Full opinion text
MOSK, J. John S. Adkins, an inventor and mechanical engineer, was engaged in the development and improvement of gyroscopes for Lear, Incorporated, a manufacturer of gyroscopes and related mechanisms. Adkins brought this action against Lear, alleging in his first cause of action that prior to 1955 he had conceived certain inventions and made them available to Lear and that on September 15, 1955, he entered into a written agreement with Lear under the terms of which he granted Lear a license to use these inventions in products manufactured by it, in exchange for which Lear was to pay a stated percentage of the net sales price to Adkins as royalties on all products incorporating his inventions. It was averred that from September 10, 1957 onward, Lear had refused to pay royalties to Adkins on numerous products which it manufactured and sold and for which royalties were required to be paid under the license agreement. The jury found in favor of Adkins and awarded him damages in the amount of $904,474.49 for the period from January 1, 1955, to May 31, 1963. A portion of this amount ($16,351.93) was based on a verdict directed by the trial court as to a product manufactured by Lear and designated as the 2156 gyro. A judgment on the jury’s verdict was entered on May 4, 1964. Thereafter, Lear moved for judgment notwithstanding the verdict or, in the alternative, a new trial. (Code Civ. Proc., § 629.) The trial court granted the motion for judgment notwithstanding the verdict, except as to the 2156 gyro, and it granted the motion for a new trial in the alternative as to all products involved in the action. Multiple appeals have been taken; that which involves the primary matters at issue between the parties, numbered L.A. 29204, will be discussed first. L.A. 29204 The license agreement contained two provisions permitting its termination by Lear, and Lear purported to exercise its right to terminate under these provisions by a letter addressed to Adkins dated April 8, 1959. Reduced to its simplest terms, the fundamental questions we are called upon to decide are whether the agreement was validly terminated by Lear and (except as to the 2156 gyro) whether there is substantial evidence in support of Lear’s claim that its products did not incorporate Adkins’ invention, We determine that under a proper construction of the agreement Lear did not effect a valid termination and that not only is there no substantial evidence to support a determination that Lear did not use Adkins’ invention in its products but that the record compels the conclusion that Lear did incorporate Adkins’ invention in the instruments manufactured by it. The 2156 gyro will be discussed separately. Before setting forth the facts, it may be helpful to summarize some fundamental principles and procedures relating to the instant ease. The federal courts have exclusive jurisdiction over actions arising under patent laws (28 U.S.C.A. § 1338) but where a plaintiff seeks to enforce a patent licensing agreement the action “is not a suit under the patent laws of the United States and cannot be maintained in a federal court as such.” (Luckett v. Delpark (1926) 270 U.S. 496, 502, 510 [70 L.Ed. 703, 705, 708, 46 S.Ct. 397] ; Farmland Irrigation Co. v. Dopplmaier (1957) 48 Cal.2d 208, 217 [308 P.2d 732, 66 A.L.R.2d 570].) Where such an action arises in a state court, the court may pass upon the meaning, scope, validity or infringement of the patent. (Lear Siegler, Inc. v. Adkins (9th Cir. 1964) 330 F.2d 595, 600.) A patent is, of course, presumed to he valid until the presumption has been overcome by convincing evidence. Ordinarily, a party will attempt to demonstrate the invalidity of a patent by showing that it has been anticipated by the prior art or that it is obvious or without utility. However, one of the oldest doctrines in the field of patent law establishes that so long as a licensee is operating under a license agreement he is estopped to deny the validity of his licensor’s patent in a suit for royalties under the agreement. The theory underlying this doctrine is that a licensee should not be permitted to enjoy the benefit afforded by the agreement while simultaneously urging that the patent which forms the basis of the agreement is void. (See 4 Walker on Patents (Deller 2d ed. 1965) p. 607.) This doctrine does not prevent one who is not a licensee from challenging the patent’s validity. The license agreement in the present case was, insofar as is relevant here, based not on an existing patent issued to Adkins but on his application for a patent which the parties believed would eventuate in the issuance of a patent. A patent application generally consists of a drawing of the invention, the specifications, and the claims. The specifications are a description of the invention, the manner in which it is constructed, and the objects it is intended to achieve. They ordinarily include an introductory clause identifying the field in which the invention is made, a recitation of the advantages of the invention and the state of the prior art, a brief description of the drawings and its parts, and, in the case of a machine, the manner in which it operates. The claims follow the specifications. The claims must point out the subject matter which the applicant regards as his invention, and they are “the real