Gilbert v. General Motors Corporation, 133 F.2d 997 (2d Cir. 1943)

*998 Hugh J. O'Brien, of Rochester, N. Y. (Werner, Harris & Tew, and Frank Keiper, both of Rochester, N. Y., and John D. Meyer, of Pittsburgh, Pa., of counsel), for appellant.

Drury W. Cooper and Alan C. Bakewell, both of New York City, and Frederic R. Twelvetrees, of Buffalo, N. Y., for appellee.

Before AUGUSTUS N. HAND, CHASE, and FRANK, Circuit Judges.

CHASE, Circuit Judge.

This suit was brought in the District Court for the Western District of New York to charge the defendant as trustee ex maleficio for the plaintiff of an unpatented invention submitted by the plaintiff to the defendant in confidence for inspection and fraudulently appropriated by it. Federal jurisdiction is based on diversity. The district court found after trial that the defendant did not appropriate any invention the plaintiff submitted to it and dismissed the bill. This appeal is from the decree then entered.

Since the dismissal was on the ground that the plaintiff had failed to prove that the defendant wrongfully appropriated his invention, our first inquiry may well be directed toward whether or not the finding of the court to that effect was based on substantial evidence.

The plaintiff insists that the finding was erroneous (1) because in interference proceedings in the Patent Office admissions binding upon the defendant were made to the effect that the plaintiff's invention was incorporated in a device made and used by the defendant; (2) that by failing to answer under oath requests for admissions which the plaintiff served upon it in accordance with Rule 36, Federal Rules of Civil Procedure, the defendant has admitted as much; and (3) that the evidence introduced at this trial does prove that the plaintiff did submit to the defendant in confidence and that the defendant did wrongfully appropriate the invention as the plaintiff has alleged.

The plaintiff alleged and attempted to prove that he was the first inventor of an *999 automatic switch used in the engine starter on automobiles sold by defendant; that before he applied for a patent upon it as improved, he submitted it to the defendant to enable the latter to determine whether it wanted to make some arrangement for its use; and that the defendant, falsely representing to the plaintiff that the switch was nothing it wanted, fraudulently used the invention so submitted in confidence to make a switch it did use. That switch will be called herein the Buick switch.

The particular kind of switch with which we are concerned is one that turns the current off and on the electric motor which runs the starter on an automobile engine. It will do well enough for present purposes to begin with one on which the plaintiff was granted Patent No. 1,635,078 on July 5, 1927, though that was by no means the first automatic switch of similar kind. He was a mechanic in a garage at Seneca Falls, N. Y., who undertook to design an automatic starter switch which would turn off the current after the engine started; turn it on again if the engine stalled; turn it off when the engine started after stalling and go through that cycle indefinitely so long as the ignition circuit was closed.

The specifications in the plaintiff's patent disclose that he accomplished his purposes by using the suction in the intake manifold of the internal combustion engine in the automobile to open the starter motor's circuit and by using gravity to keep the circuit normally closed; i.e. closed whenever the automobile engine was not in operation. He used a piston inside an upright cylinder which was grounded through the walls of the cylinder and had an electrical contact at the bottom. When gravity held the piston at its lowest point this contact it carried cooperated with one in the cylinder, and suitably insulated from it, to close the circuit from the battery to the starter's motor. Gravity thus held the contacts together and while that was the force which determined the position of the piston in the cylinder the circuit to the starter's motor would remain closed. He led his current from the battery to this switch in a circuit that was closed at the same time the ignition circuit was closed when the driver turned the key, or whatever was to be moved, to do that. In that way the patented switch became part of what is sometimes called a key starter since no other manual operation was needed, in addition to whatever was necessary to turn on the ignition, to turn current on the starter motor. But it was desirable for obvious reasons to have this motor stop running when the automobile engine started under its own power. To bring that about automatically the plaintiff connected the top of his cylinder to the intake manifold so that the vacuum pull present there when the engine started would take effect in the cylinder and would lift the piston to separate the contacts and break the circuit to the starter motor. He put a port in the cylinder above the piston which let enough air enter there to prevent suction at cranking speed from raising the piston to open the circuit, though not enough to keep it from so doing when the engine started.

