The defendant appeals from a judgment of conviction of perjury.1 He contends that his alleged false testimony was not material to the question under inquiry and consequently that the conviction cannot be sustained. Since materiality is a question of law for the court2 we shall examine the prior proceedings to determine whether the testimony was material.
The defendant and one other were indicted jointly and charged with violating restrictions and rationing orders as to new rubber tires and tubes. The indictment was in four counts. At the conclusion of the Government’s case the court directed a verdict for the defendant upon counts III and IV. The jury found the defendant not guilty upon counts I and II. Count II charged that the defendants conspired to defraud the United States on December 12, 1941 and divers dates theretofore and thereafter by making false bills or invoices, by failing to keep any records of sales or transfers, and by failing to file an inventory report of new tires and tubes.
During the course of the trial the Government proved that the Office of Price Administration and the Office of Production Management required those who had new rubber tires and tubes on hand December 12, 1941, the date of the order freezing the supply, to make out an inventory report upon Form PD-216 and that the defendant had not filed this report. Upon cross-examination the defendant testified that he received Form PD-216 on January 5, 1942, that the instructions thereon were to file the report on or before December 31, 1941 and -that since he could not file it on time he did not file it at all. He was asked by the trial judge whether he had previously stated that he did not file the report because he was winding up his business. He said this was so — that the report had to be filed every three months. The trial judge then asked the defendant whether he had filed a report for the first three months of 1942. The defendant replied “The first one I couldn’t file because on the face of it it was void, and the next one came three months later, [i. e. in March, 1942] and I had no stock.” The United States Attorney asked the defendant if he had not in fact purchased new tires in March, 1942. The defendant answered that he had not. He was thereupon shown a letter addressed to the United Petroleum Corporation, dated March 24, 1942, signed by him with his affidavit affixed, in which he stated that he had purchased 170 tires from that company on March 16, 1942. The defendant explained that the tires were in fact purchased by one Daniel L. Roatche, to whom he loaned the money for the purpose, but that he gave the letter stating that he had himself made the purchase in order to help out one Bernard E. Sweeney, an employee of the seller, who had become involved in the sale of the tires to Roatche.
In the trial upon the prejury indictment the Government undertook to prove that this testimony was false and that the defendant had in fact purchased and received 170 tires in March, 1942. The defendant contends that even if false this testimony was not material to the subject under inquiry and that the trial judge should have so decided as a matter of law and directed a verdict for the defendant.
We think that the testimony was as to a material matter. The reason for the defendant’s failure to file the report in 1941 was the subject under inquiry. The Government contended it was with the intent to conceal and thereby to defraud the United States. The defendant claimed his failure to file the report was quite fortuitous and was due solely to the fact that he received the printed form too late for timely use. His failure to file subsequent forms which did arrive on time, therefore, shed light upon the credibility of his explanation. The question is whether the credibility of a witness is material to the *206inquiry in the sense in which that term is used in the perjury statute.3 We think it is. It is well established that the false swearing need not be to the fact immediately in issue.
In United States v. Shinn, C.C.Ore., 1882, 14 F. 447, 453 the court said: “But when the superfluous 'or collateral matter is calculated and inténded to prop and bolster the testimony of the witness on some material point, as by clothing it with circumstances which add to its probability or strengthen the credibility of the witness,” it is material. In Luse v. United States, 9 Cir., 1933, 64 F.2d 776 it was held that testimony which if believed would tend to establish the credibility of the witness was material. In United States v. Slutzky, 3 Cir., 1935, 79 F.2d 504, this court pointed out that a material matter does not mean necessarily a matter that directly affects the ultimate issue of the trial but means one that is legally capable of being proved in the cause. In Goins v. United States, 4 Cir., 1938, 99 F.2d 147, certiorari dismissed 306 U.S. 622, 59 S.Ct. 783, 83 L.Ed. 1027, it was held that a witness commits perjury by swearing falsely not only as to the fact which is immediately in issue but also when he swears falsely as to material circumstances which have a legitimate tendency to prove or disprove such fact. In Blackmon v. United States, .5 Cir., 1940, 108 F.2d 572, 573, the court said “The materiality required is not as to any particular issue in the case, but as to the trial as a whole, that is, materiality is determined by whether the false testimony was capable of influencing the tribunal on the issue before it.”
Applying any "one of the above tests we conclude that the defendant’s testimony as to the purchase of tires in March 1942 was material. It follows that the indictment based upon the alleged false testimony in the prior criminal proceedings was well founded. The court accordingly did not err in overruling the defendant’s motion in arrest of judgment.
