Appellant brought suit upon three patents to Harold A. Clark, Nos. 1,807,567, claims 1 to 10; 1,807,569, claims 1 and 2; 1,819,871, claims 3 and 4. The District Court found that the first-mentioned patent is invalid for lack of invention; that one Roed was the first inventor; and that the patent if valid is infringed. The court found the last two mentioned patents invalid for lack of invention, but, if valid, infringed by certain of defendant’s devices. From a decree dismissing its bill of complaint for want of equity, appellant appeals.
These patents relate to oil seals which are used within bearings or housings of rotating parts, chiefly in automobiles and other machines employing axles and shafts, to prevent oil from escaping. Oil readily seeps through the most minute openings, and devices have long been used to furnish a seal in such places as the rear axle of an automobile which revolves within a housing upon bearings completely surrounded by oil.
The first patent covers a self-contained, unitary oil seal structure, having a cup-shaped outer member, which is closed by a washer, and having within the inclosure of the cup a sheet leather packing member comprising a barrel portion, which embraces the shaft, and a flat radial flange portion which is placed flatly against the bottom wall of the cup and hugs closely the axle or shaft. The outer cup, in turn, contains a pressure transmitting shell in the form of an inner cup located between the washer and the leather flange. The bottom of this inner cup presses closely against the flange and permanently presses the flange against the bottom wall of the outer cup. Encircling the leather flange and between it and the washer is an endless coil spring which constricts the leather. The device efficiently prevents leakage of oil. The structure is clearly shown in the description and specifications as well as the claims, of which No. 7 it will be sufficient to quote:
“In a seal, a cup-shaped cage, a cup-shaped ring nested within the cage, a flanged packing having axially and radially extending portions, said packing being positioned in the cage with the radially extending portion of the packing clamped between the bottom of the cage and the bottom of the ring, a coil spring positioned under tension within the ring about the axially extending portion of the packing, and a washer positioned under pressure *418within the cage against the rim of the ring in loose engagement with one side of the spring.”
That the device has been successful is shown by the fact that it was promptly adopted by the automobile industry and superseded every prior device for like sealing. Prior to the trial, appellant had produced and sold over 44,000,000 seals, and 80 per cent, of all automobiles manufactured in the last few years have been equipped with them.
The narrow questions submitted are whether Clark was the first inventor and whether the invention is anticipated by the prior art.
Appellee relies upon two devices not patented, namely, those of Timken and Hupp, which, for the purpose of this discussion, are equivalent. The Timken device made use' of a cup-shaped casing, but the cup in reality had no bottom against which the flange of the leather packing bore. Rather, it made use of a narrow annular inset shoulder against which the outer marginal portion of the flange of the leather packing was clamped by a washer set within the cup. The washer had a flange which extended axially substantially to the edge of the cup. The flange of the washer and the cup formed a metal-to-metal contact which was not efficient as an oil-tight seal because of distortion of the metal due to peening and press-fitting operations. The result was oil leakage through the imperfect joint. While it is true the structure clamps the extreme outer edge of the flange, the pressure was applied some distance radially from the shaft and the clamping effect was not sufficient to prevent leakage. Furthermore, this device could be utilized in only a relatively large space, as the inset shoulder had to be formed in' the outer cup radially beyond the portion of the outer cup which. contained the spring. Consequently, the diameter to which the outer cup could be reduced for any specified shaft was necessarily comparatively limited. In other words, the device could not be made small enough to fit in the small clearance spaces for seals in the modern automobile.
The appellee relies upon Aschenbach patent No. 1,507,567. That did not cover a self-contained device like that of appellant. It was necessary that its parts be built up or assembled when it was put in place. The clamping pressure on the leather packing was not inherent, therefore, in the structure itself, but such pressure could be produced only by the wheel on the axle bearing against the outer metal shell of the device. Consequently, if wheel clearance developed between the shell and the wheel by wear, after short use, any pressure effect on the leather was lost and the shell became free to assume a position where it did not even come in touch with the leather. There was, therefore, no degree of permanency in the sealing effect.
Appellee rebuilt one of Aschenbach’s structures, but in doing so, it lost Aschenbach’s idea, destroyed the original structure, and rebuilt according to the teachings of Clark. Aschenbach did not teach the essential use of an inner shell as a device for clamping the leather within a self-contained structure.
We find nothing in the Timken structure, which has been abandoned, or in the Aschenbach patent, which was never in use, by which Clark was taught how to achieve what he has accomplished in his patent.
