The appeal and cross appeal at bar concern patent rights and trade secrets respecting a "Snap Action Device" among whose uses are a switch for an automobile directional signal flasher, a thermostat, a circuit breaker and an overload protector. In the suit at bar, however, we are concerned only with the device's use as a switch for an automobile directional signal flasher. We shall discuss it primarily in this connection. The device is disclosed by Reissue Patent 24,023.

I. THE ISSUE OF VALIDITY OF REISSUE PATENT 24023.

In 1952, Schmidinger, the individual plaintiff, was issued United States Reissue Patent 24,023 for the "Snap Action Device."1 Prior to 1952 Schmidinger had granted the exclusive manufacturing rights to his device to Tung-Sol Electric, Inc., the corporate plaintiff, subject to a nonexclusive license to Signal-Stat Corporation. During the period of development the defendant James Welsh had been employed by Tung-Sol as a supervisory engineer in charge of the manufacture of Schmidinger signal flashers. In 1952 Welsh left Tung-Sol to go with its chief competitor, Signal-Stat Corporation, where he was employed until the summer of 1959. In the fall of 1959, Welsh and his wife, Marie, also a defendant, formed the defendant corporation, Welflash Inc., for the purpose of manufacturing and selling automobile directional signal flasher devices including the alleged infringing device.2

Schmidinger and Tung-Sol assert that the Welflash flasher constitutes an infringement of claims of the 023 patent and that the defendants are guilty of unfair competition in that Welsh misappropriated the plaintiffs' trade secrets for the mass production of flashers. The defendants assert that the 023 patent is invalid, deny infringement and unfair competition. Jurisdiction is based on Section 1338 and 1400(b), 28 U.S.C.

The court below held five claims (hereinafter set out) of the 023 patent valid and infringed by the Welflash device but that there had been no misappropriation of trade secrets. See 243 F. Supp. 852.3

Schmidinger and Tung-Sol have appealed at our No. 15830 from the ruling of the court below that there was no misappropriation of trade secrets. The defendants have appealed at our No. 15831 from the holding that the 023 patent is valid and infringed by the Welflash device.

The patent specification describes a snap action device that may be actuated by electrical, thermal, or mechanical forces, or by a combination of them. The device comprises three elements; (1) a disc, (2) a laterally flexible pull wire or strip secured at its ends to the disc at the opposite ends of a diameter, and (3) a support element secured to the pull strip, preferably at the center of said pull strip. The disc is of spring material and has a crease along its diameter underlying the laterally flexible pull strip. The pull strip may be continuous between the points of its attachment to the disc and is mounted on a support under longitudinal tension. The support is to be so constructed as to permit adjustment or variation of the tension in the strip. The specification goes on to state that the disc under predetermined increase of tension in the pull strip snaps from a position in which the halves of the disc on each side of the crease extend outwardly away from the pull strip at an angle of less than 180° to a second position in which the disc is bowed in the opposite direction about a diameter of 90° to the preformed crease.

As to the operation of the device, the court below stated, id. supra, at 855:

"In operation, and in the initial starting position, tension in the pull ribbon holds the buckling member in its snapped position. In this position the center of the buckling member is bowed out to form a channel positioned 90° to the permanent deformation and of opposite curvature thereto. The 90° channel of opposite curvature causes a gap to form between the buckling member and the pull ribbon. This 90° channel of opposite curvature is characteristic of the snap action of what plaintiffs termed the `hard' buckler of the patented device, which Schmidinger achieved by means of a single flexible pull ribbon arranged on the convex side of the buckling member parallel to the deformation.

"When the buckling member is in its snapped position, a gap is formed between the buckling member and pull ribbon. At such time the electrical contacts are closed. When the operator of the automobile moves the signal lever into signalling position, electric current passes through the buckling member and pull ribbon in a divided flow to generate heat and relax the tension in the ribbon. As a result, the buckling member snaps to close the gap and open the electrical contacts. Thereupon, the buckling member and pull ribbon cool and contract, until the tension in the ribbon is restored to snap the buckling member and open the gap. The electrical contacts again close, and the cycle is automatically repeated as long as the directional signal lever remains in signaling position.

