Mr. Justice Morris
delivered the opinion of the Court:
■ This is an appeal from the decision of the Commissioner [93]*93of Patents in an interference case, wherein the matter of controversy between the parties is an invention stated by the Office in the following terms:
“A fifth-wheel having an annular track-plate, the upper side edges of which are respectively provided with concentric grooves running around the track-plate at uniform distances from each other, a cover having an inverted-U-shape form, the lower edges of the sides of the cover running in the grooves of the track-plate, and the outer faces of the sides of the cover being in approximate alignment with the sides of the track-plate, and an annular traveling section running circularly within the cover and upon the track-plate, the traveling section consisting in two annular and concentric bands respectively adjacent to the inner faces of the sides of the cover, and the traveling» section also consisting in a series of rollers, mounted to loosely revolve upon spindles which are respectively engaged in the bands of the traveling section, the upper face of the track-plate and the inner face of the upper portion of the cover respectively bearing on the lower and upper sides of the rollers, so as to support the cover movably on the track-plate.”
Fifth wheels to a wagon, it seems, are not as useless as the common remark in regard to them would seem to indicate; for they have been the subject of several patents, and the present controversy is a contest between one holding a patent, which is sought to be reissued, and an applicant for a patent.
It appears from the record that the appellee, William H. Bradshaw, on December 29, 1896, filed his application in the Patent Office for a patent for the invention in controversy, and that a patent was issued to him on March 30, 1897. Subsequently, on account of an alleged misdescription in the original specifications, he made an application, on December 1, 1897, for leave to surrender the patent and to have it reissued to him. While this application was pending in the office, Christen Nielson, the other party to [94]*94the interference, on June 18, 1898, filed his petition for a patent for the same invention. For the purpose of having an interference declared, he copied Bradshaw's claim, and introduced it by amendment into his own application; and thereupon, on August 3,1898, an interference was declared.
Nielson filed his preliminary statement, in which he alleged that he had conceived the invention in the latter part of November, 1895; that he made drawings of it and explained it to others about the same time; that he never made any model of it, but embodied the invention in a.full sized machine in January and February of 1896; and that he had been manufacturing such machines since September of 1897.
Bradshaw, in his preliminary statement, alleged that he had conceived .the"invention about September 29, 1896; that he had made disclosure of it about November 20,1896, and had made a sketch of it about December 15,1896; but that he had made no model of it, had built no full sized machine, and had not manufactured or sold the article.
Bradshaw took no testimony in support of his preliminary statement. There was some testimony, it is true, on his behalf; but it was not to support his preliminary statement, but to controvert some testimony introduced by the other party. He stands, therefore, for his conception of the invention and its constructive reduction to practice, on the date of his original application for a patent, December 29, 1896 — which, however, is only a little more than one month later than the date of his alleged disclosure of it, according to his preliminary statement. „
Nielson was the junior applicant; and on him was the burden of proof to maintain his cause. The substantial testimony in the case is that taken on his behalf. Upon this testimony the acting examiner of interferences decided in his favor; but that judgment was reversed by the board of examiners, which awarded priority of invention to Brad-' shaw. The Commissioner of Patents, upon appeal to him, [95]*95affirmed the judgment of the board of examiners. And from the decision of the Commissioner the junior applicant, Nielson, has brought the case to this court by appeal.
1. A preliminary question has been raised here which should-be determined. Among the seven reasons of appeal assigned by the appellant are two to the effect that it was error on the part of the Commissioner not to have considered two affidavits alleged to have been filed or presented in the Patent Office while the case was pending before the Commissioner on appeal. One of these affidavits was that of the appellant himself seeking to correct an exceedingly important date in his own testimony, which is claimed on his behalf now to have been a clerical error. The second affidavit was that of one Daniel Macy, intended to supplement a deficiency in the appellant’s testimony. These two affidavits were not transmitted to this court originally as part of the record; but upon writ of certioram the Macy affidavit was sent up, and it appeared that the appellant’s own affidavit had been returned to him and was not in the office.
