49 F. Supp. 184 (D. C. E. D. N. Y. 1943). In some of the cases only interest remained unpaid. Raymond v. United States, 58-1 U. S. T. C. ¶ 9397 (D. C. E. D. Mich. 1958); Hogg v. Allen, 105 F. Supp. 12 (D. C. M. D. Ga. 1952). And some of the cases arose in the Third Circuit after a decision adverse to the Government in Sirian Lamp Co. v. Manning, 123 F. 2d 776 (C. A. 3d Cir. 1941). Gallagher v. Smith, 223 F. 2d 218 (1955); Peters v. Smith, 123 F. Supp. 711 (D. C. E. D. Pa. 1954), rev’d, 221 F. 2d 721 (1955). It might be noted also that Jones v. Fox, 162 F. Supp. 449 (D. C. Md. 1957), cited as a case in which the Government argued the jurisdictional question and lost, was an excise tax case in which the court distinguished our prior decision in Flora because of the divisibility of the excise tax. Another such decision during the pre-1941 period was Friebele v. United States, 20 F. Supp. 492 (D. C. N. J. 1937).
Petitioner points out that the Tax Court has no jurisdiction over excise tax cases. See 9 Mertens, Law of Federal Income Taxation (Zimet Rev. 1958), §50.08. But this fact provides no policy support for his position, since, as we have noted, excise tax assessments may be divisible into a tax on each transaction or event, so that the full-payment rule would probably require no more than payment of a small amount. See note 37, supra.
Of this $446,673,640, District Court suits involved $222,177,920; Court of Claims suits, $220,247,436; and state court suits, $4,248,284.
See note 22, supra.
The practical effects which might result from acceptance of petitioner’s argument are sketched in Lowitz, Federal Tax Refund Suits and Partial Payments, 9 The Decalogue J. 9, 10:
“Permitting refund suits after partial payment of the tax assessment would benefit many taxpayers. Such a law would be open to wide abuse and would probably seriously impair the government’s ability to collect taxes. Many taxpayers, without legitimate grounds for contesting an assessment, would make a token payment and sue for refund, hoping at least to reduce the amount they would ultimately have to pay. In jurisdictions where the District Court is considered to be a ‘taxpayer’s court’ most taxpayers would use that forum instead of the Tax Court. Conceivably such legislation could cause the chaotic tax collection situations which exist in some European countries, since there would be strong impetus to a policy of paying a little and trying to settle the balance.”
See Helvering v. Mitchell, 303 U. S. 391, 399; Treas. Regs, on Procedural Rules (1954 Code) §601.103 (a).
See Riordan, Must You Pay Full Tax Assessment Before Suing in the District Court? 8 J. Tax. 179, 181:
“1. If the Government is forced to use these remedies [distraint] *177on a large scale, it will affect adversely taxpayers’ willingness to perform under our voluntary assessment system.
“2. It will put the burden on the Government to seek out for seizure the property of every taxpayer who chooses to sue for the refund of a partial payment. Often, the Government will not be able to do this without extraordinary and costly effort and in some cases it may not be able to do it at all.
“3. The use of the drastic-collection remedies would often cause inconvenience and perhaps hardship to the creditors, debtors, employers, employees, banks and other persons doing business with the taxpayer.”
The language of Cheatham relied upon by this Court in its first opinion was the following:
“So also, in the internal-revenue department, the statute which we have copied allows appeals from the assessor to the commissioner of internal revenue; and, if dissatisfied with his decision, on paying the tax the party can sue the collector; and, if the money was wrongfully exacted, the courts will give him relief by a judgment, which the United States pledges herself to pay.
“. . . While a free course of remonstrance and appeal is allowed within the departments before the money is finally exacted, the general government has wisely made the payment of the tax claimed, whether of customs or of internal revenue, a condition precedent to a resort to the courts by the party against whom the tax is assessed. ... If the compliance with this condition [that suit must be brought within six months of the Commissioner’s decision] requires the party aggrieved to pay the money, he must do it. He cannot, after the decision is rendered against him, protract the time within which he can contest that decision in the courts by his own delay in paying the money. It is essential to the honor and orderly conduct of the government that its taxes should be promptly paid, and drawbacks speedily adjusted; and the rule prescribed in this class of cases is neither arbitrary nor unreasonable. . . .
“The objecting party can take his appeal. He can, if the decision is delayed beyond twelve months, rest his case on that decision; or he can pay the amount claimed, and commence his suit at any time within that period. So, after the decision, he can pay at once, and commence suit within the six months . . . .” 92 U. S., at 88-89.