Clark had previously made a similar device under patent No. 1,674,762, which met with success, more than a million of them having been made. But this, too, had its defects. It was small and lessened the amount of leakage found in prior seals. It was defective in that the spring was required to perform the dual function of constricting the barrel of the packing member around the shaft and pressing the flange of the packing member against the bottom of the cup. In order to achieve its dual function, the spring had to fit the space between the washer and the packing flange, but in practice and in use it was found that this pressure lessened; because the coils of the spring became embedded in or cut into the flange of the packing member and thus permitted oil' to leak between the bottom of the cup and the flange of the packing member.
In the- second attempt, that is, the first patent in suit, Clark made a revolutionary change that perfected the device. It is insisted by appellee that there was no invention involved because what Clark did was merely to put a simple spacer into an otherwise old combination. If this is an accurate statement of all that Clark accomplished, it would follow' that he did not achieve invention over his' first patent in the improved device of the patent here in suit. Electric Cable Joint Co. v. Brooklyn Edison Co., 292 U. S. 69, 54 S. Ct. 586, 78 *419L. Ed. 1131. Clark inserted an inner member, which counsel term a shell, located between the washer and the flange of the packing member, and we believe that his action in this respect involved more than a mere addition of a member to an old structure. Rather we believe this addition became a part of a successful new organization and brought about a solution of the problem of oil leakage. His structure is self-contained and unitary in character. It includes the outer cup-shaped sheet metal cage member within which is a flexible leather packing, which has a horizontally extending flange which is placed flatly against the bottom of the outer cup and clings closely to the shaft. The inner shefl constitutes a pressure transmitting and distributing ring. It, too, is of cup shape formation, having a cylindrical portion and a radial portion, which bears with a predetermined and fixed pressure against the flange of the leather over such an area as to bring about substantial and effective pressure. The sheet metal washer, located within the rim of the outer cup, bears against the end of the cylindrical portion of the inner shell and is interlocked with the rim. The coil spring within the cup encircles the axially extending portion of the leather and thus brings the latter against the shaft. This spring is independent of the washer and the inner shell.
Thus we see that this inner shell seal is a completely self-contained unit in which the parts are' all properly assembled and related one with the other with correct clearances and pressures at the time when the article is manufactured. In assembling the car, the workmen press-fit the seal as a unit into place, and it thereafter functions as intended without any further attention, adjustment, or alteration. The sheet leather packing is properly held permanently and unvaryingly after installation of the seal. We find, therefore, that the inner-shell takes the place of the spring as a pressure member for holding the leather against the bottom of the cup and provides a flat pressure area against the leather as distinguished from the interrupted line contact obtained with the spring in the first device, thereby eliminating the gradual loss of pressure which takes place when the spring becomes more embedded in the softening leather, as happened in the first Clark device. This fixed pressure remains throughout the life of the device, achieving an oil-tight joint between the packing flange and the bottom wall surface of the cup. Instead of having a thick spring which would fill the entire space, the new device makes use of a small one, without sacrifice of efficiency.
That this result achieved by Clark is not an obvious one would seem clear from the fact that none of the prior art suggested it. Aschenbach, chiefly relied upon by appellee in this respect, in our opinion, did not include such teachings. The adoption by the trade and the commercial success of the device are persuasive in this connection. We believe, therefore, that claims 3, 4, 5, 6, 7, and 8 are valid.
Claims 1 and 2 have to do with an optional suggestion, namely, that by making the inner shell of smaller diameter than the casing or cup, the leather packing and the inner shell can have a slight radial or floating movement to adapt the leather packing to any slight eccentricity in the shaft with respect to the housing. Claims 9 and 10 have to do with another optional suggestion, namely, that by placing small spurs in the bottom of the inner cup, which imbed in the packing, and by forming a notch in the rim of the inner cup, which receives a reciprocal tongue in the outer portion of the washer, an effective interlock is provided which will make it impossible for the leather to rotate in the cup.
We believe that neither of these features added anything to the invention of Clark; that his achievement as disclosed in claims 3 to 8, inclusive, was complete in itself; that the optional features are merely suggestions that would occur to the mechanic skilled in the art; and that no additional inventive features can be attributed to Clark because thereof. Accordingly, we hold that 1, 2, 9, and 10 are within the doctrine announced in Featherstone v. Cycle Co. (C. C.) 53 F. 110, 113, 116; Packard v. Lacing-Stud Co. (C. C. A.) 70 F. 66, 68; and Cons. Car Heating Co. v. American Elec. H. Corp. (C. C.) 82 F. 993, and void. The additional features are not inventive in character and can add nothing to the invention disclosed in claims 3 to 8, inclusive.
It is insisted by appellee, however, and the District Court found, that Roed preceded Clark. The latter’s filing date was June 25, 1928, and appellant claims that appellee’s evidence is not sufficient to overcome the filing date; that Roed did not complete the invention by reduction to *420practice; that his history is that of an abandoned experiment; and that the proof disclosed that Clark’s actual date of invention was earlier than 'Roed.