measure of the invention.” The drafting of claims is a highly specialized and difficult undertaking. The practice is to include both very broad claims and claims of a more restricted type. Frequently, an applicant will state his claims as broadly as is feasible in order to obtain as wide a monopoly as he can, gradually narrowing them to precisely describe his invention as the broader claims are rejected. (See 2 Walker on Patents (Deller ed. 1937) p. 770.) The subject matter of the claims may describe a machine or apparatus designed to perform some function, sometimes called an “apparatus claim,’’ or a method or process of achieving a particular result independent of the machine which produces it, sometimes called a “method claim. ’ ’ The prosecution of the application before the United States Patent Office will, except in very unusual cases, consist of at least one rejection of some or all of the claims of the application on the ground that a search has indicated they have been anticipated by prior inventions, and the subsequent amendment of the claims or specifications by the inventor. (See Seidel, What the General Practitioner Should Know About Patent Law and Practice (1956) A.L.I. p. 60.) He may then file an amendment which will alter the prior claims, cancel them, or add new claims. However, he may not enlarge the disclosure of the application as initially filed by adding new matter in an amendment. (35 U.S.C.A. § 132.) The issuance of a patent necessarily means that the Patent Office did not consider the amendment to constitute new matter in violation of the patent laws, and its determination in this regard is entitled to special weight. (Helms Products, Inc. v. Lake Shore Manufacturing Co. (7th Cir. 1955) 227 F.2d 677, 679.) The process of rejection and amendment continues until either allowance or final rejection of the claims. A rejection is presumably designated as final if the examiner concludes that there is little likelihood that the application can be amended in such a manner that the issuance of a patent would be justified. This procedure of rejection and amendment is so commonplace that one authority has stated that most patent attorneys view with some misgivings those rare instances in which an application is allowed without amendment because it raises the possibility that the applicant has failed to state his claims as broadly as he could have and that a still broader claim might have been allowed, defining the invention (and thus the protection afforded by the pair ent) in a manner broader and more advantageous to. the applicant. (Seidel, op. cit. supra, p. 48.). An application may not claim more than one distinct invention, and a patent may be issued for one invention only. If the examiner at the Patent Office discovers that more than one invention is disclosed by an application he will require a “division,” i.e., the applicant must limit his claims in the application to only one of the inventions but may file a separate application for each of the other inventions disclosed by the original application. It also happens that, after filing an application, the inventor may discover improvements which involve additional subject matter over the original disclosure and which cannot be added by amendment because they constitute new matter. He may then file another application, called a “continuation application” in which he may claim both the disclosure of the original application and the new improvements. If the applicant is satisfied with the claims which have been allowed he may cancel those which have been finally rejected. If he is dissatisfied with the final rejection he may take an appeal to the Board of Appeals or, in certain limited circumstances, may attempt another amendment. (37 C.F.E. § 1.116.) At the time of the prosecution of the application involved in the present ease, the average patent application had a pendency of somewhat over three years and was acted upon by the Patent Office from two to four times. (Seidel, op. cit. supra, p. 61.) On December 29, 1951, Adkins and Lear entered into an agreement providing that Adkins would supervise the fabrication of vertical gyroscopes for Lear, that all new discoveries, ideas and inventions relating to such gyroscopes would become the property of Adkins, and that he would license Lear to manufacture the instruments on a mutually satisfactory royalty basis. The agreement also provided that the physical instruments fabricated would be the property of Lear. A few days later. Adkins commenced employment with Lear in its Santa Monica plant and began to conduct various experiments aimed at improving the accuracy and lowering the cost of manufacturing these instruments. Gyroscopes are used in aircraft to indicate to the pilot the direction and attitude of the airplane. A vertical gyro indicates the pitch and roll of the aircraft and a directional gyro indicates the aircraft’s direction. The gyro seeks to maintain its fixed position in space, and the movements of the aircraft are measured electronically by the gyro and recorded on gauges. The essential unique element of a gyro is a rotor, a heavy disc, which, when it is electrically driven at high speed, seeks to maintain its attitude in space. In all eases the rotor is suspended by its axis, the ends of which are mounted in a ring, known as a gimbal. On