"In each of the two positions assumed by the buckling member in its back-and-forth snap action, there is a so-called `locking' effect, where the buckling member dwells in position. This dwelling in each position is due to the differential expansion and contraction of the pull ribbon and buckling member. The buckling member remains or dwells in each position while the pull ribbon is expanding or contracting, but thereafter suddenly snaps and moves very rapidly through an equilibrium position located between the two `locked' positions. The differential expansion and contraction are caused by the divided circuit flow of current through the buckling member and pull ribbon, and the ambient cooling thereof."

In an attempt to describe the operation of the 023 device as a directional signal flasher in simplest terms, we state that an electric current flows through the flexible pull ribbon to the disc. The pull ribbon when heated by the current, expands and thereby allows the disc to snap to its natural shape which breaks the flow of electricity through the disc to the flasher light. As the metals of the pull ribbon and the disc cool the disc returns to its original creased condition causing the flasher light to go on again and the cycle is repeated.4

The claims in issue are set out in the footnote.5 Claims 12, 17, 18 and 23 are combination claims having the three elements set out above. Claim 26 is described as having the first two elements stated above. The third element, the support, is not included in claim 26. All the claims in issue are directed to a flasher switch for the operation and control of automobile directional signals, the two working parts, the pull ribbon and the disc, being actuated by the passage of electrical current through them as distinguished from other claims of the patent, not in issue, some of which cover a thermostat actuated by ambient temperature. "Ambient temperature" means, of course, the temperature of surrounding air. The sole commercial use of the 023 device has been in the control of automobile directional signals.

The plaintiffs seemingly concede and we conclude, as did the court below, that the three elements of the specification as hereinbefore stated were fully anticipated by the prior art but they contend that the combination of those three elements creates a device that is new, useful and nonobvious, and therefore patentable. We agree for the reasons stated hereinafter.

We desire to make it clear that we are aware that the validity of claims of a combination patent cannot rest on any "essential" element "gist" or "heart" asserted to lie in the invention. Aro Manufacturing Company v. Convertible Top Replacement Company, 365 U.S. 336, 345, 81 S. Ct. 599, 5 L. Ed. 2d 592 (1961). The defendants contend in this court, as they did in the court below, that the patent in suit is invalid because the disclosure was obvious to any individual possessing the ordinary skill in the art. This is their principal defense on the issue of patentability. Concededly patent claims, not the specifications of a patent, measure the invention, Smith v. Snow, 294 U.S. 1, 55 S. Ct. 279, 79 L. Ed. 721 (1935), and while a specification may be referred to to limit a patent claim it cannot be available to expand it. Darsyn Laboratories, Inc. v. Lenox Laboratories, 120 F. Supp. 42 (D.C.N.J.1954), aff'd per curiam, 3 Cir., 217 F.2d 648, certiorari den. 349 U.S. 921, 75 S. Ct. 661, 99 L. Ed. 1253 (1955). But it is fundamental that claims are to be construed in the light of the specifications and both are to be read with a view to ascertaining the invention. Seymour v. Osborne, 11 Wall. 516, 547, 20 L. Ed. 33 (1871).

If we turn to Claim 12 of the patent which we deem to be substantially similar to the other claims in issue save Claim 26, we find that it claims a snap-action device "according to claim 10"; the tension pull ribbon and the disc to be varied by the differential expansion and contraction of the pull ribbon and the disc by passing an electric current through them.

