It is worthy of notice that hitherto in the progress of this litigation, upon the same testimony, different persons have reached different conclusions. Thus, in the opinion filed June 5, 1888, by the examiner of interferences and assistant examiner, it was found that the defendant was the original inventor. On an appeal from that decision the examiners in chief (two members being present) came to a different conclusion, and awarded priority to the plaintiff. On a further appeal the commissioner of patents on March 22, 1889, reversed the judgment of the examiners in chief, and sustained that of the original examiners. A motion for rehearing was brought before a succeeding commissioner, and overruled. When this case was submitted, without any additional testimony, to the circuit court, the conclusion finally reached in the patent office was dissented from, and it was found that the plaintiff was the original inventor. An examination of the opinions filed by these different officers indicates that by each of them the matter was carefully considered. Evidently, therefore, the question as to which was the prior inventor is not free from doubt. What, then, is the rule which should control the court in the determination of this case? It is insisted by counsel for the appellant that the decision of the patent office should stand, unless the testimony shows beyond any reasonable doubt that the plaintiff was the first inventor, and, in support of their contention, they cite the cases of Coffin v. Ogden, 18 Wall. 120, 124, and Cantrell v. Wallick, 117 U. S. 689, 695, 6 Sup. Ct. 970. In the first of these cases, which was a suit for infringement, the defense was a prior invention, and in respect to this defense the court observed: 'The invention or discovery relied upon as a defense must have been complete, and capable of producing the result sought to be accomplished; and this must be shown by the defendant. The burden of proof rests upon him, and every reasonable doubt should be resolved against him.' In the other case, the same defense in a suit for infringement was set up, and there the court thus stated the rule: 'The burden of proof is upon the defendants to establish this defense, for the grant of letters patent is prima facie evidence that the patentee is the first inventor of the device described in the letters patent, and of its novelty. Smith v. Vulcanite Co., 93 U. S. 486; Lehnbeuter v. Holthaus, 105 U. S. 94. Not only is the burden of proof to make good this defense upon the party setting it up, but it has been held that 'every reasonable doubt should be resolved against him."

These two cases are closely in point. The plaintiff in this case, like the defendant in those cases, is challenging the priority awarded by the patent office, and should, we think, be held to as strict proof. In the opinion of the court below, the rule is stated in these words: 'The complainant, on the issue here tendered, assumes the burden of proof, and must, I think, as the evidence stands, maintain by a clear and undoubted preponderance of proof that he is the sole author of that drawing.' 42 Fed. 451. This language is not quite so strong as that just quoted. The case as presented to the circuit court was not that of a mere appeal from a decision of the patent office, nor subject to the rule which controls a chancellor in examining a report of a master, or an appellate court in reviewing findings of fact made by the trial court. There is always a presumption in favor of that which has once been decided, and that presumption is often relied upon to justify an appellate court in sustaining the decision below. Thus, in Crawford v. Neal, 144 U. S. 585, 596, 12 Sup. Ct. 759, it was said: 'The cause was referred to a master to take testimony therein, 'and to report to this court his findings of fact and his conclusions of law thereon.' This he did, and the court, after a review of the evidence, concurred in his findings and conclusions. Clearly, then, they are to be taken as presumptively correct, and unless some obvious error has intervened in the application of the law, or some serious or important mistake has been made in the consideration of the evidence, the decree should be permitted to stand.' See also Camden v. Stuart, 144 U. S. 104, 12 Sup. Ct. 585, and Furrer v. Ferris, 145 U. S. 132, 12 Sup. Ct. 821.

But this is something more than a mere appeal. It is an application to the court to set aside the action of one of the executive departments of the government. The one charged with the administration of the patent system had finished its investigations and made its determination with respect to the question of priority of invention. That determination gave to the defendant the exclusive rights of a patentee. A new proceedings is instituted in the courts,—a proceeding to set aside the conclusions reached by the administrative department, and to give to the plaintiff the rights there awarded to the defendant. It is something in the nature of a suit to set aside a judgment, and, as such, is not to be sustained by a mere preponderance of evidence. Butler v. Shaw, 21 Fed. 321, 327. It is a controversy between two individuals over a question of fact which has once been settled by a special tribunal, intrusted with full power in the premises. As such it might be well argued, were it not for the terms of this statute, that the decision of the patent office was a finality upon every matter of fact. In Johnson v. Towsley, 13 Wall. 72, 86, a case involving a contest between two claimants for land patented by the United States to one of them, it was said: 'It is fully conceded that when those officers (the local land officers) decide controverted questions of fact, in the obsence of fraud, or impositions, or mistake, their decision on those questions is final, except as they may be reversed on appeal in that department.'

