Me. Justice Bean
delivered the opinion of the court.
This is a suit to impeach and set aside a decree of this court on the ground of newly-discovered evidence, and comes here on an appeal from a decree granting the relief demanded. In brief, the facts are that in October, 1893, Homer Nessley, Frank Nessley, and Charles Nessley, being in possession of certain lands in Union County, and claiming to own the same in fee under a warranty deed from the plaintiff, brought a suit in the circuit court of that county against the present defendant to *239quiet their title to such lands, which resulted adversely to them. They subsequently appealed to this court, and, while the appeal was pending, reconveyed the land to the plaintiff. The decree of the court below was after-wards affirmed (29 Or. 354, 45 Pac. 904.), whereupon this suit was instituted for the purpose above stated. In the original cause of Nessley against the defendant it was claimed by the Nessleys, as it is claimed by the plaintiff in this suit, that the land in question had been sold and conveyed by the defendant in the year 1873 to his brother, John R. Ladd, the predecessor in interest of the Nessleys and plaintiff, and that the deed of conveyance was burned, without having been recorded, in a fire which destroyed John R. Ladd’s house in the year 1876. The execution of such deed was the real and only point in issue in the case, and upon it much evidence was given on both sides, the material portions of which, as they influenced the decision of this court, are summarized in the opinion of Mr. Justice Wolvekton. The Nessleys produced upon the hearing several witnesses who testified to having signed the deed; one of them claiming to have read it over, and undertaking to give its contents, the manner in which it was signed, how it was witnessed and acknowledged, etc. Other witnesses were called who testified as to statements made by the defendant inconsistent with a claim of ownership to the land. The defendant, however, positively denied ever having made this deed, and gave other evidence tending to corroborate his denial; and this court found, as a conclusion from all the .testimony, that he was right.
The object of the present suit is to overturn this decree and set it aside on the ground of newly-discovered evidence tending to prove the execution of such deed, and which it is claimed could not have been produced on the former trial by the exercise of reasonable diligence. This *240newly-discovered evidence consists of the oral testimony of five witnesses (McComas, Deal, Cramblitt, Martin, and Reeves), who testified to facts and circumstances tending to support the plaintiff’s theory of the case. McComas testifies that he was a notaiy public in Union County in 1873 ; that in March of that year, by request of the defendant and his brother, he drew up a deed of conveyance of the premises in controversy from the former to the latter; that the defendant signed it by his mark, and acknowledged it before him as a notary; and that he then signed as a witness. Martin says that in the spring of 1873 he was employed by McComas ; that upon a certain day McComas prepared a deed of the premises in controversy, from the defendant to his brother, and that it was duly signed by the defendant with his mark ; and that he attested it as a witness. Deal says that he was in the office of McComas some time in the spring of 1873, and saw the defendant execute a deed of the premises in controversy to his brother, and saw him sign it by his mark. Cramblitt’s testimony is substantially to the same effect. And Reeves says that, in the spring of one of the years in the early ’70’s, he was working for the defendant’s brother; that upon a certain day the defendant directed him to hitch a horse to the buggy, saying that they were going to La Grande to make a deed “to that land that the two brothers went away, and upon their return the same evening defendant informed witness that the deed had been made out. The defendant insists, among other things, that the evidence by which the former decree is sought to be impeached, is, at best, merely cumulative, or, rather, an accumulation of witnesses upon a point directly in issue in the former case, and is therefore insufficient to support the bill; and, second, that, if the newly-discovered evidence is competent, it is not sufficiently material, satisfactory, or con*241vincing to justify or warrant the decree appealed from.
1. By our statute (Hill’s Ann. Laws, § 877), bills of revieiv are abolished, and it is provided that a decree in equity may be impeached or set aside by an original suit. But it is only the form that is abolished. The substance remains the same. And an original suit, under the statute, to set aside or impeach a decree on the ground of newly discovered evidence can only be maintained upon similar grounds to those which- would be sufficient to maintain a bill of review, or a bill in the nature of a bill of review', at common law': Crews v. Richards, 14 Or. 442 (13 Pac. 67).