The defendant next assigns as error the refusal of the trial judge to charge that in a perjury case the Government must establish the falsity of the statement made by the defendant by the testimony of two independent witnesses or one witness and corroborating circumstances.4 This rule of evidence has been the subject of cogent criticism.5 Nevertheless, save for some exceptions not here relevant,6 it continues to be applicable to perjury cases in the federal courts.7 We, therefore, must begin with the premise that in the federal courts no defendant may ordinarily be convicted upon a charge of perjury which is supported by the testimony of but a single witness. It does not necessarily follow, however, that a defendant in a perjury case is entitled to have the jury instructed as to this rule of evidence. On the- contrary not only historical considerations but practical considerations as well lead to the conclusion that the jury have no role to perform in its application. Moreover, as we shall see, even those courts which in theory *207have accepted the view that the jury are to be instructed upon the rule have in practice shown an increasing willingness to condone tlie refusal of the trial judge to give such instructions.
We will first glance briefly at the history of the rule and its nature.7 8 In early times most perjury cases were tried in the Court of Star Chamber. That court, which followed the canon law, had absorbed much of the civil law in its procedure. In particular it had adopted the quantitative theory oí evidence. This involved a counting of the oaths given in a judicial proceeding with victory going to the party with the greater number on his side. It was the intrinsic value of the oath rather than the credibility of the witness’s testimony which was considered important. The Court of Star Chamber accordingly required two oaths against the defendant’s one to sustain a charge of perjury. In 1640 when the Court of Star Chamber was abolished by statute jurisdiction over the crime of perjury was transferred to the common law courts and with it was transferred the two oaths rule. That tire rule thus transferred was quantitative in its nature is indicated by the reasons which the common law courts gave for its acceptance in perjury cases but not in other criminal cases. In the Seventeenth Century when the transfer of jurisdiction took place defendants in criminal proceedings in the common law courts were not permitted to testify in their own defense. In a perjury case, however, the defendant’s allegedly false oath was before the court and jury. The courts accordingly reasoned that the prosecution should be required to produce two oaths to contradict the defendant’s one since otherwise there would be the legal stalemate of one oath against another. This, it will be noted, was a purely quantitative idea.
As originally taken over into the common law the rule, being quantitative in nature, required that the oaths of the two prosecution witnesses should both be directly adverse to the defendant’s oath.9 It is true that the rigor of the two oaths rule has long since been ameliorated by permitting evidence of corroborating circumstances to take the place of the second oath. But this modification came about in order to facilitate the conviction of perjurers. It was certainly not designed to provide them with additional opportunities to escape conviction. It would thus appear that the modification was not intended to change the fundamental nature of the rule to the extent of presenting a jury question where none had theretofore existed. Considered quantitatively the rule is purely one of law. The trial judge is in a position to determine whether the burden imposed by the rule upon the prosecution in this respect has been met. If it has been met he can submit the case to the jury. If the burden has not been met he can direct a verdict of not guilty.
When we turn to a consideration of the rule in its practical operation we likewise reach the conclusion that it should be considered as purely quantitative in nature and as raising no question for the jury. Considered from the' standpoint of the jury’s task the interjection of the rule into their deliberations would impede rather than aid them in reaching a right decision upon the sole issue with which they are concerned, the guilt or innocence of the defendant. For if such an instruction is to mean anything it must mean that the members of the jury must make an independent decision as to the credibility of each government witness before they determine the guilt of the accused and, if they find themselves unable to credit the corroborating witness whom the Government has offered, they must acquit the defendant even though they may be so well satisfied of the credibility of the principal government witness as to be convinced beyond doubt of the defendant’s guilt. Anyone who has had practical experience in the conduct of jury trials knows that the members of a jury who are thus convinced of the defendant’s guilt cannot be expected to pay much attention to a .rule such as this which, even if understood, is likely to be viewed by them as legalistic and technical.
The traditional sphere of the jury is to determine from the evidence whether the guilt of a defendant has been established beyond a reasonable doubt. This they do by considering all the evidence which is presented to them, appraising its credibility, *208weighing it' and reconciling it, if possible. This function is one which Anglo-American experience has shown that a jury is well qualified to perform. But it is quite a different matter to ask the twelve laymen who constitute a jury to engage in the difficult mental feat of giving simultaneous but wholly independent consideration to the question of the defendant’s guilt from the standpoint of the testimony of each individual witness without regard to the testimony of the others and to declare the defendant not guilty if they find the story of a corroborating witness not credible even though they are so thoroughly satisfied by the other testimony of the defendant’s guilt that the discredited testimony does not even raise a reasonable doubt in their minds. We think it is unreasonable to ask the members of a jury to perform such a task and unrealistic to think that they will be able to do so if asked.