Of course, these affidavits constituted no proper part of the record, and were not proper to be considered by the Commissioner as affecting the testimony in the case.. Testimony in patent cases of interference is taken substantially according to the rules which prevail in equity; and it is one of the cardinal principles of our jurisprudence that testimony, both in equity and at common law, must be taken under the safeguard of the right of cross-examination. Testimony by affidavit has no part or place in our system, except for collateral purposes in exceptional cases. Affidavits may lay the foundation for the introduction of testimony, but they are not themselves testimony; and least of all could they be allowed to contravene testimony already taken. Affidavits may be proper in the Patent Office, as they are in the courts of law, as the basis for an application to reopen a cause and to permit further testimony to be taken; but it is an unheard-of thing that they themselves [96]*96should be received as testimony. This is a fundamental rule of our law which seems to have been ignored by the appellant in this instance; and it is clearly no answer to the requirement of the rule to argue, as it is argued on behalf of the appellant, that the appellee had due notice of the affidavits and had opportunity to controvert them. The appellee was called upon to answer, only testimony presented in due form of law. Not only was there no error in the action of the Commissioner in regard to these affidavits, but it would have been gross error to have considered them at all for the purpose for which they were presented.
2. There is considerable discussion in the briefs of counsel, as there was in the oral argument before us, with reference to the meaning of a term of considerable importance in the case, that of “Evans rollers” as the designation of the traveling section of the fifth wheel which constitutes the issue of interference in this case. It appears that, on December 4, 1883, a patent had been issued to one Thomas Evans for a fifth wheel containing rollers of a certain kind different from the rollers of this issue; so that if the “Evans rollers,” claimed by the appellant to have been used by him in the construction of the machine introduced by him in evidence as his reduction to practice of the invention of the issue, are the rollers of the patent to Thomas Evans, his claim to the invention of the issue wholly fails; for this issue involves the use of rollers radically different from those of the Evans patent. It is urged, on behalf of the appellant, that while the board of examiners and the Commissioner of Patents were disposed to regard the “Evans rollers” mentioned in the testimony as the rollers of the Evans patent, although their decision is not based on that point, yet the parties in the course of the testimony understood by the term the same rollers as those involved in the issue. This understanding does not seem to be conceded on behalf of the appellee; and the Macy affidavit, already referred to, was filed by the appellant for the purpose of affecting or [97]*97modifying the testimony in that regard. It is alleged, as one of the reasons of appeal, that the Commissioner erred in arbitrarily assuming that the “Evans rollers” referred to in the record were the rollers of the Evans patent, and not the rollers of this issue.
It is unnecessary for us further to pass upon this point than to say that, inasmuch as the burden of proof was upon the appellant to prove his case beyond a reasonable doubt, it was incumbent upon him to remove the doubt in this regard. This he has not done. It could not have been done by the Macy affidavit. There was no doubt a proper way wherein to do it; but the appellant has not thought proper to pursue the course usual and proper in such cases. Undoubtedly, the appellant must have known of the Evans patent and of the rollers therein mentioned; and if he did not' have actual knowledge of them, which is not probable, he is chargeable with such knowledge. It was therefore very plainly incumbent upon any applicant for a patent for the invention involved in this issue to have distinguished in his testimony between the rollers of the Evans patent and the rollers of this issue, if both were commonly known by the name of “Evans rollers,” as the appellant would have us infer. It is not sufficient to do this in argument; it should have been done in testimony. It is not sufficient to say that both parties understood the term in the same way. The correctness of that statement is not admitted by the appellee, and the understanding should appear from the record, and not be advanced merely in the argument of counsel.
3. Biit the principal and the important question in this case is one of dates. Asssuming that the appellant did at some time have the invention in issue, and that he did not first take it from the appellee’s patent, did his conception of it and reduction to practice antedate the 29th day of December, 1896, the day of the appellee’s application for a patent? In order that this question should be solved in [98]*98bis favor, the appellant is required to show beyond a reasonable doubt that he did have a conception of the invention and that he reduced it to practice before the day specified. For, inasmuch as the appellee is the holder of a patent, the burden of proof is on the appellant, as the junior applicant, not merely to prove his case by a preponderance of evidence, but to establish it beyond reasonable doubt. This is the well-established rule, repeatedly affirmed in the decisions of the Supreme Court of the United States and of this court.
Now, how does the testimony on behalf of the appellant conform to the requirements of that rule? It may be that, if the case were one at common law and to be determined by a mere preponderance of evidence, a verdict of a jury in favor of the appellant would be approved and sustained. But has the appellant established his case beyond a reasonable doubt? We think that there are very grave doubts in the case which militate against the appellant, not sufficient perhaps to cause the evidence in our opinion to preponderate against him, but certainly sufficient to justify the application against him of the rule to which we have referred.