Roed made oil seals for Clark while employed by the Chicago Tool & Die Company in the spring of 1927 similar to Clark’s first device in hearing in suit. He testified that he studied at home upon the problem because he found that Clark’s first device would leak. He claims to have made a discovery as a result of this study, which he produced in court as a,completed device, which, he says, he showed to one Lyngaas in the summer of 1927 and to a cousin in July or August. Lyngaas testified that he saw the device on or about the 28th or 29th of May, 1927. The cousin claimed to have seen the seal in the latter part of July or August, but it was then assembled and he did not see the interior. Peterson testified that he saw the device in the summer of 1927 and Aschersen that he saw it in June, 1927. Roed was in plaintiff’s plant quite often, while working for his employer, and had free access to it. He had two devices and there was opportunity for the witnesses to confuse the two. One of them did not touch the question before the court. The witnesses who testified for Roed were not familiar with oil seals, and Roed claimed that his ideas had their inception from what he learned while working on Clark’s device. Roed never applied for a patent, never showed the. device to any representative of his employer, never put' the device into practice, and never tested it out on an axle where the hoúsing was filled with oil. His only test was with his fingers pressing on the side. There was no test on a shaft; no ascertainment of whether it would leak or whether in practice it would perform. We think this was not reduction to practice within the meaning of the decision in Jobski & Griswold v. Johnson, 47 App. D. C. 230, where the court said: “Bryant failed utterly to show by sufficient competent evidence that he made the device described in the counts or reduced it to practice before Johnson did. The only test made by him of the device, which is a part of the means used in retaining an automobile tire on its rim, was, according to his own testimony, the bouncing upon the floor of a wheel having the tire and the device attached thereto. When it is considered what strain a device of this character would be subjected to while the machine of which it was a part swept along at a high rate of speed, the test described by Bryant must be regarded as negligible. What constitutes a reduction to practice 'is a matter the determination of which must largely depend upon the nature or purpose of the machine or device in the particular case, as well as the particular circumstances surrounding the alleged construction or use.’ (Mason v. Hepburn [C. D. 1898, 510, 84 O. G. 147], 13 App. D. C. 86-89.) In Henderson v. Gilpin [C. D. 1913, 310, 187 O. G. 231], 39 App. D. C. 428, this court said: ‘It is not enough, as contended by • appellant, that these shop tests indicated the operation of the device would be successful. To constitute reduction to practice the test must amount to a demonstration in fact as contradistinguished from one in theory.’ ”
Further, moreover, it appears that the witness Agger saw Clark’s drawings of his device on June 11, 1927; that he saw the actual device which was preserved and put in evidence three or four weeks before that date. If the inspection was four weeks before, it was not later than May 14th; if it was three weeks earlier, it was not later than May 21st. Either date is prior to any suggested as to any disclosure by Roed. Other witnesses saw the device in the spring of 1927 and in the early part of June, 1927. It was submitted to the Ford Motor Company in June, 1927. Samples for testing were made in the fall of 1927 and were tested and approved for production.
We conclude, therefore, that Roed did not make an invention; that he did not' reduce his device to practice and that Clark’s device was first.
The second patent sued upon, No. 1,807,569, pertains to the expedient of tapering one end portion of the seal to ease its entrance into the axle housing. The court found there was no invention in such tapering and that the same was an old and well known expedient. Each of the two claims include only three elements, the sheet metal cup, the annular packing member or leather and a closure member. After prolonged prosecution of the application, the claims were allowed upon representation that “the constriction or tapering” between the inner and outer cups serves to space such cup inwardly from *421the inner wall, thereby effectively preventing the drawing action of the housing on the cylindrical periphery of the cage from disturbing the connection. We agree with the District Court that no invention was involved in the expedient thus adopted.
The disclosure of claims 3 and 4 of the third patent in suit, No. 1,819,871, differs from that of the other two patents in that no spacer or inner shell is utilized, but two cup-shaped cage members are telescoped in inverted relation to each other. We agree with the court below that invention was not involved in these claims.
The District Court found that appellant’s Exhibit 1-C infringed claims 3 to 8, inclusive, if valid. No appeal from that finding is before the court, and from the record, we find that Exhibits 1-A, 1-C, and 1-D infringe claims 3 to 8, inclusive, of patent No. 1,807,567, and that Exhibit 1-B infringes claim 3 of patent No. 1,807,567.
Accordingly,, it is necessary that the decree be reversed as to the validity of claims 3 to 8, inclusive, of the first mentioned patent. In all other respects the decree is affirmed.