Opinion of the Court. 322 U.S. issues of fraud itself instead of sending the case to the District Court. An order was thereupon entered denying the petition as framed but granting Hazel leave to amend the prayer of the petition to ask that the Circuit Court itself hear and determine the issue of fraud. Hazel accordingly amended, praying that the 1932 judgments against it be vacated and for such other relief as might be just. Hartford then replied and filed additional exhibits and affidavits. The following facts were shown by the record without dispute. In 1926 Hartford had pending an application for a patent on a machine which utilized a method of pouring glass into molds known as "gob feeding." The application, according to the Circuit Court, "was confronted with apparently insurmountable Patent Office opposition." To help along the application, certain officials and attorneys of Hartford determined to have published in a trade journal an article signed by an ostensibly disinterested expert which would describe the "gob feeding" device as a remarkable advance in the art of fashioning glass by machine. Accordingly these officials prepared an article entitled "Introduction of Automatic Glass Working Machinery; How Received by Organized Labor," which referred to "gob feeding" as one of the two "revolutionary devices" with which workmen skilled in bottle-blowing had been confronted since they had organized. After unsuccessfully attempting to persuade the President of the Bottle Blowers' Association to sign this article, the Hartford officials, together with other persons called to their aid, procured the signature of one William P. Clarke, widely known as National President of the Flint Glass Workers' Union. Subsequently, in July 1926, the article was published in the National Glass Budget, and in October 1926 it was introduced as part of the record in support of the pending application in the Patent Office. 238 Opinion of the Court. January 3, 1928, the Patent Office granted the application as Patent No. 1,655,391. On June 6, 1928, Hartford brought suit in the District Court for the Western District of Pennsylvania charging that Hazel was infringing this "gob feeding" patent, and praying for an injunction against further infringement and for an accounting for profits and damages. Without referring to the Clarke article, which was in the record only as part of the "file-wrapper" history, and which apparently was not then emphasized by counsel, the District Court dismissed the bill on the ground that no infringement had been proved. 39 F. 2d 111. Hartford appealed. In their brief filed with the Circuit Court of Appeals, the attorneys for Hartford, one of whom had played a part in getting the spurious article prepared for publication, directed the Court's attention to "The article by Mr. William Clarke, former President of the Glass Workers' Union." The reference was not without effect. Quoting copiously from the article to show that "labor organizations of practical workmen recognized" the "new and differentiating elements" of the "gob feeding" patent owned by Hartford, the Circuit Court on May 5, 1932, held the patent valid and infringed, reversed the District Court's judgment, and directed that court to enter a decree accordingly. 59 F. 2d 399, 403, 404. At the time of the trial in the District Court in 1929, where the article seemingly played no important part, the attorneys of Hazel received information that both Clarke and one of Hartford's lawyers had several years previously admitted that the Hartford lawyer was the true author of the spurious publication. Hazel's attorneys did not at that time attempt to verify the truth of the hearsay story of the article's authorship, but relied upon other defenses which proved successful. After the opinion of the Circuit Court came down on May 5, 1932, quoting the spurious Opinion of the Court. 322 U.S. article and reversing the decree of the District Court, Hazel hired investigators for the purpose of verifying the hearsay by admissible evidence. One of these investigators interviewed Clarke in Toledo, Ohio, on May 13 and again on May 24. In each interview Clarke insisted that he wrote the article and would so swear if summoned. In the second interview the investigator asked Clarke to sign a statement telling in detail how the article was prepared, and further asked to see Clarke's files. Clarke replied that he would not "stultify" himself by signing any "statement or affidavit"; and that he would show the records to no one unless compelled by a subpoena. At the same time, he reinforced his claim of authorship by asserting that he had spent seven weeks in preparing the article. But unknown to Hazel's investigator, a representative of Hartford, secretly informed of the investigator's view that Hazel's only chance of reopening the case "was to get an affidavit