Moreover it seems to us that the application of the rule to the deliberations of a jury would necessarily thwart rather than further the ends of justice since it would have no other practical effect than to require the acquittal of a defendant whom the jury, after considering all the evidence, believed beyond a reasonable doubt to be guilty. If the corroborating evidence is so incredible as to raise a reasonable doubt in the jury’s minds as to the defendant’s guilt they will acquit him under their general instructions without reference to the rule under discussion. If, on the other hand, the evidence of the principal witness is so clear and convincing as to remove from the jury’s minds any doubt as to the defendant’s guilt despite their inability to credit the corroborating evidence, they will under their general instructions convict him. It is, therefore, only in this situation that the two witnesses rule could have any effect upon the jury’s deliberations and its only possible effect could be to bring about the acquittal of a person whom the jury have found from the whole evidence to be guilty. We do not think that the processes of the criminal law should be permitted to accomplish any such jail delivery as this. Although, as we have indicated, the rule has been much criticized, nevertheless considered as a quantitative rule its application in perjury cases may perhaps be justified as affording a guarantee of the good faith of the proser cution. But to apply the rule to the deliberations of the jury is to transform it from an earnest of good faith on the part of the government into a loophole for the escape of perjurers from justice. No principle of law or public policy requires that of all criminals perjurers alone10 should be singled out for such special protection.
Finally we think that recent decisions upon this question indicate that in practice the courts no longer adhere to the theory that the defendant is entitled to have the jury instructed upon the rule, even though they may still pay lip service to it. While it is true that the rule has not been expressly rejected as a guide to the jury’s action, it has been deprived of all practical application as such by the decisions which have held that the refusal of the trial judge to give instructions upon the rule did not substantially prejudice the defendant where it appeared that the quantum of evidence required by the rule was in fact offered by the Government.11 Certainly, if the jury *209have any function to perform under such instructions, the performance of which might conceivably lead to the acquittal of the defendant rather than his conviction, the refusal of the instructions could hardly be said not to prejudice his rights. We have shown that the only possible effect of such instructions could be to secure the acquittal of a defendant who would otherwise he convicted.
Our consideration of the subject convinces us that the question whether the government has offered sufficient proof to meet the rule as to the quantum of proof required in a perjury case is and should be treated by the trial judge as purely a rule of law to be applied by him in determining whether the case should be submitted to the jury or a verdict of not guilty directed. If the Government has met its burden by offering the evidence of two witnesses or of one witness and corroborating circumstances the case should be submitted to the jury for consideration upon all the evidence, as in any other criminal case. Under these circumstances to instruct the jury as to the requirements of the rule of law which the trial judge has already applied is not only unnecessary but, we think, unwise and inappropriate since the introduction of such an irrelevant matter may well tend to confuse the members of the jury and to deflect their attention from the question which has been submitted to them, namely, whether upon the entire evidence the defendant’s guilt has been proved beyond a reasonable doubt. We accordingly conclude that the trial judge in this case was right in refusing to give the instructions requested.
It remains to be seen whether the evidence presented by the Government did as a matter of law satisfy the requirement of the quantitative rule of evidence applicable to perjury cases. We think that it was sufficient to satisfy the strictest requirement of the quantitative rule. The defendant’s testimony which the Government alleged to be false was that he had not purchased tires in March, 1942. The Government proved by Bernard E. Sweeney, tire sales manager of United Petroleum Corporation, that the defendant agreed to purchase tires from that company and that on March 16, 1942, at Paulsboro, New Jersey, he supervised the delivery oí 170 tires to Evans, the defendant’s manager. Sweeney identified the truck operated by Evans as one which he had often seen about the defendant’s place of business. Daniel L. Roatche testified that he was present when the defendant agreed to purchase the tires and when the tires were placed upon the defendant’s truck at Pauls boro, New Jersey, and into the custody of Evans. Joseph J. Gabel and John F. Oates, president and office manager respectively of United Petroleum Corporation, testified that the defendant acknowledged to them on or about March 24, 1942 that he was the purchaser of the 170 tires. The Government also offered in evidence a letter signed and sworn to by the defendant in which he acknowledged that he had purchased the 170 tires.
It will thus be seen that the fact of the purchase was testified to by more than two witnesses and was shown as well by corroborating evidence, the letter to which we have already referred. The jury was, of course, at liberty to believe the defendant’s version that the apparent sale of the tires to him was made just to cover up the sale of the tires to Roatche and Sweeney and to disbelieve Sweeney’s and Roatche’s denial that the tires were actually purchased by Roatche. This, hswever, was solely a question of the credibility of the evidence and not of its amount.
We conclude that the trial judge properly decided as a matter of law that the quantitative rule had been met by the Government’s evidence and that he did not err in refusing the defendant’s request that the jury be instructed upon the rule.
We have examined the^ other assignments of error and find them to be wholly without merit.
The judgment is affirmed.