Upon principle and authority, therefore, it must be laid down as a rule that, where the question decided in the patent office is one between contesting parties as to priority of invention, the decision there made must be accepted as controlling upon that question of fact in any subsequent suit between the same parties, unless the contrary is established by testimony which in character and amount carries thorough conviction. Tested by that rule, the solution of this controversy is not difficult. Indeed, the variety of opinion expressed by the different officers who have examined this testimony is persuasive that the question of priority is doubtful, and, if doubtful, the decision of the patent office must control.

What was the invention which the parties each claim to have made, and in respect to which an interference was ordered in the patent office? It was thus stated by the examiner: 'In a device for coiling wire or wire rods, the combination with a rotating coiling receptacle or reel for receiving and laying in coils the rod as it is delivered from the rolls of a spider or platform for supporting the coil mounted on a vertical shaft concentric with the reel-supporting shaft, and means for elevating the platform shaft independently of the other.'

Plaintiff claims to have conceived the idea of this combination in July, 1878, and to have made in that and the succeeding year sketches and drawings which fully disclosed it. It is conceded that a machine embodying the invention was first constructed and put into successful operation in the spring of 1886, and that this was done under the direction and superintendence of defendant.

During the years 1878 and 1879, the defendant, a man about 47 years of age, was, as he had been since 1864, the general superintendent of the Washburn & Moen Manufacturing Company, a corporation engaged in the manufacture of iron and steel wire; while the plaintiff, 24 years of age, was in the employ of the same company as a draughtsman, working under the direction of the defendant. The business of the company had, during the years of defendant's superintendency, grown to be enormous. In the actual work of the mill, as the finished wire came from the rolls it was coiled on reels. At first this was done through the agency of an attendant seizing the wire by a pair of tongs and engaging it with the reel, but this operation was attended with both danger and delay. To obviate this, the parties interested in this manufacture sought the invention of machinery which should seize the finished wire and engage it with the reel, and thereafter dislodge the completed coil therefrom.

The defendant locates the time of his conception of the idea embodied in this combination in October, 1878, and that which enables him to locate it is a transaction with Daniel C. Stover. It appears that in October, 1878, Stover (who was a manufacturer and inventor) came to Worcester, Mass., and, while engaged in examining the machinery in the mill, had his attention directed to the way in which the wire was coiled on the reels, and thought that some device could be invented for picking up the wire immediately after its leaving the rolls. After some reflection, he conceived the idea which he subsequently embodied in patent No. 219, 124. He suggested his idea to the defendant, who replied that it was not new, and that others were working at it. He prepared a model which contained, not only a device for picking up the wire, but also one for discharging the completed coil, and exhibited it to defendant, who, on his part, showed Stover a model which he had prepared. It is not pretended that either the model of Stover or that of defendant disclosed the exact combination for which a patent was claimed in this case. Nor is it important to notice all the details of the transactions at the time between Stover and the Washburn & Moen Company. It appears that Stover sold and assigned a one-half interest in his invention to the Washburn & Moen Company. The time of Stover's visit is established by the date of that assignment and the application which he made for a patent, and by other writings. There is significance in the fact that, although the plaintiff was present when Stover's model was shown, it does not appear that he made any suggestion that he had invented anything of a similar character; and yet, if his present claim is true, he had for months been considering the matter, and had at least three months before conceived the very idea of the combination now in dispute. But it is enough to say in respect to this branch of the case that the story of the invention by defendant at the time stated is reasonable, probable, and to a certain extent supported by the testimony of Stover.