So long as the engine was running under its own power the piston might in theory remain high enough in the cylinder to prevent the closing of the circuit to the starter motor and that would be so in fact so long as the vacuum pull was greater than the force of gravity. The vacuum pull was not enough in practice, however, for the suction so obtained from the intake manifold of an internal combustion engine varies inversely as the throttle is opened when the engine is working under enough load to keep its speed from increasing much, if any. And so the piston might drop in the cylinder to close the starter motor circuit while the engine was running under a heavy load with the throttle well, or wide, open because the suction pull became less than the gravity pull. To do away with that undesirable possibility, the plaintiff altered his patented switch by making the immovable contact at the bottom of the cylinder a moveable one. He put this moveable contact on a swinging arm on the outside of the cylinder and connected the arm to the throttle. After the engine started, the contact would be pulled away from the cylinder, as the throttle was opened, so far from the contact on the piston that, so long as the accelerator pedal was depressed, the closing of the starter motor's circuit was prevented no matter what might be the position of the piston in the cylinder. He also put in a spring to hold the piston in closed contact position when the engine was not running. This construction will be called the plaintiff's second switch.

What will be called his third switch was a key start one like his second with a piston and cylinder performing the same functions in the same way. It did not, however, have a moveable contact which could be pulled away and held away from the cylinder to keep the switch open no matter where the *1000 piston was in the cylinder. The only moveable contact was, as in his patented switch, that carried on the piston. He made his connection from the accelerator pedal to the piston this time. After the piston had been moved by the suction from the manifold to pull the contacts apart, the piston was held from moving back into closing position, so long as the accelerator pedal was depressed, no matter how the vacuum might drop in the cylinder. It will be noticed that the plaintiff kept the key start feature of his patented switch in both his second and third switches but that in neither of the latter could there be any automatic restarting of a stalled engine while the accelerator pedal was depressed. There would be automatic restarting as soon as that pedal ceased to be depressed or at least there would then be an automatic closing of the circuit to the starter's motor.

What the plaintiff had contrived was made known to the defendant in the following ways. The trial judge found on ample evidence that on September 10, 1931, a man named Watkins, who was acting for the plaintiff, drove an automobile equipped with his third switch to Rochester, N. Y., where he demonstrated it to Findlay, one of defendant's engineers in charge of new devices. Findlay was then employed by Delco Products, a subsidiary of the defendant then making appliances for household uses only. Both Watkins and Findlay thought the switch being demonstrated was the one the plaintiff had previously patented and Findlay, who died some years before this suit was brought, neither took the switch apart nor made a sketch of it when he examined it; nor was there any evidence that he ever told any of the defendant's other employees what he learned about it.

The plaintiff also attempted to show that he later submitted his third switch for examination to some engineer working for the defendant at its Chevrolet plant in Detroit but the proof of that was too meagre to support any finding to that effect and none was made.

There was a submission of the third switch as the result of a letter one Frantz, who was interested in it with the plaintiff, wrote on December 26, 1931, to the Delco-Remy Corporation, a subsidiary of the defendant at Anderson, Ind. The chief engineer at Anderson replied expressing a desire to see the switch. By then an application for a patent on it had been filed and been given Serial No. 568,572.^[1] On December 31, 1931, Frantz wrote Delco-Remy enclosing a photostat of application Serial No. 568,572 and also sent a switch which was received and examined in due course and Dyer, one of defendant's engineers, whose work will now become important learned all about the switch early in 1932.