2. Now', it is generally agreed that mere cumulative evidence upon a point in issue in a former trial is not sufficient to support a bill of review, although there is perhaps not an entire uniformity of opinion in the authorities as to whether the mere discovery of new witnesses to prove a matter which was in issue in the former case is cumulative evidence, within the meaning of this rule. In the early and leading case, in this country, of Respass v. McClanahan, Hardin, 342, it is maintained in an opinion, the reasoning of which is so full and clear that Mr. Justice Story said in Dexter v. Arnold, 5 Mason, 315 (Fed. Cas. No. 3,856), that he should hesitate longer before acting against it, and expressed regret that he could not transfer it to his opinion in Wood v. Mann, 2 Sumn. 335 (Fed. Cas. No. 17,953), that the discovery of new witnesses to prove a matter which was in issue in the original cause is not a ground for a bill of review-, but that the new proof to support such a bill must be matter of record, or a writing not known before. In the course of the opinion the court said that, after a most careful search, it could not find “one case reported, in which a bill of review has been allowed on the discovery of newr witnesses *242to prove a fact which had been before in issue, although there are many where bills of review have been sustained on the discovery of records and other writings relating to the title which was generally put in issue. The distinction is very material. Written evidence cannot be easily corrupted, and, if it had been discovered before the former hearing, the presumption is strong that it would have been produced, to prevent further litigation and expense. New witnesses, it is granted, may easily be procured by it, and the danger of admitting them renders it highly impolitic.” Mr. Chancellor Kent, in Livingston v. Hubbs, 3 Johns. Ch. 124, arrived at a like conclusion, and seemed to think that such new evidence must not be a mere accumulation of witnesses to the same fact, but some stringent written evidence or newly-discovered papers.
Mr. Justice Story, in Dexter v. Arnold and Wood v. Mann, manifestly leans to the same limitations, for in the latter case he says : “I am not able to satisfy myself that this objection [the one made by the Court of Appeals of Kentucky] to the evidence is not well founded. On the contrary, the more I reflect, the more I feel the difficulty of the admissibility of merely cumulative and corroborative testimony, though newly discovered, to the facts in issue. If I were to decide in favor of its admissibility, I should, as far as I know, be the first judge who ever acted upon so broad a doctrine. I am not bold enough to adventure upon such a course. On the contrary, if I were called upon to frame a rule, it would be to exclude all testimony of newly-discovered witnesses to any facts in issue, unless connected with some newly-discovered documents:” 2 Sumn. 335 (Fed. Cas. No. 17,953). In Brewer v. Bowman, 3 J. J. Marsh. 492, it is declared that ‘ ‘it is not sufficient for a review to depend upon reswearing the same or other witnesses, with a *243view to reach the truth and in a monographic note to that case in 20 Am. Dec. 171, it is said: “In regard to newly-discovered evidence, it appears to be settled that, if it be documentary, it will be considered sufficient to warrant the granting of a bill of review, it being in other respects sufficient; but mere oral proof, ‘an accumulation of witnesses,’ — as it has been termed, will not be entertained on a petition for leave to file the bill.” The case of McDougald v. Dougherty, 39 Ala. 409, involved practically the precise questions presented by this record. It was a bill to review and set aside, on the ground of newly-discovered evidence, a decree in a suit brought to foi’eclose a deed of trust, rendered for want of sufficient proof of the execution of the deed; and, in deciding the case, Mr. Justice Stone said : “The testimony newly discovered is parol proof — consists in swearing — and is cumulative. It relates to a fact or facts — the execution and last-known custody of the deed — which were in issue, and were attempted to be proved, in the former suit. The rule is inflexible, and rests in the soundest public policy, that parol cumulative testimony to a fact in issue in an original cause cannot be the foundation of a bill of review. To allow such testimony would invite to the greatest abuse — nay, sometimes, to subornation of perjury — and would tend to almost endless litigation. 1 Interest reipublicss ut sit finis litium.’ ” — citing Caller v. Shields, 2 Stew. & P. 423 ; Dexter v. Arnold, 5 Mason, 312 (Fed. Cas. No. 3,856); Respass v. McClanahan, Hardin, 342; Livingston v. Hubbs, 3 Johns. Ch. 124; Her v. Routh’s Heirs, 4 Miss. (3 How.) 292; Foy v. Foy, 25 Miss. 212 ; Love v. Blewit, 1 Dev. & Bat. 108 ; Head v. Head’s Adm’r, 3 A. K. Marsh. 121; Norris v. Le Neve, 3 Atk. 36 ; Taylor v. Sharp,. 3 P. Wms. 371; Willan v. Willan, 16 Ves. 88; Young Y. Keighly, 16 Ves. 351.