some occasions the gimbal ring is itself suspended in a second ring, in which event the rings are known as the inner and outer gimbal rings and axes. The entire structure is mounted through the outer gimbal ring in a frame fixed to the aircraft or other craft served by the gyro. Bach gimbal ring is provided with two short pins or axle ends called trunnions, which project outward from the circumference of the ring directly opposite each other. They lie on a common axis and are inserted into a bearing element which, in the case of the inner gimbal, is fixed to the inner circumference of the outer gimbal ring and, in the case of the outer gimbal, to a supporting frame. There is, accordingly, one pair of bearings in the outer gimbal for the inner gimbal, and one pair of bearings in the frame for the outer gimbal. The bearings are of the ball-bearing type, each consisting of an inner and outer race or ring with a number of balls between the two rings. Among the most important factors in achieving accuracy in a gyro is alignment on the same axis of the pair of trunnions on the opposite sides of the gimbal, and the alignment in parallel planes of the faces of the bearing elements. Accuracy in these coaxial and parallel alignments reduces friction and the exertion of undesirable forces which cause the gyro rotor to drift from its desired stable position in space, which in turn results in inaccurate recordations. In all gyros here in issue, the bearings are fixed in place, either directly to the next outer gimbal or frame, or to intermediate bearing cups affixed to the outer gimbal or frame. Adkins’ invention and the original claims related thereto involved both a mechanism or apparatus which fixed and held each pair of bearings in coaxial and parallel alignment, as explained above, and a method by which pairs of bearings could be positioned in coaxial and parallel alignment. In subsequent amendments the method claims were eliminated. When Adkins was first employed bjr Lear its practice was to bore, tap and ream a pair of coaxial holes directly opposite each other on the gimbal ring or frame which supports the outer ring. The bearings were then fitted into the seats formed by these holes. In order that alignment within acceptable tol- erances could be achieved by this method, precision work by skilled machinists was required and there was a high incidence of rejects. Adkins’ apparatus for alignment, in its simplest form, consists of a pair of bearing cups mounted in oppositely placed receiving holes in the gimbal by use of a mandrel. A mandrel, in this instance, is a finely tooled rod or shaft, and is used by Adkins only as an assembly tool. Its cylindrical circumference is such that a bearing cup will fit snugly over each end of the mandrel. The mandrel, with cups firmly in place, is then inserted within the gimbal. In this instance the gimbal is not a ring but a hollow sphere which is made in two halves, capable of being joined to form the sphere. There is a receiving hole, designed to receive a bearing cup at the bell top of each half sphere. The mandrel, with cups fitted, is first inserted within a half sphere by placing a cupped end of the mandrel into the receiving hole. The other half sphere is then placed over the assembly, so that the opposite cupped end of the mandrel fits into the receiving hole in that half sphere. The mandrel is of such length that when the half spheres are joined the cups project into the receiving holes. These holes are oversize, that is, the cups do not fit snugly into the holes but are allowed to freely position themselves. When the gimbal sphere is in the assembled position, the cups are cemented into the precise position as aligned by the mandrel. As the mandrel is machined to the precise dimensions of the rotor shaft or trunnions of the inner gimbal, as the case may be, the cups will receive the shaft or trunnions in what Adkins claims to be greatly improved alignment. Moreover, aside from the mandrel, which is a reusable tool, and the inner surfaces of the cups which are machined to receive the pair of bearings, no precision machine work is required and production costs are thus substantially reduced. After the cups have been firmly fixed in place, the sphere is disassembled, the mandrel removed, the rotor or inner gimbal is installed, and the sphere reassembled. At the same time that Adkins was engaged in developing his invention, Lear was experimenting in its Grand Eapids, Michigan, plant with a number of other devices to achieve gyro accuracy. Lear claims that it developed the devices used in its models 2151, 2152, 2153 and 2171 independently of Adkins and that these models do not incorporate any of Adkins’ inventions. Adkins disagrees. These models will hereafter be collectively referred to as the “steel gyros” and will be discussed more fully in the portion of this opinion devoted to the question whether Lear utilized Adkins’ invention in its products. On February 24, 1954, Adkins filed an application for a patent in the United States Patent Office relating to his invention. It contained, inter alia, several claims relating to bearing alignment. Claims 7 and 8 related to a method of aligning bearings, and Claim 9 was denominated an alignable bearing structure for supporting an element movable with respect to another element. These