In 1930, however, before the plaintiff's switch had been submitted to the defendant at all, its engineers had begun to develop a switch for an automatic starter control of the key-start kind. They made use of the vacuum from the intake manifold working on a diaphragm to supply the initial automatic movement needed to permit a spring to open the circuit to the starter motor and to hold one contact in such a position that the circuit would remain open regardless of variations in the vacuum. Blake and Hill had submitted such devices to the defendant in December, 1930, and they had been tested early in 1931. At least as early as April or May in 1931 Dyer began to work on the development of this type of automatic control. On September 10, 1931, the same day Watkins demonstrated the plaintiff's switch to Findlay at Rochester, N. Y., Dyer at Anderson, Ind., made a drawing of a starter control switch which was not of the keystart sort but required a turning-on of the starter current after the ignition circuit had been closed. He broke the starter motor's circuit by using as the automatic feature the generator's output voltage after the engine started. That apparently was not entirely satisfactory. Dyer kept at it and by September 29, 1931, had the problem worked out enough to make a drawing of a control embodying the important features which have since been built into what has been called the Buick switch. In doing his work Dyer had the benefit of some other devices which had been submitted to the defendant and which should be mentioned briefly. One was by Collins whose switch was responsive to the movement of the accelerator pedal and to vacuum made effective by means of a tube and piston much as it was in the plaintiff's switch. But Collins, unlike the plaintiff, did not put his piston into the electric circuit and so could have it fit more loosely in the tube. A difference presently more important, however, was that Collins did not open the starter motor's circuit by direct vacuum pull as did the *1001 plaintiff but had the vacuum act to release a spring that opened the switch. No variation in the vacuum could bring about a reversal of the action of the spring, however, and so the circuit would remain open regardless of vacuum. A man by the name of Kauffman also submitted several devices which Dyer had available for use before he perfected the Buick switch and before he became acquainted with the plaintiff's switch. One had a connection to the accelerator pedal, similar to the plaintiff's to prevent the closing of the starter motor's circuit after it had been opened by suction should there be such a vacuum drop that the suction alone would not keep the switch open. Another had means for disabling the connection which had opened the switch so that a reversal of that movement could not close it. That was the type of construction used by Collins and also by Dyer in making the Buick switch. Of course the foregoing does not include all of the evidence but it will suffice when the Buick switch has been described to show that there was adequate evidential support for the finding of the trial judge that, "The Buick switch has nothing in common with that of Gilbert except use of vacuum control and a mechanical connection to the accelerator. These features are entirely different in construction and function in the Buick device and that of Gilbert. These features are both found in the prior devices of Kauffman and Collins."

The Buick switch is not of the key-start type; i.e. the circuit to the starter motor is not closed merely by turning on the ignition. It does employ the intake manifold vacuum to start automatically the operations that open the circuit to the starter's motor. Yet it controls this method not as the plaintiff did but by a disabling means which so changes the mechanical connection that no reversal of the movement caused by vacuum pull can, upon any vacuum drop, have any effect beyond the point where the mechanical connection has been made unable to pass it on and consequently cannot have any effect upon the circuit controlled by the switch.

Dyer accomplished this by applying his knowledge and experience to the perfection of the switches of Collins and Kauffman in such a way that a fairly simple, compact and apparently efficient one way automatic switch emerged. He connected a diaphragm, which was made responsive to the vacuum in the intake manifold, with a pin to one member of what will be called a clutch. He connected the other clutch member to a spring which if let alone would keep it turned out of engagement with the part of the clutch connected to the diaphragm. When it was so turned out of engagement, the switch in the starter motor's circuit would be open and when that clutch member was turned into engagement with the cooperating member of the clutch that was connected to the diaphragm the switch was closed. He connected the accelerator pedal to the spring held member of the clutch so that depressing that pedal after it had been allowed to rise freely would make the clutch members engage. This turning movement would not only "throw in" the clutch to hold its two parts together but would at the same time close the starter motor's circuit. That circuit would remain closed until the engine started. With the starting of the engine the vacuum would pull out the part of the clutch responsive to it. That would "throw out" the clutch and the spring, no longer held inactive by the clutch, would turn the part responsive to it away from engaging position and that would open the switch. No matter how the part of the clutch connected to the diaphragm might thereafter move as the diaphragm responded to the variations in suction caused by the variations in the vacuum pull from the intake manifold, the spring held part of the clutch would not move. And so the circuit to the starter's motor remained open until the clutch was thrown in again by the force needed for that as before described.

This construction did not permit the automatic closing of the circuit to the starter's motor after it had been opened by spring action released by suction on the diaphragm. There would always have to be some action by the driver of the automobile before the starter would run again. In practice this was simple enough as the control was but the up and down movement of the conventional accelerator pedal held up by a spring and depressed by the driver's foot. The pedal, if depressed, would merely have to be released and then depressed again each time the engine was started.

As the evidence thus shows, the plaintiff put the old vacuum method of automatic control into use in his switch by preventing the reversal of a connection which vacuum drop might otherwise cause to reverse itself and close the circuit. Dyer used a different connection, which had to be reset *1002 each time after use before it would work again, to put the vacuum to work as an automatic control and by so doing got his result in a different way. And he did so before he knew what the plaintiff had done.

