American Graphophone Co. v. Emerson Phonograph Co.

MAYER, District Judge. This patent has been the subject-matter of considerable litigation, and was adjudged valid in American Graphophone Co. v. Universal Talking Machine Mfg. Co., and Same v. American Record Co., 151 Fed. 595, 81 C. C. A. 139 (January 14, 1907). and American Graphophone Co. v. Leeds & Catlin Co. et al., 170 Fed. 327, 95 C. C. A. 511 (April 30, 1909). The invention, as stated by Jones in his specification, ‘"i elates to the commercial production of sound records, and has for its object the production of a number of copies of an original record characterized by later:'! undulations of substantially uniform depth.” And he claimed: “1. The herein described method of producing sound records, which consists in cutting or engraving upon a tablet of suitable material, by means of the lateral vibrations of a suitable stylus, a record groove of appreciable and practically uniform depth and having lateral undulations corresponding to the sound waves, next coating the same with a conducting material, then forming a matrix thereon by electrolysis, and finally separating this matrix and pressing the same into a tablet of suitable material, substantially as described. ”2. The process of producing commercial sound records of the type Indicated, which consists of first preparing a flat tablet or disk of soft wax-like material, then engraving thereon by means of the material vibrations of a suitable stylus a record groove of appreciable and uniform depth and having lateral undulations corresponding to sound waves, next rendering the surface thereof electrically conductive, then forming a matrix thereon by electrolysis, next separating the matrix from Ihe original record disk without the use of heat, and finally impressing said matrix into a disk of suitable material to form the ultimate record, substantially as described.” [1] The value of the opinions supra consists, not only in the fact that they state the conclusions of the court as to the questions then presented, but also that they make clear what it was which the court-then considered to be a differentiation fropi and an advance beyond the prior art sufficient to characterize an invention. It is an interesting and entirely human characteristic of patent litigation that, as time goes on, the owner of the patent seeks to extend the scope of its claims as far as possible, while those who seek its benefits are constantly contending for a construction which shall narrow that scope. Thus it is, where a patent has once been declared valid and a controversy later arises in a suit inter alios, that so much argument is presented as to what the court in a prior litigation really decided. Where, as in this case, the court wrote many years before, — 576 — it is helpful to remember (although not controlling) that the question of invention was looked upon with eyes which saw the art as it then seemed. Errors, of course, may occur, especially in abstruse arts; but, speaking generally, the view entertained by the courts in the earlier days of the life of a patent is usually a safer guide with'which to judge invention’ and scope of claims than new and later contentions, which, as the case may be, seek to enlarge or defeat the inventor’s accomplishment. [2] In the case at bar (except for the Wurth and Johnson defenses infra), the record on the question of invention seems to be substantially the same as that which received such careful consideration in the reported opinions supra. The claims are, of course, for a combination, and plaintiff contends that the great value of the Jones process lay in the matrix, and that the act of cutting the original record groove is only one of five steps in the complete Jones process. The opinion of Judge Townsend, however, in 151 Fed. 595, 81 C. C. A. 139, supra, demonstrates that the inventive feature rests in the lateral cutting step, which was then regarded as new in the art. The result was the production in the sound record of lateral undulatory grooves of even depth, corresponding to sound waves, and that result achieved a notable commercial success. Several extracts from Judge Townsend’s opinion could be quoted to support the construction of his opinion here stated, but it will suffice to extract the following observation made in relation to the Young British patent, No. 1,487, which was then considered the closest reference in the prior art: “And we conclude, in the light of the prior art, that the changes from Young to Jones involved invention, because, inter alia, Jones was practical, Young was impractical: Young was before the public for six years before any ‘skilled artisan’ succeeded ‘in adjusting the various elements so that a flat sound record of the type in question could be produced,’ and no one prior to Jones saw that it could be adapted to a practical disk record with lateral undulations; there were inherent objections to the practical production of varying depth records, which Jones found did not exist when the known or suggested processes were applied to laterally undulating grooves of even depth.” Indeed, it was in respect of what the District Court described as a “step forward,” but within limit of a man skilled in the art (i. e., “the step by which the groove is cut or engraved by the lateral movement of the stylus, instead of undulations being traced or etched”), that the Circuit Court of Appeals differed from the District Court, and reversed the decrees by which the District Court had declared the patent invalid. The file wrapper, whose history need not be recited, fully confirms the view as to the “step forward,” and of several expressions in the specification itself none is more convincing than the statement of Jones: “For the foregoing reasons I do not, claim my new process in connection with sound records characterized by vertical 'irregularities, but limit it to records characterized by lateral undulations of practically uniform depth.” — 577 — When Jones filed his application, the art was familiar with (1) the Bell & Tainter, or Edison, type, known as “graphophone” records, and (2) the Berliner type, known as “gramophone” records, both of which have survived in the art. The former, known as “hill and dale,” are characterized by vertical undulations; the latter, known as “zigzag,” by lateral