Opinion
John M. Cole, of New York City (Joshua R. Potts and Basil H. Brune, both of Chicago, Ill., of counsel), for plaintiff.
Fred E. Tasker, of New York City (E. W. Marshall, of New York City, of counsel), for defendant.
PATTERSON, District Judge.
The suit is for infringement of patent. The patent said to have been infringed is patent to Arnold, 1,390,267, applied for November 30, 1920, and issued September 13, 1921. The patent is on a machine for making tufts or balls and applying them to fringes. The product, ball fringe, is used for decorative purposes on various fabrics. The defenses are invalidity and noninfringement.
The Arnold machine is one whereby a warp of threads is tightly bound by wire, a loop of fringe being bound in by the wire at the same time; the warp is cut above the binding, so that a tuft fastened to the fringe is formed. The tuft, when steamed, will take the shape of a ball. The operation of the machine in making and attaching the tufts may be briefly described. The machine has a head which has vertical movement and carries dies and cutting edges actuated in timed succession by cams. At the start a loop of the fringe is placed by hand over a prong of a clinching die. The prong is pointed toward the operator and has a concave side to facilitate placing the loop. With the machine in operation a short length of wire is fed forward and is cut by a knife attached to a bending die; the piece of wire is then bent into a V-shape by a mandrel pin and the bending die. A movement of a plunging die and a clinching die toward one another causes the wire to encircle the warp and loop simultaneously and to grip them securely. The head drops and pulls down the warp for the next cycle. The warp is then cut a short distance above the wire binding to make a tuft. The final step is the ejection of the tuft. The machine seems to have had commercial success. Prior to its appearance, balls were generally made by hand and attached to the fringe by hand, a slower and more expensive process.
There is a question of invention. Arnold was not the first to build a machine for making tufts and attaching them to fringe. Nor was he the first to use wire for binding the tuft, though the Arnold specification might suggest otherwise. The patents to Swoboda, issued in 1891 and 1893 (German 63,337, United States 508,645, and German 76,792), describe a machine for making and attaching tufts, and the tufts were bound by wire. The Swoboda machine is quite different in structure from Arnold's, and evidently had no commercial career. Tufts continued to be made and attached by hand. Swoboda does show, however, a die action for clamping the wire around the warp that is basically similar to that of Arnold, and Swoboda attached the tuft to the fringe in the same manner as Arnold, that is, by slipping a loop by hand over a curved prong of the clincher die. The place of Rehfuss in the prior art is of importance. In 1901 Rehfuss built a machine for making tufts and sold it to Maurer & Sons, a Philadelphia house that marketed ball fringe. Arnold was employed by the Maurer concern from 1900 to 1920. The Rehfuss machine was never patented. It was used to a limited extent by the Maurer Company for making tufts, but was not capable of attaching them to fringe. In parts and in method of operation the Arnold machine is a close copy of the Rehfuss machine. Despite Arnold's denial, there is not the least doubt that he made a close study of the Rehfuss machine and took most of the elements of his machine from it. The tufts are formed in the same way. But Arnold went further than Rehfuss. He tapered down one side of the Rehfuss clincher die, so that there would be a curved prong accessible to the operator on which a loop of the fringe might readily be placed by hand. He also changed the location of the knives to advantage. On the Rehfuss machine the knives were set below the dies. Arnold put them above the dies. By these changes the machine was made a tuft-attaching as well as a tuft-making device. Arnold's prong is quite similar to that of Swoboda. It appears then that Arnold took the greater part of his machine from Rehfuss and took the prong of the clincher die from Swoboda. While the parts were old, the combination was new, and in my opinion there was patentable invention in what Arnold achieved. Webster Loom Co. v. Higgins, 105 U.S. 580, 26 L.Ed. 1177; Hobbs v. Beach, 180 U.S. 383, 21 S.Ct. 409, 45 L.Ed. 586. If the case were one of doubt, the commercial success of the machine would be persuasive of invention. Smith v. Goodyear Dental Vulcanite Co., 93 U.S. 486, 23 L.Ed. 952.
But the question remains as to prior public use of the Arnold machine for more than two years preceding his application for patent. The act allows a patent to an inventor on an invention 'not in public use or on sale in this country for more than two years prior to his application' for patent. 35 U.S.C.A. § 31. In the leading case, Smith & Griggs Mfg. Co. v. Sprague, 123 U.S. 249, 8 S.Ct. 122, 31 L.Ed. 141, it was laid down that prior use by an inventor in an experimental way does not result in loss of the invention, even though incidentally the product of the use is sold, but that a prior use for trade or profit for more than two years before application is fatal. In the latter event, it does not aid the patent that experiment is carried on incidentally. It is also law, by the Smith & Griggs Case, that where a prior use for more than two years has once been proved, the patentee then has the burden of showing, by 'full, unequivocal and convincing' proof, that the use was primarily by way of experiment. Later authorities to the same effect are collected, and the rule restated by Judge Learned Hand in Aerovox Corporation v. Polymet Mfg. Corp., 67 F. (2d) 860 (C.C.A.2).
Arnold's application was filed on November 30, 1920, shortly after he quit the employ of the Maurer Company. The critical date is therefore November 30, 1918. Arnold's first machine was built in early 1917. As early as May 12, 1917, this machine was in operation in the Maurer plant. It was recognized as a success from the start, and turned out merchantable ball fringe. The product was sold to the trade in course of business. That machine was practically the same as the machine described in the patent. Mrs. Thomas, the first operator, could see no differences between the first machine operated by her and an Arnold machine in court conceded to be substantially the machine as patented. The proof is that the first machine had the curved prong substantially as in the machine later patented. Two more machines like the first were straightway built, so that by the fall of 1917 three were in operation. The operation of the three machines was fairly continuous, except for time out for repairs. Beyond doubt they were being run for profit fully three years before Arnold applied for his patent. If the use had been primarily for experiment, the building of these additional machines was a singular thing; one machine would have sufficed for testing. By November 30, 1918, a fourth machine had been added and was in operation.
Arnold testified, to be sure, that the use was for the purpose of testing, that till 1919 or 1920 changes were being made in wire feed, prong, guards, and posts, that the earliest date of satisfactory work was just before he filed application. But the issue is one of fact, and the weight of evidence is the other way, that the new machines worked well from the start, that they were in commercial operation all along, that the changes made were slight and did not affect the principle of the machine, that while some loops were missed by the first machines, the missing was due to inexperience of the operators, and there is a certain amount of missing on to-day's machines. Certainly from the fall of 1917 on, commercial use was the principal use, experiment no more than the incident.
The fact is undisputed that three Arnold machines were in use much longer than two years before Arnold filed application. That being the case, it was for the plaintiff to show by full, unequivocal, and convincing evidence that the use was experimental. The plaintiff's evidence on the point does not measure up to that standard. The defendants' evidence that the prior use was primarily for profit is more potent. The use was not secret; the machines were operated in the Maurer factory as a regular part of the Maurer business, with Arnold's consent, and under his direction. See Worley v. Loker Tobacco Co., 104 U.S. 340, 26 L.Ed. 821; Aerovox Corporation v. Polymet Mfg. Corp., supra. The finding is that there was public use of the invention for more than two years before Arnold filed his application, and on that finding the inevitable conclusion is that the Arnold patent is invalid.
The proof of infringement on the part of both defendants is strong enough to warrant a finding for the plaintiff on that issue. But in view of the invalidity of the patent, there will be a decree dismissing the bill, with costs to the defendants.