Although the foregoing would ordinarily require the affirmance of the decree the plaintiff insists that in the interference proceedings Dyer, whose application was then owned by the defendant, admitted that his invention was identical with that of the plaintiff. For the purposes of those proceedings the adoption by Dyer of a claim identical with that of the plaintiff was an admission that their conceptions were the same. Ewing v. United States ex rel. Fowler Car Co., 244 U.S. 1, 37 S. Ct. 494, 61 L. Ed. 955. But in those proceedings the sole issue was priority of invention. United States ex rel. Lowry v. Allen, 203 U.S. 476, 27 S. Ct. 141, 51 L. Ed. 281; Ewing v. United States ex rel. Fowler Car Co., supra. Neither Dyer nor the plaintiff prevailed and no authority has been called to our attention, and we have found none, to the effect that either was thereby estopped from later asserting as against the other such rights as might be his on the actual facts under the law. This is what happened: The plaintiff filed his application for a patent on what has been called his third switch on Oct. 13, 1931, and it was given application serial No. 568,572. Over a year later Dyer filed, on Nov. 19, 1932, an application for a patent on what has been called the Buick switch. In 1934 an interference was declared. It involved Dyer's application; one by McKinney and Eagle; and one by Lachappelle. While those proceedings were pending undetermined and in March 1935, the Commissioner redeclared the interference and added the plaintiff's application serial No. 568,572 as well as applications by Blake, Collins, and Saxe.

As is customary, the Commissioner suggested a claim which all adopted. Under the rules of the Patent Office each had to elect to do so or concede that he was not entitled to priority as to any invention covered by the claim. Neither Dyer nor the plaintiff was successful in the interference proceedings. Lachappelle was awarded priority over all. The plaintiff now contends that those proceedings were irregular in that the defendant owned all the applications in interference and failed to disclose that fact. However that may be, it is obviously of no moment on this appeal but wholly a collateral issue. The plaintiff undertook to pursue its remedy in respect to the award in the interference proceedings by bringing a suit under § 4915 R.S., 35 U.S.C.A. § 63, in the District Court for the District of Columbia to obtain the patent despite the award of priority to Lachappelle. It was dismissed, though not on the merits, and the dismissal was affirmed on appeal. Gilbert v. Lachappelle, 75 U.S. App.D.C. 395, 127 F.2d 750.

All this left both the plaintiff and Dyer simply two of the defeated parties to the interference proceedings. They were not precluded from contesting the question of interference in fact later. Westinghouse v. Hien, 7 Cir., 159 F. 936, 24 L.R.A.,N.S., 948. Were the appellant right in his insistence that the parties to the interference all conclusively admitted that the invention each claimed was the same it would be of no help to him on this appeal. The defendant does not defend this action merely on the strength of the effect of what Dyer did. It also is the assignee of Lachappelle who won in the interference proceedings. If it were the law that each of the contestants in those proceedings were estopped from claiming later that there was interference in fact the plaintiff would now be unable to question the defendant's right as the assignee of Lachappelle to be accorded the position of the first inventor of whatever subject matter was involved in the interference.

The only other contention made by the plaintiff to avoid the force and effect of the actual evidence as reflected by the decision below is that he filed requests for admissions in accordance with Rule 36 Federal Rules of Civil Procedure and that the defendant failed to answer them under oath and so must be held to have admitted as requested. If so, the plaintiff must prevail for the requests for admissions were comprehensive enough to include the material allegations in the amended complaint. These requests for admissions were introduced in evidence. There is nothing in the record to show that they were not answered under oath as Rule 36 requires to prevent the requests being treated as admissions. In this court the defendant has moved in diminution of the record for leave to add thereto answers not under oath which are represented to have been duly made and served upon the plaintiff as well as filed in the office of the clerk of the district court. The point is made that the plaintiff's failure *1003 to object to the answers on the ground that they were not under oath is a waiver of that defect. See City of Detroit v. Grummond, 6 Cir., 121 F. 963; Shell Petroleum Corporation v. Caudle, 5 Cir., 63 F.2d 296; Manke v. United States, 9 Cir., 38 F.2d 624. Whether the defendant may be right or wrong about waiver we are not disposed to grant its motion for the reason that the plaintiff is in no position in this court to rely upon any failure to answer his requests for admissions. He did not prove such failure below and the record shows none. The rule is not self executing and if one would take advantage of its provisions all the facts necessary to invoke the consequences must be made in some way to appear. That answers not sworn to were filed in the office of the clerk below is a fact which we may notice but that does not show that answers under oath were not duly served on the plaintiff as the rule contemplates. We have no reason to believe that they were but in the absence of a record which shows affirmatively that they were not the plaintiff has failed to lay the needed foundation for asserting that the requests were not answered under oath.

Affirmed.

[1] This was later in the interference already mentioned and there denied priority.