As against this, the plaintiff claims to have conceived the idea of this combination in July, 1878, and relies mainly on the testimony of two witnesses, Lambert and Fowler. Though he was a draughtsman, he presented no sketch or drawing made prior to November, 1878, which in any manner pictures his invention. It is true he testifies to having made sketches prior thereto, but none have been preserved. One of them he claims to have shown to Lambert in July, 1878, and Lambert was called as a witness to support this statement; but Lambert's testimony does not, it seems to us, carry the weight which is claimed for it. He was a tinsmith by occupation, employed at times by the manufacturing company, and testifies that in July, 1878, or about that time, very soon after Daniels' return from Europe, he came to his shop and showed him a sketch, and asked him to make a model of it. He declined, saying that he was too busy. Nothing more took place at that time, but in the fall of 1886, at Daniels' request, he made a model of the machine which, as he says, was disclosed by that sketch. This model was in evidence. Now, it is possible that one seeing for a few minutes a sketch of a complicated machine can eight years there after remember the details of that sketch so clearly as to make an accurate model; but, if it is possible, it surely is not probable. If the invention disclosed in the sketch impressed either Daniels or Lambert as something of great value, and therefore fixed itself firmly in the mind, it is strange that neither seemed anxious to impress it upon the attention of others, carry in into actual use, or derive profit from it. On the other hand, if it was one which did not impress either as of any special value, it seems almost morally certain that the details of the sketch and the precise character of the invention would not have been accurately remembered through all those eight years. We must not be understood as imputing to Lambert intentional falsehood. He was familiar with the machinery actually in use in the wire mill. He saw, in the course of his acquaintance with that machinery, many models and many machines, and the machine embodying this invention, as perfected, had been in actual use in the mill several months before he made this model. He may well have gotten some of these matters confused in his mind, and introduced into this model elements and parts which were not seen by him in the sketch displayed in 1878, and which were in fact taken from other sketches and drawings, or from models or machines. At any rate, when there was nothing to specially arrest the attention, it taxes credulity for one to claim that he bears in mind for over eight years the details of a sketch of a complicated machine which is casually shown to him, and which he sees but for a short time, and is enabled to thereafter reproduce the details of that sketch in a model.

Equally unreasonable is the testimony of Fowler, which is to the effect that on July 20, 1878 (the date being fixed by a memorandum in his diary), Daniels came to visit him. This diary notes the fact simply of the visit of Daniels, but contains nothing in respect to the matters involved in this case. His testimony as to this was in these words: 'He explained by rough sketches an arrangement he had for coiling the wire after it was delivered from the rolls, which was a round box mounted on an upright shaft, and in the bottom of the box was a plate, perhaps I should say a movable plate, which could be lifted for the purpose of raising the coil of wire. That was his method of getting hold of the coil, to get it out of the box.' None of these rough sketches were preserved. There was apparently nothing to impress the matter upon the mind of Fowler, and his attention was not directed to it until some time in the early part of 1887, nearly nine years thereafter, when Daniels called on him for the purpose of securing his testimony in this case. Doubtless Fowler means to be truthful. There is no reason to impugn his integrity, but his testimony is subject to a criticism similar to that placed upon the testimony of Lambert. As one of the matters which was being considered and discussed by the manufacturing company was how to do the work which is now being done by the machine finally constructed by defendant, it is not at all unlikely that Daniels, one of the employes of that company, spoke of the matter to Fowler at the time named; very likely he may have drawn rough sketches to suggest the ideas which were in his mind; but it is not probable that, in the absence of some special reason therefor, the memory would carry for eight or nine years the details of the plan or idea suggested by Daniels and illustrated by these sketches. While, of course, it is possible, yet such testimony is not of a character to carry great weight.

There is other testimony on both sides of this controversy. It is unnecessary to notice it in detail. It is enough to say that the testimony as a whole is not of a character or sufficient to produce a clear conviction that the patent office made a mistake in awarding priority of invention to the defendant; and because of that fact, and because of the rule that controls suits of this kind in the courts, we reverse the judgment, and remand the case, with instructions to dismiss the bill.

Mr. Justice JACKSON did not hear the argument or take part in the decision of this case.