The issue presented, therefore, is whether 023 meets the test of Section 103 of the Patent Act of 1952, 35 U.S.C. The standard was declared by the Supreme Court by Mr. Justice Clark in Graham v. John Deere Co., 383 U.S. 1, 86 S. Ct. 684, 15 L. Ed. 2d 545 (1966).6 Mr. Justice Clark stated, id. at 3, 4, referring to the Patent Act of 1952 in relation to earlier Patent Acts: "[T]he Congress has for the first time expressly added a third statutory dimension to the two requirements of novelty and utility that had been the sole statutory test since the Patent Act of 1793. This is the test of obviousness, i. e., whether `the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains. Patentability shall not be negatived by the manner in which the invention was made.' § 103 of the Patent Act of 1952, 35 U.S.C. § 103 (1964 ed.).7

"The questions, involved in each of the companion cases before us, are what effect the 1952 Act had upon traditional statutory and judicial tests of patentability and what definitive tests are now required. We have concluded that the 1952 Act was intended to codify judicial precedents embracing the principle long ago announced by this Court in Hotchkiss v. Greenwood, 11 How. 248, 13 L. Ed. 683 (1851), and that, while the clear language of § 103 places emphasis on an inquiry into obviousness, the general level of innovation necessary to sustain patentability remains the same." Mr. Justice Clark went on to make clear that the factors by which non-obviousness are to be tested are substantially the same as were those of the decisional law as it existed prior to the time of the 1952 Patent Act. See also United States v. Adams, 383 U.S. 39, 86 S. Ct. 708 filed concurrently with the John Deere Cases. The defendants insist that Schmidinger developed a mere improvement over the prior art and that the device disclosed in his combination patent could have been made by a skillful mechanic and was not effected by the acts of an inventor. This is the main thrust of the attack upon the patent. We come therefore, to an examination of the prior art in part and a study of the circumstances under which Schmidinger devised the switch disclosed and claimed by him in 023.

Schmidinger at the trial described how he developed his 023 snap-action flasher. Until 1954 motor vehicle flashers had operated by magnetism. Tung-Sol had manufactured a magnetic flasher devised by Schmidinger which had wide use in the motor vehicle field. Schmidinger had secured a patent No. 2,177,671 for a "Thermal Switch and the Like", issued on October 31, 1939 on an application filed on April 30, 1937. He also had received a patent No. 2,299,767 for a "Thermally Operated Switch and the Like", issued on October 27, 1942 on an application filed on February 2, 1940. The teachings of the 671 and 767 patents are prior art in this case. Schmidinger testified that in 1943, attempting to improve his non-magnetic flasher, he directed his attention to the device shown in Fig. 11 of 671. He removed the glass bead at the end of the left arm at numeral 52. He then changed the position of the pull wire to the proximity of the vane, at numeral 12. The design thus created was somewhat similar to that shown at Fig. 25 of 767.

Schmidinger then applied the electric current to the center of the pull wire. There was nothing new about this as is shown by the Hanson (Netherlands) Patent No. 6,711 published March 15, 1922 for an "Automatic Circuit Breaker." Schmidinger next bent his buckling member in the center at right angles to the pull ribbon as shown by Fig. 8 of the Klahn Patent No. 1,784,450, issued on December 8, 1930 on an application filed February 25, 1926. Klahn's Fig. 8 discloses such a deformation but what it fails to show is a deformation along the long axis. The defendants insist that a diametrically deformed disc-shaped buckling member created by creasing it through the diameter would be obvious to a man having ordinary skill in the art. But with this we cannot agree. Schmidinger next employed a pull ribbon in lieu of a pull wire. There can be nothing non-obvious in exchanging a pull ribbon for a pull wire but the pull means is described in 023 as both "laterally flexible" and "expansible". Moreover, as disclosed by 023 the pull strip is attached to the disc in such a manner as to bow the disc in the opposite direction about a diameter of 90° to the preformed crease. Schmidinger then experimented with disc-shaped buckling members, as disclosed in the Spencer Patent No. 1,883,252 issued on October 18, 1932 on an application filed August 15, 1931.