claims were rejected by the Patent Office on September 8, 1954, as being fully met by a previous patent issued to one Carlson. Thereafter, on March 7, 1955, Adkins filed Amendment A to his application. This amendment canceled Claims 7 and 8 and replaced them with Claims 18 and 19, which also related to a method of bearing alignment. Three words were added to Claim 9. This was the condition of the application when the license agreement was executed on September 15, 1955. Several years of negotiation preceded the consummation of this agreement, and the parties were represented by counsel in these negotiations. The license agreement expressly provided that it superseded and canceled all prior agreements between the parties, including the one executed on December 29, 1951. Adkins represented in the agreement that he was the owner of the inventions disclosed or intended to be disclosed in the application, which was attached and referred to as Exhibit B. He granted to Lear an exclusive license “under all the claims of . . . Exhibit ‘B’ . . . and any patents issued . . . thereon.” The definition of the word “claims” was limited to those claims which were patented or patentable. Consideration in the amount of $500 was acknowledged, and it was provided further that while the licenses were in effect a stated royalty would be paid on products incorporating Adkins’ inventions and that, in any event, a minimum of $500 a year would be payable as royalties. Two provisions permitted Lear to terminate the agreement. One was contained in paragraph 2(a) and provided; “Lear shall have the right on ninety days’ prior written notice to Adkins, to terminate any one or more of the licenses herein granted.” The other, set forth in paragraph 6 of the agreement, stated, “In the event that . . . the U. S. Patent Office refuses to issue a patent on the substantial claims of the application attached as Exhibit ‘B’, or if such a patent so issued is subsequently held invalid . . . Lear at its option shall have the right forthwith to terminate the specific license so affected or to terminate this entire Agreement and no further royalties shall thereupon be payable under the license so terminated or under this Agreement if Lear shall have elected to terminate this Agreement in its entirety. ’ ’ On March 21, 1957, the Patent Office rejected Claims 9, 18 and 19 on the ground that they had been fully met by a patent granted to one Grenat. Before Adkins filed any amendment in an attempt to meet this objection, Lear wrote him on September 10, 1957, that it had reviewed his patent application and concluded it did not disclose any inventions utilized in any Lear equipment except the 2156 gyro and certain components thereof, that the steel gyros manufactured by it did not come within the scope of the application, and that Lear had made a search of the Patent Office files and believed the method of bearing alignment used in assembling the steel gyros was not patentable. Lear also informed Adkins that it would no longer pay royalties on the steel gyros but would continue to do so on the 2156 gyro. Adkins protested this action and on February 1, 1957, he resigned from Lear’s employ because of its refusal to continue the payment of royalties on the steel gyros. Thereafter, Adkins again amended Claims 9, 18 and 19 in an attempt to meet the objections of the Patent Office relating to the. Grenat patent. These claims were also disallowed and, after another amendment of the application but before any action by the Patent Office on the amendment, Lear wrote Adkins on April 8, 1959, that it was exercising its right to terminate the license agreement under paragraphs 2(a) and 6. -Royalty payments on the 2156 gyro were also discontinued. Subsequently, Adkins canceled the claims in his application going to a method of aligning bearings and substituted therefor a series of claims referring to an apparatus for support bearings. These apparatus claims were allowed by the Patent Office and letters patent issued thereon on January 5, 1960. On the same date, Adkins filed the present action. Lear stipulated that it manufactured and sold the gyros in issue both before and after its letter of termination of April 8, 1959. In summary, a history of the prosecution of Adkins’ patent application, after the license agreement was executed, reveals that Lear refused to pay royalties on the steel gyros after one rejection of Adkins’ claims by the Patent Office, and it notified Adkins that the agreement was terminated after a second rejection at a time when the claims still involved a method of aligning bearings. The trial of the action occupied many weeks, and voluminous testimony was introduced by both parties. The trial court held that as a matter of law Lear had terminated the license agreement insofar as the steel gyros were concerned. Adkins had urged at the trial that even if the agreement was terminated Lear would be liable for damages under the theory of unjust enrichment set forth in Seagren v. Smith (1944) 63 Cal.App.2d 733 [147 P.2d 682]. The court submitted to the jury for determination whether, as to the steel gyros, Adkins’ patent was valid and whether Lear used the invention. The jury deliberated for eight days, and its verdict in favor of Adkins implied that it had determined both that the patent was valid and that Lear had