Apart from the fact that the whole testimony on behalf of the appellant is not as definite and satisfactory as it seems to us it could have been made, if he was in fact, what he claims to be, the first and true original inventor of the device in controversy, there are two or three features of the testimony which give rise to a remarkable degree of doubt. The first of these is when the appellant, in his own deposition, being interrogated by his counsel as to the time when he conceived the invention set forth in the issue, promptly answered, “The latter part of February, 1897.” That the counsel fully understood the fatal character of this answer, if it was allowed to stand, seems to us to be clearly evidenced by the line of inquiry immediately thereafter pursued by him, and especially by the ■ question which he first [99]*99propounded, which was whether the witness understood the meaning of the term “conceived,” to which inquiry the appellant returned the answer that he did understand it. Now, if this testimony of the appellant is correct and states the truth, it disposes absolutely and conclusively of his case; for Bradshaw’s unquestioned and unquestionable date, that of his application, is anterior.
But it is argued by counsel that the appellant’s statement that he understood the meaning of the term “conceived,” and that he had conceived the invention in issue in “the latter part of February, 1897,” is, as he styles it, “a manifest clerical error.” A “clerical error” it certainly is not; a mistake or misapprehension it may have been, and probably was. But if it was a mistake or misapprehension, there was a way to correct it; and that way was not by the presentation to the Commissioner of Patents of the appellant’s affidavit, to which we have already referred, and by which, as we understand, — for very properly that affidavit is not in the record before us, — it was sought to have the Commissioner accept the contents of the affidavit as evidence in the place of the corresponding portion of the appellant’s deposition. The appellant might have been recalled, or other testimony might have been introduced in the regular way, to correct the mistake, if mistake there was. And if the fact was, as to which we are not satisfied, that the mistake was not discovered until the decision of the board of examiners was promulgated, we do not understand that there was no way open for its correction even at that stage of the case. That counsel felt the necessity for correction is shown by the fact that he presented an affidavit to the Commissioner for that purpose; and by the further fact that he has alleged as error here the refusal of the Commissioner to accept such affidavit as evidence.
The testimony of the appellant in this regard, even if the remainder of his deposition be regarded as satisfactory, leaves room for-very grave doubt upon a vital point in his [100]*100case; and the remainder of his testimony does not suffice to remove the doubt.
An exhibit 5 has been introduced in evidence by the appellant, which, as it now appears and as it appeared when offered in evidence, is a full-sized machine embodying the invention of the issue, and which the appellant claims to have made in the latter part of June, 1896. Upon the making of this exhibit at the time mentioned the appellant’s case mainly, if not exclusively, depends, for the proof of his conception and reduction to practice of the invention in anticipation of Bradshaw. And yet the testimony in regard to this exhibit is most unsatisfactory, when, if the appellant had a good case, it might have been made clear and satisfactory beyond question.
No witness except himself testifies as to the kind of'rollers that were used in this machine. Indeed, it is not to be deduced with any great certainty from the testimony of the other witnesses that there were any rollers in this machine in the year 1896, during which they say it lay in the appellant’s shop. The testimony of the appellant, apparently uncorroborated, is that he put “Evans rollers” in it; but we have seen what doubt there is in regard to the meaning of the term “Evans rollers,” a doubt which it was incumbent on the appellant to remove, as well, as other doubts in the case. These doubts can not now be removed by conjecture, or by any plausible theory as to the probability of the situation. It was an essential part of the appellant’s case to prove that he used the special rollers of this issue in the machine which he claims to have made in June, 1896, and to prove that fact beyond reasonable doubt. He leaves the fact in very great doubt, even though it should be conceded that the preponderance of the evidence is in his favor.
From this important exhibit No. 5 a piece was cut and taken to the appellant’s counsel “for advice,” as the appellant testifies, “if there was anything to prevent me from making a wheel of it,” — to quote his own words — by which, [101]*101as he explains afterwards, he meant whether there was anything to prevent him from manufacturing the article. And this he says he did some time in July, 1896. On the contrary, his principal witness, Eugene Bastion, who worked for him, testified that he, Bastion, cut off the piece himself in 1897 — under what circumstances or for what purpose he does not state. Here is an important discrepancy in dates, which there is no attempt to explain or reconcile.
Again, if the piece so cut out was taken to the appellant’s attorney for advice, it would seem not to be difficult to produce it, or to account for its loss; and it would seem not difficult to find some corroboration or further proof of the time at which it was taken to the attorney. The attorney does not now hesitate to state in argument that it was taken to him in 1896 ; but there was no offer of testimony by him to show that. Nor was there any offer of testimony to show what the condition of the section was at the time, and whether it had in it a section of rollers such as are involved in the issue of the interference. All this was vital to the case, and might have been decisive of it; and yet there was a strange omission to introduce any testimony upon any of these points. Such omission is greatly suggestive of grave doubt whether testimony in that regard, if offered, would have been beneficial to the appellant.