If these authorities are to be followed — and they are *244certainly entitled to very great consideration, both on account of their numbers and of the recognized ability of the several courts and judges — this case is at an end. There is no pretense that the new proof or newly-discovered evidence upon which it is sought to impeach and set aside the former decree consists of anything more than the oral testimony of witnesses concerning a fact in issue in the former case, and upon which there was much testimony on both sides. But whether it can be truly said that there is any universal or absolute rule which prohibits courts from allowing the introduction, upon a bill of review, of the oral testimony 'of newly-discovered witnesses to prove an issue on the former hearing, it is agreed by all the authorities that such evidence, if allowed at all, should be permitted with very great caution, and only in extreme cases, for fear of opening the doors to perjury and subornation of perjury, and that, where the new evidence consists in the mere accumulation of witnesses to a fact once litigated, a bill of review should rarely, if. ever, prevail: Long v. Granberry, 2 Tenn. Ch. 85 ; Craig v. Smith, 100 U. S. 226; Peters v. Bowman, 98 U. S. 56 ; McDowell v. Morrell, 5 Lea, 278 ; Kimberly v. Arms, 40 Fed. 548; Society of Shakers v. Watson, 23 C. C. A. 263, 77 Fed. 512. The newly-discovered evidence sufficient to support a bill of review, whatever its nature, should be of so clear and decisive a character as to leave no doubt that it would of itself compel a reversal of the former ruling; and, as said by the chancellor in Long v. Granberry, 2 Tenn. Ch. 85, “New proof of so conclusive a character will rarely if ever be found, unless it be documentary,” for, as said by McFarland, J., in Burson v. Dosser, 1 Heisk. 763, “whether it [the new testimony] would be sufficiently strong to change the result could not be determined from the bill, but only upon consideration of the additional testimony, when taken *245and considered with the other evidence in the original cause. If this be a good ground for a bill of review, there would probably be but few seriously contested cases where grounds equally strong might not be presented. Parties, after a trial, and after discovering the ground upon which they failed, would too often discover that upon another trial they could maintain their side of the contest with more evidence and greater skill. This would open the door to endless frauds and perjuries.”
So that whether we look at this case from the standpoint of the authorities which hold that the discovery of new witnesses upon a point already litigated will under no circumstances support a bill of review, or those which refuse to be bound in all cases by a rule so strict and inflexible, the decree of the court below cannot be upheld. The oral testimony of the newly-discovered witnesses, as given on the trial of this suit, tends strongly, no doubt, to support the theory of the plaintiff in the suit of Nessley v. Ladd, and might have been sufficient, had it been given on the trial of that suit, to have brought about a different result. But manifestly it is not that conclusive and irrefragable evidence which is required in a proceeding of this kind. It is the mere recollection of witnesses of an event occurring a quarter of a century ago, and its probable effect can only be determined when it is considered in connection with the evidence in the original cause; and, when so considered, it leaves the matter in more or less doubt, and there still remains room for a very strong argument, upon the undisputed and undeniable facts, in support of the defendant’s position. If, under such circumstances, an original suit in the nature of a bill of review can be maintained, there will scarcely ever be an end to litigation. The policy of the law is not to protract or continue litigation, but to end it; but if whenever a new witness or witnesses can, honestly *246or by subornation, be found, whose testimony might probably have been sufficient to have justified a different finding and decree in a suit already terminated, an original suit can be maintained on that account, there never could be a certainty that a cause once tried out and passed to a final decree was ended. The danger and mischief to society which would naturally flow from a rule of that kind are too great and too universal to justify its sanction by the courts. It follows that the decree of the court below must be reversed, and a decree entered here dismissing the bill. Reversed.