utilized Adkins’ invention in its products. Lear’s motion for judgment notwithstanding the verdict and for a new trial in the alternative was based on numerous grounds, including the asserted invalidity of the patent and noncoverage. The trial court, in granting the motion for judgment notwithstanding the verdict as to the steel gyros, wrote an opinion in which it held that Adkins’ patent was invalid as a matter of law. The court analyzed Adkins’ invention and the steel gyros manufactured by Lear and compared them with the prior art. The holding of invalidity was based on the ground that Adkins’ device did not constitute an invention but was merely an aggregation of old parts and elements performing no new or different function. Although the court purported to do so, it did not decide whether Lear utilized Adkins’ device. The motion for new trial was granted on the grounds of irregularity of the proceedings, excessive damages, errors of law occurring at the trial, insufficiency of the evidence to justify the verdict and that it was against the law. (See Code Civ. Proc., § 657.) Under the doctrine of licensee estoppel, Lear would be prohibited from challenging the validity of Adkins’ patent if the agreement had not been validly terminated. Where a license agreement specifies the conditions under which termination may occur, those conditions must be satisfied in order to effect a valid termination. (Thomson Spot Welder Co. v. Oldberg Mfg. Co. (1932) 256 Mich. 447 [240 N.W. 93, 94] ; see Ellis, Patent Licenses (3d ed. 1958) pp. 365-366 ; 69 C.J.S. p. 785.) Here, the contract provided that it would continue in force until the patent relating to the subject matter of the agreement expired, unless sooner terminated pursuant to its provisions. We must, therefore, turn our attention to the question whether Lear effected a termination of the license agreement. We will consider first the question whether Lear validly-terminated the agreement under paragraph 6. As we have seen, paragraph 6 provided that Lear could terminate the agreement in the event the Patent Office refused to issue a patent on the substantial claims of the application (Exhibit B). The crux of Lear’s argument is that Adkins was required to obtain a valid patent based on method claims and that the agreement related only to those claims set forth in the application at the time the contract was entered into, or their substantial equivalents. Lear also contends that section 6 did not require final action by the Patent Office in rejecting Adkins’ application before the right to terminate arose and that it was justified in terminating the agreement on April 8, 1959, because the Patent Office had refused by its rejections of March 21, 1957, and June 20, 1958, to allow Adkins’ claims relating to bearings. The rebuttal to these contentions is contained in the provisions of the agreement, which neither party asserts is ambiguous. An examination of those provisions demonstrates Lear incorrectly asserts that the license which was the subject of the agreement related solely to method claims contained in the application at the time the agreement was executed. Indeed, the parties could hardly have defined the scope of the license in a broader manner. It was to cover not only patented or patentable claims disclosed or intended to be disclosed by the application attached as Exhibit B to the agreement but also such claims in other applications filed for different but related inventions revealed by the exhibit. Nowhere in the agreement was there a restriction as to the type of claims (i.e., apparatus or method claims) to which the license was to apply, nor any implication that the coverage of the agreement was to be restricted to claims in the application at the time of the agreement. Under section 2(a) Adkins granted to Lear an exclusive license under all the patented or patentable claims of Exhibit B and “any patents hereafter issued thereon’’ to manufacture and sell products containing his inventions. The term “claims’’ was defined as those claims which were patented or patentable. Paragraph 3(a) provided that royalties were to be paid on products incorporating Adkins’ “inventions’’ and that term was defined (par. 1(d)) as “all claims . . . disclosed or intended to be disclosed’’ in Exhibit B “but only to the extent that such claims . . . are patented or patentable ’ ’ by Adkins. Thus it appears that Adkins granted Lear a license on all patented or patentable claims of Exhibit B and the claims of any patents issuing thereon, and Lear agreed to pay royalties on all products incorporating patented or patentable inventions disclosed or intended to be disclosed by Exhibit B, which is defined in paragraph 1(b) as follows: “The phrase Exhibit ‘B’ as used in this Agreement shall be deemed to refer to and include the application for United States Letters Patent filed on February 15, 1954, attached hereto, and all continuations and divisions thereof, and all patents issued or issuing thereon to Adkins with respect to the inventions disclosed or intended to be disclosed therein and all improvements thereof, and all renewals, extensions, reissues of any patents issued purusant thereto. ’ ’ It clearly appears from the provisions quoted above that the parties intended to license not only those claims eventuating in a patent which were