from someone, to the effect that this article was written" by Hartford's attorney, also had traveled to Toledo. Hartford's representative first went to Toledo and talked to Clarke on May 10, three days before Hazel's investigator first interviewed Clarke; and he returned to Toledo again on May 22 for a five-day stay. Thus at the time of the investigator's second interview with Clarke on May 24, representatives of both companies were in touch with Clarke in Toledo. But though Hartford's representative knew the investigator was there, the latter was unaware of the presence of the Hartford representative. On May 24, Hazel's investigator reported failure; the same day, Hartford's man reported "very successful results." Four days later, on May 28, Hartford's representative reported his "success" more fully. Clarke, he said, had been of "great assistance" and Hartford was in a "most satisfactory position"; it did not "seem wise to distribute copies of all the papers" the representative then had or 238 Opinion of the Court. to "go into much detail in correspondence"; and Hartford was "quite indebted to Mr. Clarke" who "might easily have caused us a lot of trouble. This should not be forgotten. ." Among the "papers" which the representative had procured from Clarke was an affidavit signed by Clarke stating that he, Clarke, had "signed the article and released it for publication." The affidavit was dated May 24 the very day that Clarke had told Hazel's investigator he would not "stultify" himself by signing any affidavit and would produce his papers for no one except upon subpoena. Shortly afterward Hazel capitulated. It paid Hartford $1,000,000 and entered into certain licensing agreements. The day following the settlement, Hartford's representative traveled back to Toledo and talked to Clarke. At this meeting Clarke asked for $10,000. Hartford's representative told him that he wanted too much money and that Hartford would communicate with him further. A few days later the representative paid Clarke $500 in cash; and about a month later delivered to Clarke, at some place in Pittsburgh which he has sworn he cannot remember, an additional $7,500 in cash. The reason given for paying these sums was that Hartford felt a certain moral obligation to do so, although Hartford's affidavits deny any prior agreement to pay Clarke for his services in connection with the article. Indisputable proof of the foregoing facts was, for the first time, fully brought to light in 1941 by correspondence files, expense accounts and testimony introduced at the trial of the United States v. Hartford-Empire Company et al., 46 F. Supp. 541, an anti-trust prosecution begun December 11, 1939. On the basis of the disclosures at this trial Hazel commenced the present suit. Upon consideration of what it properly termed this "sordid story," the Circuit Court, one Judge dissenting, held, first, that the fraud was not newly discovered; sec Opinion of the Court. 322 U.S. ond, that the spurious publication, though quoted in the 1932 opinion, was not the primary basis of the 1932 decision; and third, that in any event it lacked the power to set aside the decree of the District Court because of the expiration of the term during which the 1932 decision had been rendered. Accordingly the Court refused to grant the relief prayed by Hazel. Federal courts, both trial and appellate, long ago established the general rule that they would not alter or set aside their judgments after the expiration of the term at which the judgments were finally entered. Bronson v. Schulten, 104 U. S. 410. This salutary general rule springs from the belief that in most instances society is best served by putting an end to litigation after a case has been tried and judgment entered. This has not meant, however, that a judgment finally entered has ever been regarded as completely immune from impeachment after the term. From the beginning there has existed alongside the term rule a rule of equity to the effect that under certain circumstances, one of which is after-discovered fraud, relief will be granted against judgments regardless of the term of their entry. Marine Insurance Co. v. Hodgson, 7 Cranch 332; Marshall v. Holmes, 141 U. S. 589. This equity rule, which was firmly established in English practice long before the foundation of our Republic, the courts have developed and fashioned to fulfill a universally recognized need for correcting injustices which, in certain instances, are deemed sufficiently gross to demand a departure from rigid adherence to the term rule. Out of deference to the deep-rooted policy in favor of the repose of judgments entered during past terms, courts of equity have been cautious in exercising their power over such judgments. United States v. Throckmorton, 98 U. S. 61. But where the occasion has demanded, where enforcement of the judgment is "mani |