It appears that there was a period of experimentation by Schmidinger as to the size and operation of the members of the device evolved by the disclosures of the patent in suit. This period of experimentation was not as extensive as the plaintiffs contend. The court below laid emphasis upon the fact that Schmidinger had begun his work on automobile signals installations in 1931 and progressed through electromagnetic flashers to mechanical ones. But it is necessary to point out that Schmidinger's experiments on mechanical flashers, resulting in the device of 023, were not prolonged and were brought to fruition in not more than two years. The period required by Schmidinger to perfect his device is not too helpful to him here.

The court below stated, 243 F. Supp. at 860: "The interrelationship of the parts, as put together by Schmidinger, and their functioning dependency on each other, has produced a unit capable of performing the reciprocating snap action with its `locking' effect, and of generating the heavy snap forces required for opening and closing the electrical contacts with firm contact pressure, fast make and break of contacts, and uniform cyclic beat. None of the prior art structures contain [sic] the interrelationship of the parts found in the Schmidinger device, and none of them were [sic] capable of performing the operation or producing the result of the structure of the patent in suit."

This statement is undoubtedly supported by the record, but we point out that the court below employed functional terms in describing the operation of the 023 device and that patent claims at least under the circumstances at bar cannot be sustained by functional conclusions. We do not undervalue Schmidinger's experiments; nor do we conclude that he was merely routineering. We note again that the last sentence of Section 103 states: "Patentability shall not be negatived by the manner in which the invention was made." An invention made even accidentally is patentable,8 and a non-obvious result whether it be obtained by inspiration or by infinite pains is equally within the protection of Section 103 and the principle of the John Deere Cases.

We have given consideration to the substantial commercial success of Schmidinger's flasher, well founded in the record, commencing with the sale of the 023 device in 1956 to the Pontiac division of General Motors. The record shows that by 1963 approximately 99% of the flashers employed in the automotive industry were Schmidinger's. While a portion of these large sales may well have been brought about by the decreased electric current presently employed to operate flashers,9 a decrease that Schmidinger's device could and would sustain, this possibly accidental result does not render the Schmidinger disclosures less valid.

Considering the prior art and weighing the whole record on this phase of the case, we conclude that Schmidinger's disclosures were not "obvious * * * to a person having ordinary skill in the art." 35 U.S.C. § 103. We therefore hold all claims in issue valid,10 save Claim 26.

As to Claim 26, no support is claimed. The court below stated, id. supra, 243 F.Supp. at 860, "However, while a support may be necessary for an operable structure, it is not a basic element of the invention, and need not, therefore, be set forth in the claim. It is quite apparent that one skilled in the art would know how to support the device of claim 26 in order to make it operable." This is not enough and the court below was in error in so concluding. A support is necessary to make Schmidinger's device operable and in a combination patent there is no single essential element. Every element in the combination must join in setting out and defining an operable device possessing utility. Aro Manufacturing Company v. Convertible Top Company, supra, 365 U.S. at 344-345, 81 S. Ct. at 603-604. We hold Claim 26 to be invalid.

The defendants contend that during the prosecution of 023 in the Patent Office the patentee agreed to certain limitations of his asserted claims. Having agreed to these limitations in order to obtain allowance of 023, the defendants contend that Schmidinger cannot now claim that which he previously had released to the public domain. Specifically, the defendants argue that Schmidinger, in order to overcome Patent Office objections based on prior art, limited his device to one which has the buckling member solely supported by the pull means, the stationary support attached to the pull means, not to the buckling member, and to a buckling member which is substantially circular. We agree with these contentions.