set forth in the application at the time of the agreement, but also any claims intended to be disclosed in the application, without any restriction as to the type of claims allowed. The license extended even to related inventions disclosed by the application or improvements of a disclosed invention for which a divisional or continuation application would be required and which would eventuate in a patent. Lear’s obligation to pay royalties was, of course, based on the use of such inventions in its products. The unimportance of particular application claims at the time the agreement was consummated is indicated by the fact that Exhibit B did not constitute a copy of the application as it then existed in the Patent Office but contained claims which, with the knowledge of both parties, had already been rejected. The application was, as we have seen, filed on February 15, 1954, and the claims going to a method of bearing alignment as well as Claim 9 had been rejected on September 8, 1954. Amendment A, also relating to a method of bearing alignment, was filed on March 7, 1955. Both parties were aware of this but nevertheless Exhibit B did not include Amendment A and indicated the claims in the form in which they had appeared in the original application. Lear knew that the method claims in the original application had been rejected prior to the execution of the license agreement, yet the terms of the contract not only failed to restrict the license to method claims or to claims in the application at the time the contract was signed but defined the scope of the license in an extremely broad manner. Under these circumstances, Lear’s contention that it bargained only for a license of claims relating to a method of aligning bearings cannot be sustained. Moreover, there is nothing in the contract to indicate that Lear could unilaterally anticipate Patent Office action and determine, prior to final action by that agency, that Adkins could not obtain a patent on the invention disclosed or intended to be disclosed by the application. Section 6 provided that Lear could terminate the agreement in the event the Patent Office refused to issue a patent on the substantial claims of Exhibit B. There was no way to determine in advance of final action whether a refusal would eventuate or whether Adkins would receive a patent. Neither of the rejections of the Patent Office prior to Lear’s termination of the agreement were denominated final rejections. Lear had the advice of counsel experienced in patent matters in negotiating the agreement, and it must have known that the prosecution of an application consists of the rejection and amendment of the claims of the application and that this procedure is time-consuming. Nevertheless, it unilaterally determined after the first rejection of the claims subsequent to the agreement that it Avould no longer pay royalties on the steel gyros and, after a second rejection, it terminated the agreement altogether on the ground, inter alia, that the Patent Office had refused to issue a patent on the substantial claims of the application. There is not the slightest indication in the agreement that the parties were concerned with intermediate actions of the Patent Office in rejecting particular types of claims, and it cannot be said that there ivas a refusal to grant a patent as to the claims of Exhibit B, as defined in the agreement, on the basis of these intermediate rulings. Lear relies on Marvin v. Mills Allays, Inc. (1939) 31 Cal.App.2d 549 [88 P.2d 230], in support of its contention that it could unilaterally determine prior to final Patent Office action that Adkins would not obtain a patent. There, the agreement provided that the plaintiff granted to the defendant a license to use a described invention. The invention was a one-element device which the agreement stated had been manufactured by the defendant previously. It was also provided that the plaintiff would file an application for a patent on that device and that if he was finally unsuccessful in obtaining a patent for it, royalty payments would cease. The plaintiff filed an application in accordance with the agreement, but the application was denied on the ground that patents had been obtained on the same invention by another. The defendant refused to pay further royalties. Thereafter, the plaintiff obtained a patent for a three-element invention, which included the single-element device the defendant was manufacturing. It does not appear whether the application upon which the patent was issued was the same one as originally filed. The court held that the three-element device of the patent was different from the device manufactured by the defendant, that the plaintiff was unable to secure a patent on his original invention, which was the subject matter of the agreement, and that, therefore, the defendant was not liable for payment of royalties for manufacturing the original invention. The differences between Marvin and the instant case are obvious. The question decided there was whether the plaintiff obtained a patent on the same device as was specifically referred to in the agreement. Here, the agreement related to any patented or patentable invention disclosed or intended to be disclosed by the claims of the application, not to a particular device described. There is dictum in Marvin, not necessary