undulations. It was to the latter that Jones devoted his attention, and what he accomplished and all he accomplished was to so cut his groove as to obtain his characteristic lateral undulations of practically uniform depth. In this case, defendant has presented two defenses, not heretofore directly before our Circuit Court of Appeals — the Wurth use and the Johnson defense. The former may be passed by as merely experimental, and, in any event, of no consequence in this litigation. The Johnson defense has two aspects. It is claimed (1) that Johnson was prior to Jones, and that Johnson’s prior use fully covered the Jones invention; and (2) that a decree in a suit between the Victor Company, as plaintiff, and the American Graphophone Company, as defendant (the plaintiff here), and the Johnson patent, are admissible in evidence. The decree referred to was the result of the opinion of Judge Ray in 1911, in Victor Talking Machine Co. v. American Graphophone Co. (C. C.) 189 Fed. 359. That decree was affirmed by consent of the parties by the Circuit Court of Appeals, 190 Fed. 1023, 111 C. C. A. 675. The Johnson patent, which was the subject-matter of the decree, was No. 896,059, granted to Eldridge R. Johnson August 11. 1908, which was found to have been a divisional application of patent No. 778,975. The original application of Johnson was filed on August 16, 1898, and the divisional on November 12, 1904. That the decree and Johnson patents supra are not admissible is too plain for argument; hut the testimony as to the Johnson use is clearly admissible. Judge Ray, in his opinion, drew a very close line of distinction between the Jones patent and the Johnson patent, holding the view that they were both valid. The case on this point was so debatable that there well might be doubt as to the result in the Circuit Court of Appeals, if that controversy were reviewed by that court on the merits. The question here, however, is whether the testimony of Johnson, and those who corroborate him in various respects, establish under familiar principles that prior use which will defeat the Jones patent. Taking the Johnson defense testimony at face value, it is apparent that Johnson’s work went no further than ideas, and the formative or experimental stage, until after the Jones application date. No record produced by Johnson was put on the market until 1900. In September, 1896, Johnson, with one Haddon, visited one Du Bois, a friend of Haddon. Johnson testified: t-I explained, to Mr. Du Bois substantially what I was seeking to accomplish. I asked him if wax surfaces could be electroplated with' copper, and if such electroplate would be accurate reproductions of the surface of the wax. I told him. they must be very accurate. Mr. Du Bois assured me that this could he done. I thou gave Mr. Haddon a fragment of one of the wax — 578 — tablets upon which I had made a record. Mr. Haddon, after further consultation with Mr. Du Bois, brought to my office in a few days the piece of wax record which X had given him, having sound waves cut with laterally undulating lines of even depth on its surface. He had succeeded in making a perfect deposit of copper On the fragment of record. I took the record from him and carefully separated it, the wax from the copper, without injuring either one. I immediately made a careful examination of the same copper deposit with the strongest magnifying glass in my possession at that time. This observation and examination convinced me that the copper reproduction was very accurate, and I knew from that time on that my process could be used commercially, and that I could manufacture disc records with laterally undulating sound waves of even depth, of a superior quality to anything theretofore known in the art.” This fragment of irregular shape cannot be called a matrix for the purposes of this case. After 1896 Johnson did nothing until December, 1897. It was not until then that he “succeeded in finding an expert” (one Nafey) whom he thought “competent to carry out” his plans. Nafey started to work in January, 1898, and the first copper matrix was made about April, 1898. Duplicate records were not manufactured until about the same date. Even then, Johnson said, “the records were not shown indiscriminately. A number of people saw them, and I reproduced them for a number of people, but it was always confidentially,” and, indeed, he made every effort to keep his work secret until 1900. Without further reference to the testimony of Johnson and his associates and friends, it must be concluded that the so-called Johnson use did not anticipate Jones. One fact, however, is established by Johnson’s testimony which disposes of any emphasis as to electroplating, and that is as to the general use of the Berliner process for reproducing records by the electroplating process; such records having been electroplated and duplicated by the Duranoid Manufacturing Company, of Newark, N. J., as far back as 1896. [3] Defendants offered in. evidence the Clark and Johnson patent, No. 624,625, granted May 9, 1899, but applied for prior to the date of the Jones application. This is not prior art in this case. Autosales Gum & Choc. Co. v. Ryede (D. C.) 138 C. C. A. 648, 222 Fed. 956, affirmed 223 Fed. 1021, and cases cited. And the motion to strike out .the testimony in respect thereof must be granted. With the record in this situation, the patent in suit must be held valid, and this patent must be excluded, and the case thus comes down to the question of infringement. Emerson, for many years, was in plaintiff’s employ, and plaintiff urges this fact as having some bearing on the question of infringement. So far as this record discloses, Emerson was not guilty of any wrongful or improper acts. He availed of existing knowledge open to all the world, and endeavored to devise a process different from that of the patent in suit. This he was fully entitled to do. There are, of course, cases' when the conduct of employés may affect their standing ,in a court of equity; but this is not one of them. What Emerson did, no one else had done in all the years during which this beneficial art was progressing. He adopted what might be called the — 579 — midway between the vertical and lateral systems and tilts the cutting tool at an angle of 45°. What Emerson sought commercially was the production of a