delivered the opinion of the Court:

This is an appeal from the decision of the Commissioner of Patents in an interference proceeding. Judgment of priority was awarded to appellee, Lester O. Smith, by the Examiner of Interferences, which judgment was reversed by the unanimous decision of the Board of Examiners in Chief, which, in turn, was reversed by the Commissioner. Originally there were three parties to this interference. One Hine, who, like appellee, was an employee of the Coe Brass Manufacturing Company, was the junior applicant, but, since these appeals were taken, this company, the owner of the Hine application, has purchased the application of appellee, and the Hine appeals have been withdrawn.

The issue is embodied in the following counts:

“1. The combination with a rotary wire-drawing block having a drawing face on its periphery, of a vise pivoted directly to the block, and arranged to turn on its pivot, so that the holding faces of the jaws will take a radial position with reference to the block, and swing to carry the wire onto the drawing face, a die, a die holder pivoted at a point more remote from the block than the die, to allow the die holder to move from a line with the opening of the die directed substantially towards the axial line of the block to a line tangential to the periphery of the block.

“2. The combination in a wire-drawing machine of a rotary block, having a drawing face on its periphery, a vise pivoted on *286the block, and located at one side of tbe drawing face, and arranged to have tbe gripping jaws swing to tbe radial line o£ tbe block, a die and die bolder pivoted at a point to swing from a line, with tbe axis of tbe die directed toward tbe axis of tbe block and tbe vise, to a position with tbe axis of tbe die directed tangential to tbe block and the drawing face.

“3. Tbe combination in a wire-drawing machine of a rotary block, having a drawing face on its periphery, a vise pivoted directly to tbe block, and located at one side of tbe drawing face, and arranged to bave tbe gripping jaws swing to tbe radial line of and project beyond tbe face of tbe block and swing to carry the wire onto tbe periphery of the block, a die and a die holder pivoted to swing from a point with tbe die contiguous to, and in line with, tbe jaws when in radial position, to a point with tbe axial line of tbe die directed tangentially on tbe drawing face of tbe block.

“4. In a wire-drawing-machine, the combination of a rotary drawing block, having a vise pivoted directly thereon to swing to a radial position, with reference to tbe block, and swing to carry tbe wire onto tbe periphery of tbe block, a die and a die bolder arranged to permit a movement of tbe die from a position contiguous to, and in line with tbe vise, when in radial position, to a line directed tangentially on tbe periphery of tbe block.

“5. Tbe combination in a wire-drawing machine of a rotary drawing block, a vise pivoted directly on tbe block between tbe circumference and tbe axis of tbe block, and adapted to swing to carry tbe wire onto tbe periphery of the block, and a swinging die holdei’, all three arranged so that tbe axis of tbe block, tbe pivot of tbe vise, and tbe opening of tbe die may be brought into tbe same line.”

Kegarding the nature of the invention in issue, tbe Board of Examiners said: “In tbe old practice of wire drawing, after tbe end of tbe wire rod was pointed by swaging, it was inserted in tbe die, and, on an ancillary machine, sufficient length of wire drawn through the die to extend from tbe die bolder of tbe main machine to tbe drawing block, as tbe drum or reel which *287draws the wire from the die and on which it is wound is called. A sufficient length of wire had to be provided in the reduced end also for extending into the bight of the vise or tongs, as the gripping mechanism on the drum or block is called. After a suitable length of wire was drawn through the die for this purpose, the latter, with the reduced end of the wire extending through it, was then transferred to the die holder of the main machine, the projecting end of the wire inserted in the jaws of the block, and the latter rotated to draw and reel the wire. The preliminary drawing operation above referred to took time, and also required the end of the wire so drawn to be removed and discarded or scraped by reason of an irregularity formed thereon, and to obviate these objections a block known as the Morgan block had been devised and introduced into use, in which the vise for receiving the end of the wire is pivoted to the block so that it.can swing from a position in which it extends beyond the periphery to a position lying within the circumference of the block, the die holder also being pivoted so that it may swing the die into a position in close proximity to the vise when the latter is swung to its outer position. By thus modifying the old block, but a short length of wire is necessary to extend through the die and be gripped by the vise of the block, and it is therefore possible to dispense with the preliminary step of drawing the length of wire through the die, referred to above. The construction just described was in common use long prior to the date when the invention which constitutes the issue of this interference was made hy any of the parties. * * * Now it is well known that, if a speed which is usually practicable for the major part of the drawing operation be given to the wire initially, the shock of the first tug of the block before the metal of the wire begins to be displaced by the die will frequently cause it to snap. Various expedients have been adopted to prevent this mishap, such as a spiral on the block, leading the wire from a point towards the center of the block gradually to the periphery. This feature, though not comprised in the present invention, is present in the structure disclosed by Smith and Hiñe. The invention of the issue is another means of *288meeting the same difficulty. For this purpose the pivotal points of the die-holding arm, the axle of the block, and the vise on the block are so arranged with reference to the length of the arm and vise respectively that these elements may be brought into substantial alignment in a line connecting the axis of the block and the pivot of the die holder, the die and the jaws of the vise in this position registering. When the drawing is started with the parts in this initial position, the wire is carried during a portion of the first revolution from its radial direction with respect to the block, to the periphery, so that the initial pull is very slow, but is accelerated rapidly until it equals the peripheral speed of the block. By this method of operation no snap is given to the wire when the machine starts, so that a much higher speed can be given to the block as a whole in starting without corresponding risk of snapping the wire. The essence of the invention, in fact the only invention involved, is therefore so proportioning and arranging the parts, themselves old in the art, that the machine starts with the die holder and vise in a position radial to the axis of the block.”