An understanding of the current scope of Reissue Patent 24,023 requires an analysis of proceedings before the Patent Office started on July 6, 1944. On that date, Schmidinger filed application No. 543,623 for a "Thermal Control Means". During the prosecution of that application, the Patent Office rejected Claims 1, 18, 19, and 20 on prior art claiming Schmidinger's device disclosed no new invention over Schmidinger 767 and 309 and Newton 2,267,164. Schmidinger responded by amending Claims 1, 18, 19, and 20, noting in his "Remarks" that such amendments were directed specifically to attaching the support solely to the pull means in order to overcome the prior art. A reading of the amended claims does indeed disclose that the support is connected only to the pull means. By his oath Schmidinger repeated the importance of this arrangement and argued that it overcame successfully the prior art. Subsequently, Schmidinger added Claims 41-44 to application 623. These new claims described the buckling member as being supported solely by the pull means. 623 was later abandoned in favor of application No. 237,958, filed July 21, 1951, which resulted in Patent No. 2,615,106 issued on October 21, 1952, which in turn was the basis for the Reissue Patent in suit. In the prosecution of 958 Schmidinger continued Claim 41 of 623 as new Claim 10. Claim 41, as noted, provided that the buckling member be supported solely by the pull means. After 958 had been allowed and Schmidinger had experimented with a device which had the support attached to the buckling member, on advice of counsel, he requested a reissue patent which would include claims for a support attached to the buckling member. In a supplemental oath to the reissue application he swore: "[T]he claims of * * * said patent are inoperative to protect adequately [the] invention in that they are limited to attachment of a support to the pull means * * *." In order to achieve a broadening, Schmidinger included Claim 25 in his reissue application, a claim which could be read to include a support on either the buckling member or the pull means. The Patent Office rejected Claim 25 on Norwegian Patent 957, and subsequently rejected as new matter the language in the specifications which implied that the support could be attached to the buckling member. The Patent Office demanded deletion of all like language and stated in a letter of September 8, 1954: "The support was not originally disclosed as being connected directly to the disc and to refer to such a construction * * * is to disclose new matter * * * Claims 25-27 are rejected as containing new material * * *." Schmidinger complied by deleting all of the alleged offensive language. Moreover, Schmidinger had previously amended Claim 25 to remove any implication that the support could be attached to the buckling member. Thus, Schmidinger did not succeed in broadening his claims to include a device in which the support was attached directly to the buckling member. He cannot now claim that such a device infringes his patent.

As noted, the defendants also contend that Schmidinger limited his device to a disc-shaped buckling member. The definition of the word "disk" or "disc" is given in Webster's New International Dictionary of the English language, 2 ed., as "A flat circular plate; as, a disk of metal * * *." The definition is one of common usages and we accept it.

In the original application 623 and its accompanying figures Schmidinger claimed buckling members of varying shapes. The Patent Office responded in a letter of October 28, 1946, stating: "In view of the rejection, below, of the generic claims, election of a single species (i. e. a single form of support and a single form of buckling member) is required * * *." This demand was clearly in reference to the shape of the buckling member. In a responsive letter dated April 24, 1947, Schmidinger's attorneys insisted that the device was not restricted to a circular buckling member, at the same time, however, choosing to elect for prosecution the device as embodied in Figs. 2, 2a and 6, figures showing substantially circular buckling members. The plaintiffs' witness, Norton, explained that these embodiments were probably chosen in order to include figures which depicted a means for adjusting tension in the pull ribbon and not to elect a shape of buckling member. We do not agree. It is clear that the Examiner was demanding an election of a shape of buckling member. This conclusion is supported by the subsequent rejection by the Patent Office of Claims 12 and 15 as not being readable on the elected species. Claim 12 claimed an oval buckling member and Claim 15 a flat strip. In all other respects both claims were identical to claims not rejected as not readable on the elected species because of shape. Compare Claims 1111 and 12. Claims 12 and 15 were then rewritten as, inter alia, Claims 29, 30, 31, 33, and 34, which were described respectively, as "a disc, a disc being circular, a disc being oval shaped, a flat strip, and a dished metal disc." Thus, Schmidinger had not yet abandoned his claim for a strip buckling member. However, in a letter of August 24, 1950, the Patent Office rejected, inter alia, Claims 33 and 34 as not being readable on the elected species, no generic claims having been allowed. The Patent Office allowed Claims 29, 30, and 31 among others. The only differences between allowed Claim 31 and rejected Claim 33 are the description of the shape of the buckling member and the attachment of the pull means at diametrically opposite points in Claim 31. Schmidinger subsequently cancelled Claim 33 and later abandoned the 623 application in favor of prosecuting the 958 application. In this latter application there was no attempt to include any specific claim for a strip buckling member and 20 of the 21 claims included a buckling member which was to be either disc shaped or substantially circular in form. Moreover, Schmidinger's attorneys admitted that these claims were more narrow structurally than were the abandoned claims of 623. The one claim, Claim 4 of application 174,411 filed July 18, 1950 but abandoned, like 623, in favor of 958, which could have been read to include a strip buckling member was not prosecuted and therefore must be presumed to have been abandoned. It seems clear from the foregoing that Schmidinger limited his device to a disc buckling member but if further proof were needed Schmidinger's attorneys made his position crystal clear during prosecution of 958. In attempting to overcome a Patent Office objection to Claims 10 and 11 as lacking in invention over Klahn 450, Schmidinger's counsel argued that "Klahn's device is a strip, not a disc". Finally, all the claims of 023, the reissue patent of 106, in dispute in this litigation describe only a disc, a disc-like or a substantially circular buckling member. Schmidinger was forced to employ this restrictive language by the Patent Office. The case on file wrapper estoppel is conclusive.