to the decision, that when the Patent Office denied the plaintiff’s application “no patent could be obtained by him except upon a different invention. ’ ’ In view of the normal procedures followed in patent applications, we disapprove this language insofar as it implies that a single rejection of an application by the Patent Office constitutes a final determination that no patent may be issued on an invention disclosed in an application. The next question to be determined is whether the license agreement was validly terminated under paragraph 2(a). Lear had the right, under paragraph 2(a) to cancel the license upon 90 days’ written notice to Adkins. Lear maintains that this provision granted it an absolute right to cancel the agreement at any time without discontinuing the use of Adkins’ invention, while Adkins asserts that when the license agreement is viewed as a whole it shows that Lear could not exercise its right to terminate unless it also stopped utilizing his invention. Lear urges that we should not require the cessation of manufacture as a condition precedent to 2(a) termination because another provision of the agreement (paragraph 3(g)) indicates that the parties had cessation of manufacturing in mind when they entered into the agreement, and if they expressed such a condition precedent as to one contingency they would have, if they so intended, expressed it as a condition to 2(a) termination as well. Paragraph 3(g) provided that Lear must pay Adkins minimum royalties of $500 a year from January 1, 1957 onward so long as the license was still in effect (e.g., whether or not Lear used Adkins’ invention) and that if Lear failed to pay minimum royalties Adkins could cancel the agreement and thereafter Lear would be prohibited from using the invention. A number of cases are cited in support of Lear’s argument and, while their holdings favor Lear, they make clear that the ultimate consideration in determining whether to imply a covenant in an agreement is the intention of the parties as shown by the contract construed as a whole. A covenant may be implied where it is indispensable to effectuate the intention of the parties, where it is clear from the language used that the provision was so clearly within their contemplation that the parties deemed it unnecessary to express it, where it would have been inserted into the contract if attention had been called to it, and where the subject is not completely covered by the contract. (Lippmam v. Sears, Roebuck & Co. (1955) 44 Cal.2d 136, 142 [280 P.2d 775].) There seems little doubt that when the contract here in issue is viewed as a whole, it shows an intention that Lear would discontinue the manufacture of products using Adkins’ invention if it terminated the agreement under paragraph 2(a). One indication of such an intention is that the termination provision of paragraph 6, under which Lear could terminate in the event the Patent Office refused to issue a patent on the substantial claims of the application, would be almost wholly redundant if Lear could, for any reason or no reason, exercise a right of termination under paragraph 2(a) and still continue to use Adkins’ invention. Another indication of the intention of the parties is provided by a comparison of two provisions of the agreement permitting Adkins to terminate in the event Lear defaulted in its payment of royalties. Under paragraph 3(g), as set forth above, Adkins could terminate the agreement if Lear failed to pay $500 in minimum royalties yearly, and thereafter Lear could no longer manufacture products containing Adkins’ inventions. Paragraph 10(a) gave Adkins the right to terminate the agreement in the event Lear defaulted in the payment of earned royalties. However, this provision did not state that Lear would thenceforth be prohibited from manufacturing products incorporating Adkins’ invention. It would be absurd to assume that the parties intended that Adkins could require Lear to stop using his invention if Lear failed to pay royalties of $500 a year but if Lear defaulted in its payment of earned royalties in a very much larger amount, Adkins could terminate the agreement under paragraph 10(a) and nevertheless Lear could continue to use his invention. It seems evident that the parties did not expressly state in some provisions the requirement that Lear discontinue manufacturing products utilizing Adkins’ invention in the event of termination, although they so intended. Also persuasive are the provisions of two agreements entered into between Adkins and Lear on September 15, 1955, the same date as the license agreement. These agreements, relied upon by Lear in its answer as the basis of an affirmative defense, provided that Adkins released Lear from any claims he might have against Lear prior to September 15, 1955, including claims arising out of the agreement of December 29, 1951, except that he retained those rights which were secured to him by the simultaneously executed license agreement. Aside from the rights set forth in the license agreement, all inventions of Adkins prior to and after September 15, 1955, were to belong to Lear When viewed in the light of these contemporaneous agreements, Lear’s insistence that it could exercise paragraph 2(a) termination and still continue to use Adkins’ invention amounts to an assertion that the parties