Appellee’s application was filed on July 6, 1903, while appellant’s application was not filed until March 11, 1904. The record clearly shows that appellant, Hugh L. Thompson, conceived the invention in issue in November or December, 1901, and reduced to practice in an operative device in December, 1903. Prior to appellant’s date of conception, appellee made drawings of a wire block, which block (described in the record as the No. Y block) was installed at the Coe Brass Works and operated on August 24, 1901. It is contended by counsel for appellant that this block did not embody the invention in issue, and upon this point the case must turn.

Appellee testified that block No. Y was started with the vise and die holder in radial position as called for by the issue, and that it was constructed with that end in view. It is well settled that the uncorroborated testimony of an inventor is not sufficient to establish either a conception or a reduction to practice. Although the drawings from which the No. Y block was made fail to disclose the feature of the radial start, appellee at*289tempts to point out that such an operation must necessarily follow from the construction shown. Speaking of the drawings, the Board of Examiners in Chief said: “These blue prints, however, show nothing more than that the drawing of the wire must have begun with the wire in a line secant to the drawing face of the drum, and, as it proceeded, the wire swung outward to the periphery of the block. From these prints it cannot be determined whether the machine was adapted to swing into a radial position or not.”

None of the witnesses who testified on behalf of appellee corroborate him as to the radial start. One Westerman, foreman of the wire department of the Coe Brass Company, upon being asked what were the new features of this block, said: “The new features were the absence of the drawbar. In pulling in wire in a die, the old way is to pull it in about 2 feet by means of a drawbar; then it is put into the vise of the block and started up and drawn through; and the improved block was to do away with the drawbar and start in the wire direct on the block.” Iline, one of the parties to the interference in the tribunals below (the issue between him and appellee being one of originality), on being asked if the various elements were so arranged as to permit of a radial start, answered, “I presume they were;” but, when pressed for an answer from personal knowledge, he said: “Why, this is a technical question that I wouldn’t be able to answer positively.” However, when asked whether any of the machines constructed before the filing of his application in June, 1905, had the radial start, his answer was an unqualified “Yes.” He. could not state, however, how long prior to his application these machines had been constructed.

The Commissioner found corroboration of appellee’s statement as to the radial start in the testimony of one Blakeslee, a witness called in rebuttal on behalf of appellant, that the shorter the point of the wire was made, the more the die holder and vise would have to be swung toward a radial position in order to get a sufficient grip upon it. The evidence shows that the points were made from 4 to 8 inches long. The Commissioner also found corroboration in the testimony of one Northrop, also *290called in rebuttal by appellant, that “the wire would be placed in the vise in a radial position, but the wire would not start through the die until the vise had left the radial position.” There is positive evidence, however, that it was impossible to start the drawing operation with the wire directed radially on the block. This the Commissioner admits, saying: “While it may be true that the actual drawing- or reduction of the wire did not begin until after the parts had left the radial position, it does not appear that the issue contemplates an actual reduction of the wire while the parts are still in the radial position, or that this operation takes place in the Thompson machine.” With his conclusion, we cannot agree. The invention in issue differs from the prior art only in providing a means for preventing the initial shock to the wire when the drawing operation is begun. This purpose is not subserved if the wire has left the radial position before it starts through the die, even though it may be possible to place the wire in the vise in such position.

From a careful examination of the record, we are forced to the conclusion that the radial start was not in appellee’s mind when he constructed the No. 7 block, and was not communicated to anyone else until after appellant entered the field. It is contended by counsel for appellee that it was proved by certain drawings made in 190Y, after all the evidence in this case was submitted, that the die holder and vise could be brought into the radial position. These drawings were properly rejected by the tribunals below as improper evidence to be considered in the case. The drawings, if admissible, would not have availed much, in the absence of evidence that this position was ever utilized to secure a radial start of the drawing operation. The evidence is conclusive that it could not be so adjusted. But, .conceding for the moment that appellee’s die holder and vise were capable of being moved into the radial position, it was accidentally discovered after appellant’s invention had become known; and appellee cannot be held, as against the rights of appellant, to have contributed anything to the art. Mere accidental practice does not constitute reduction to practice. Gayler v. Wilder, 10 How. 477, 13 L. ed. 504. In Tilghman v. *291Proctor, 102 U. S. 707, 26 L. ed. 279, Mr. Justice Bradley, delivering the opinion of the court, said: “The accidental effects produced in Daniell’s water barometer and in Walther’s process for purifying fats and oils preparatory to soap-making, are of the same character. They revealed no process for the manufacture of fat acids. If the acids were accidentally and unwittingly produced, whilst the operators were in pursuit of other and different results, without exciting attention, and without its even being known what was done or how it had been done, it would be absurd to say that this was an anticipation of Tilghman’s discovery.” Our conclusion is that, with regard to the radial start, the invention here in issue, appellee made no such contribution to the art as to constitute an anticipation of appellant’s invention, which is conceded by all parties to meet perfectly every element of the issue.

The decision of the Commissioner of Patents is reversed, and the clerk is directed to certify these proceedings as by law required. Reversed.

A motion by the appellee for a rehearing was overruled May 4, 1909.