record which could be played sufficiently well for commercial purposes upon all types of phonograph machines. The artistic result might not be (and, indeed, is not) equal to that attained by the Berliner or zig-zag; but Emerson evidently appreciated the well-known fact that the majority of the buying public is well enough satisfied with pleasing music, and does not concern itself so much with that accurate and fine reproduction so necessary to the comfort of those gifted with discriminating cars and trained in artistic appreciation. His appeal, therefore, was to owners of all types of machines, and I am satisfied that in seeking this field he has made an honest effort not to entrench on the Jones patent. On first impression, it may seem that the Emerson angle is a mere evasion; but the testimony of the experts and the demonstration of actual playing of the records in the courtroom soon make clear that the question of infringement is both real and difficult. Preliminarily, it is important to ascertain whether the Jones patent was generic., in the sense that it covered broadly any process other than the hill and dale, up and down, or Edison (as variously called), or must he confined to the precise terms of its claims with a reasonable range of equivalents. The court arid file wrapper history concur in showing that, while the Jones patent was a valuable advance, it was in no sense generic. The District Court, on two occasions (in the Universal and Leeds & Catlin Cases, supra), had held the patent invalid, and the result, as hereinabove indicated, was that the appellate court found the inventive feature, in effect, to be that which had to do with lateral undulations. The outstanding features of the claims cannot now be modified or lightly laid aside. They are (1) a record groove of appreciable and practically uniform depth, and (2) having lateral undulations corresponding to sound waves. The word “practically,” does not appear in claim 1, and is inserted merely by way of safeguard in claim 2, In effect, in this case, “uniform” and “practically uniform” mean the same thing. Further, they are words of limitation, and not of description. The question, then, is whether the Emerson record groove is of uniform depth and has lateral undulations corresponding to sound waves, and whether the Emerson has vertical undulations which are idle, or which perform no useful function. Ocularly the grooves of plaintiff and defendants appear different. There are variations o f depth iri defendants’ groove and vertical undulations in the bottom thereof. Defendants contend that these differences are caused by the different method and angle of cutting, and result in different aural impressions between a pure Victor or Columbia zig-zag record played on plaintiff’s machine and an Emerson record played on the same machine, while, in addition, the Emerson can be well played on an Edison machine. Plaintiff insists that the vertical undulations in the Emerson record arc inert and do not perform any useful function. — 580 — The following diagrammatic drawings in evidence of models in evidence show the grooves and the cutting tool in the three instances and will save much verbal description: — 581 — The Jones undulations are latex-al — i. e., confined structurally and' operably to the side walls — and do not extend to the bottom of the groove, and are necessarily of uniform depth, because, as plaintiff’s expert stated, “what we mean ordinarily by a laterally vibrating tool is one which cuts a groove of substantially uniform depth; the two thoughts go together.” Such is not the case with the Emerson groove. The Jones lateral undulations are in both sides of the groove alike and undulate in absolute parallelism. In the Emerson groove, according to the testimony of Prof. Morris, a highly skilled microscopist, on behalf of plaintiff, “there is a correspondence” between either side with the other side, or either side with the bottom, but “there is no exact parallelism.” It is but fair to add that Prof. Morris characterized the lines as coswerving; “that is to say, all swung toward thq inner part of the disc or all toward the outer part of the disc together ; they were not parallel but they were according.” With these important physical differences (1) in position of cptting tool, and (2) in appearance of the grooves under the microscope, came the battle of the experts. It is impracticable to go into the many contentions in detail. Both experts, Mr. Wadsworth and Mr. Dyer, have lived with the art, and are frank as well as able. Their testimony represents their sincere convictions, and I have no doubt that a court of scientists would be quite as much at variance as they are. I am disposed to conclude, however, that the Emerson groove oscillates vertically to the same extent as it oscillates laterally, and that the Emerson tool vibrates, not in a straight vertical line, nor in a straight horizontal line, but in a single straight oblique line, incidentally lowered and raised while swinging from side to side. When, however, it stops at any point of the oblique line of its travel, it stops completely, and does not swerve to the side. The vertical, or “hill and dale,” undulations of Emerson are therefore real, and do perform an active and useful function. Whether this conclusion is correct or not, it is at least apparent that the burden of proving infringement has not been sustained by plaintiff, and tire situation is, as matter of law, very much like that discussed in General Electric v. Sundh, 251 Fed. at page 286, - C. C. A. -. I have given little consideration to the experiment (the good faith of which is not questioned) of the “buffed” matrix and the comparison with the “unbuffed” matrix. Such experiments occasionally may< be reliable, but, generally speaking, they are full of uncertainties. The moment there is departure from the precise device, the controversy is diverted. To add to or subtract from a device — especially where we are dealing with infinitesimal variations — leads surely to doubt and confusion. Finally, it is an important, outstanding, and emphatic fact in this case that, whether well or badly done, the Emerson record may be played on either a hill and dale or zigzag type of phonograph — a result which was never before commercially attempted. The bill is dismissed, with costs.