Outside of the appellant’s own testimony, there is but little evidence to sustain his case. No one of the other witnesses testifies that the wheels alleged to have been manufactured by the appellant in 1896 had in them the rollers of the present issue; and indeed from the testimony of these other witnesses alone it is not at all certain that they had any rollers whatever in them at that time. The appellant alone testified with distinctness that the wheel manufactured by him, designated as “Exhibit No. 5,” had Evans rollers in it; and we have seen what hopeless uncertainty there is in regard to the meaning of the term “ Evans rollers.” All this uncertainty, it seems to us, might have [102]*102been cleared up by the production of the section cut from the wheel and taken to the attorney for his opinion, and proof of the time when it was so taken.
A reasonable implication from the account given by the appellant of his taking the section to his attorney is that he desired to know whether he would infringe any rights of others by manufacturing such a wheel. He was threatened by Bradshaw in October, 1897, with a suit for the infringement of his (Bradshaw’s) patent, if he persisted in manufacturing the wheel in question; and it is no more than reasonable to refer his consultation with his attorney to this time. But if he did consult his attorney in 1896, as he claims to have done, to see whether he was infringing upon any rights, and the attorney then saw, as he must have seen, if such was the fact, that Nielson had a new and patentable invention, it is most extraordinary that he did not then apply for a patent for himself for the invention, and most extraordinary that the application was postponed for nearly two years thereafter, when Bradshaw in the meantime conceived the invention, made and prosecuted his application, and procured his patent. Assuredly such delay under the circumstances, while not claimed to constitute laches on the part of the appellant, is a remarkable circumstance that tends to throw discredit on the appellant’s whole story in regard to the date of his alleged invention.
Were we to consider the testimony with reference simply to the question of the preponderance of the evidence, we would still be left in very grave doubt.
The appellant claims to have conceived the invention in controversy as early as February, 1896, assuming that his statement in his testimony that he did so in February, 1897, is merely a mistake; and he claims to have then embodied it in a wheel. But this wheel he admits he showed to no one, and no one saw it; and it was immediately broken up, and never submitted in any manner to trial or test for use. This statement is wholly uncorroborated by evidence of any [103]*103kind; and it may therefore be dismissed from our consideration.
The appellant’s case depends entirely o.n his alleged manufacture of exhibit 5 in June, 1896. As we have stated, exhibit 5, as introduced in evidence, embodies the invention of the issue, and if it was made by the appellant in June, 1896, he should be held to have anticipated the appellee. But the testimony to show that this exhibit 5 was made in June, 1896, is unsatisfactory. The appellant himself testifies with sufficient positiveness to the fact, always assuming that his statement that his conception of the invention was in February of 1897 is a mistake; but there is no written or record evidence of any kind to support him; and his witnesses, in our opinion, fail to support him satisfactorily.
His statement is that, in June, 1896, he made ten such wheels; that one of them was subsequently sold, he does not know to whom ; that the other eight were still lying in his shop; and that exhibit 5 was the only one of the ten which had rollers fitted into it. And he admits that exhibit 5 was the only wheel ever made by him before the date of the appellee’s application which contained the invention of the issue.
Two witnesses were called by the appellant in support of his statement, workmen in his service, one of them his brother in law, David Kaiser, and the other one Eugene Bastion. Kaiser, who worked on the wheels, testified that he knew nothing of any rollers being put into any of them. The importance of his testimony is evident when it is remembered that the rollers are an essential element of the issue, and that the wheels, without rollers, or with any other rollers than those specifically described in the issue, would not constitute the invention in controversy. Bastion, the other witness, testified that he first saw the wheel exhibit 5, with the rollers in, the last of June, and before he took charge of the engine on the premises, which was on July [104]*10411, 1896; that he did not work on the wheel himself at that time; that two men, whom he calls Lew and Dave, and the latter of whom he identifies as the preceding-witness, David Kaiser, worked upon it, and put the rollers in; that he himself did not go to work on the wheels until after July 11, 1896; that the wheels were then ready, with the exception of the rollers, and that he went to work to put the rollers in, and that the section which was wanting in exhibit 5 had been cut out by him in 1897; but that he did not remember in what month.