The alleged infringing devices variously have rectangular buckling members, not discs and their pull means are not fastened to the support.12

The court below despite the fact that it found that there was no literal infringement of 023 nonetheless disregarded the particular structure and the elements of the claims, broadly employing the doctrine of equivalence. As was said in the John Deere opinion, supra, 38 U.S. at 33, 86 S. Ct. at 702: "Claims as allowed must be read and interpreted with reference to rejected ones and to the state of the prior art; and claims that have been narrowed in order to obtain the issuance of a patent by distinguishing the prior art cannot be sustained to cover that which was previously by limitation eliminated from the patent."

Here, Schmidinger abandoned too much and now faces the limitations imposed on his claims by those abandonments. He cannot be permitted to recover what he lost in the Patent Office proceedings. We hold that there was no infringement.13

III. UNFAIR COMPETITION.

The issues presented under this heading will be discussed briefly. The court below was requested not to disclose any of the details concerning the plaintiffs' asserted trade secrets in its opinion. We need not refer specifically to facts for it is clear from the record that the court below had substantial bases for its findings. The court stated, 243 F. Supp. at 865, as follows: "A careful consideration of that part of the record relating to plaintiffs' cause of action for unfair competition leads the Court to conclude, in light of applicable law, that plaintiffs have not sustained their burden of proving by a preponderance of the evidence that what they claim as confidential information and trade secrets are in fact such, or, if they could be so classified, that defendants have utilized such data in their mass production of the infringing device."

In respect to the trial judge's concept of the law of trade secrets he stated that Schmidinger's work in measuring the snap and unsnap forces of his buckling member "involved no more than the application of ordinary engineering skill to obtain uniformity of these forces." Id. supra, at 865. This smacks of the test of patentability of Section 103, 35 U.S.C., and is not applicable to trade secrets. We quote with approval the Restatement, Torts, Vol. 4, Section 757, at p. 6.14 The applicable rule of law under circumstances substantially similar to those at bar was applied by the Court of Appeals for the Eighth Circuit in McGraw-Edison Co. v. Central Transformer Corp., 308 F.2d 70, 73 (1962). In the case at bar as in the cited case, the sine qua non of a claim of unfair competition consists in the existence of a trade secret. Here there was none and therefore, of course, no information useful to a competitor could have been transmitted.15

There can be no point in remanding the case so that the trial judge may simply rephrase his conclusions of law. His decree was correct. Accordingly, we will affirm the decision of the court below on the issue of unfair competition.

The judgment of the court below on the issue of validity of the claims, save only Claim 26, will be affirmed. The judgment of the court below on the issue of infringement, as indicated, will be reversed.