One Henry L. Mangan, to whom the appellant in his testimony referred as one of the persons who had seen the wheels in question, including exhibit 5, in June, 1896, was called by the appellee; and he testified that he'had never seen these wheels, or exhibit 5, until the latter was introduced into this case, and that in 1896 the appellant did not use the rollers involved in this issue, but certain Berger rollers. The force of this testimony it was sought to break in rebuttal by showing a condition of ill-feeling between this witness and the appellant.
Such is substantially the whole testimony on the vital point of the case; and its insufficiency is quite palpable. There is a remarkable inconsistency between the testimony of the appellant and that of his witness Bastion in two important particulars. The former says that the missing section of exhibit 5 was removed in July, 1896, in order to have it taken to his attorney for his opinion on the question of its infringement of existing patents. The latter says that he himself cut out the section in 1897, for what purpose and under what circumstances he does' not state. It was very important to the appellant’s case to determine at what time this cutting was done; and yet we are left in doubt upon the point; or rather we must accept the testimony of the witness Bastion that it was done in 1897 in preference to that of the appellant that it was done in 1896, there being nothing whatever to corroborate one rather than [105]*105the other. Again, the appellant testifies that rollers were pnt into only one of the wheels claimed to have been made in June, 1896, and that no rollers were ever put into the other nine, while Bastion says that when he went to work on July 11, 1896, the wheels were all ready, except the putting in of the rollers, and that he himself then proceeded to put in the rollers. These statements can not both be correct; and both statements are discredited by the statement of Kaiser, the other witness, the man who worked upon the wheels in June, 1896, and who certainly must be presumed to have accurately known their condition, to the effect that he knew nothing of any rollers being put in any of the wheels.
Messrs. Church & Church, for the appellee, opposed the motion.
Apart from any consideration whatever of Mangan’s testimony, it is quite clear that the testimony on behalf of the appellant is open to very grave doubt as to its sufficiency to establish even a prima facie case on his behalf; and certainly, if the rule is applicable that he must prove Ms case beyond a reasonable doubt, there can be no question that he has fallen far short of the requirement. Under any aspect of the case, we find nothing in the record that would warrant us to overthrow the effect of two well-considered decisions of the Patent Office against the appellant.
For the reasons stated, we are of opinion that the appellee is entitled to judgment of priority of invention, and that the decision of the Commissioner of Patents awarding it to him should be affirmed.
The clerk of the court will certify this opinion and the proceedings in this court to the Commissioner of Patents, according to law.
On March 9,1900, a motion was made by Messrs. Schreiter & Matthews, on behalf of the appellant, for leave to apply to the Patent Office to reopen the interference and introduce additional evidence.
[106]*106On March 15, 1900, the motion was denied, Mr. Chief Justice Alvey delivering the opinion of the Court:
This motion by the appellant made in this court, for leave to apply to the Patent Office, to have the interference that has been finally determined adversely to the claim of the appellant reopened in that office, with permission to introduce additional evidence to supply defects and omissions that are supposed to exist, must be denied. There is no precedent or practice that would justify the granting of such motion; and to grant it, under the circumstances of this case, would make a bad precedent, and one that would likely be productive of mischievous consequences.
The defects in the evidence had been called to the attention of the appellant in the Patent Office while the case was pending there, and before the appeal was taken to this court. If the appellant had desired .to supply the defect in his evidence, then and there were the time and place to make the motion for leave to do so, under the rules and according to the practice of the office, if there be any established practice upon the subject that would authorize such proceeding. The appellee should not be subject to the trouble and expense attending an appeal to this court by the appellant, upon a defective state of the record which the appellant could have corrected in the Patent Office, by making a seasonable application for leave to do so. This court can only dispose of the ease upon the record as it is filed here; and it has no power to authorize the taking of new or additional testimony upon a motion like the present. It can only deal with the patent appeals in the manner directed by the statute; and the most that could be done in such case, under any circumstances, and before transmitting the certificate to the Commissioner, as required by the statute, would be to give leave to the appellant to make a motion in the Patent Office upon the filing of the certificate to reopen the interference with leave to produce the proposed evidence. But that motion would necessarily be [107]*107addressed to the discretion of the Commissioner of Patents, and it would rest with him alone to determine whether the motion to reopen and allow additional evidence should be granted or not.
The present motion must be denied, and the appellant if he thinks he has a case that can be sustained by the evidence in his control must resort to his remedy under section 4915 of the Revised Statutes. Motion denied.