intended by the license agreement to grant to Lear Adkins’ right in the invention which was the subject of that agreement for the $500 recited as consideration for its execution. That is, acceptance of Lear’s contention would mean that although it was clearly understood prior to September 15, 1955, that Adkins’ inventions conceived in his employment would be his own property, Lear could, the day after the license agreement was signed, simply by giving notice of termination under paragraph 2(a), extinguish Adkins’ right in the invention covered by the license agreement because the simultaneously executed contracts released all of Adkins' other rights against Lear arising out of his employment. We cannot construe the license agreement in a manner so manifestly alien to its entire purpose. Adkins argues persuasively that the purpose of paragraph 2(a) was to permit Lear to avoid the payment of minimum royalties in the event it no longer wished to use his invention. Under paragraph 3(g) such royalties were payable while the license was in effect, whether or not Lear used the invention, and if there had not been some provision permitting termination of the license by Lear in the event of nonuse, it would have been liable for the payment of minimum royalties throughout the 17-year life of any patent subsequently issued. We comment on one other matter briefly before concluding the discussion of the license agreement. Lear placed in evidence a letter from one of its employees, stating that he believed he had introduced cemented-in bearing cups using a mandrel procedure in 1953, which was a year before Adkins filed his application for a patent. The device of cemented-in bearing cups using a mandrel procedure constitutes the essence of Adkins’ invention, as described in his application. Lear’s contention amounts to no more than a claim that it entered into a bad bargain when it signed the license agreement. Since we have concluded that the license agreement was not properly terminated by Lear, it follows that Lear has been in breach of the contract since September 10, 1957, the date on which it ceased to pay royalties to Adkins on the steel gyros. It follows also that the doctrine of licensee estoppel is applicable and Lear is foreclosed from challenging the validity of Adkins’ patent. The trial court thus erred in granting Lear’s motion for judgment notwithstanding the verdict as to the steel gyros on the ground that Adkins’ patent is invalid. As set forth above, the court also granted a new trial in the alternative on several grounds, including insufficiency of the evidence. One of the major bases of Lear’s motion for a new trial was the assertion that the evidence was insufficient to support the jury’s implied determination that Lear incorporated Adkins’ invention in its products. The agreement clearly sets forth that royalties are to be paid only for patented or patentable claims which Lear utilized in its products. We are therefore presented with the second major issue in this ease, i.e., whether there is any substantial evidence to support the trial court’s conclusion, implicit in its order granting the new trial, that the steel gyros do not incorporate Adkins’ invention as defined in the claims of the patent. (See Richardson v. Ham (1955) 44 Cal.2d 772, 775 [285 P.2d 269].) We conclude, for the reasons hereinafter stated, that the evidence establishes that Lear used Adkins’ invention in its steel gyros and that there is no substantial evidence to the contrary. In determining the question before us we must ascertain what it is that Adkins invented. The resolution of this issue requires that the claims of the patent be construed, and this is a question of law for the court. (Rogers v. Hensley (1961) 194 Cal.App.2d 486, 491 [14 Cal.Rptr. 870] ; see 4 Walker on Patents (Deller 2d ed. 1965) pp. 66-67.) Although the present action is not for infringement of a patent but rather one for the recovery of royalties under a license agreement, the test whether the products manufactured and sold by the licensee come within the coverage of the patent is the same as that employed in a patent infringement suit. (American Photocopy Equipment Co. v. Ampto, Inc. (1964) 82 N.J. Super. 531 [198 A.2d 469, 475].) Claims must be construed in the light of the specifications and drawings of the patent and the file history of the patent’s prosecution. Claims which have been allowed cannot, under the doctrine of file-wrapper estoppel, be held to cover what has previously been eliminated from the patent application in order to avoid a rejection. (Oregon Saw Chain Corp. v. McCulloch Motors Corp. (9th Cir. 1963) 323 F.2d 758, 762, 768 ; Moon v. Cabot Shops, Inc. (9th Cir. 1959) 270 F.2d 539, 543.) A third rule applicable to the construction of the claims of a patent is that an accused infringer may cite the prior art to construe and narrow the claims of the patent, and if he is successful in demonstrating that he has built his device solely according to the teachings of the prior art, the mere fact that the device constructed reads upon the claims of the patent assertedly infringed does not render him liable. (Westinghouse Electric & Mfg. Co. v. Formica Insulation Co. (1924) 266 U.S. 342, 351 [69 L.Ed. 316, 322, 45 S.Ct. 117] ; Casco Products Corp. v. Sinko Tool & Mfg. Co. (7th Cir. 1940) 116 F.2d 119,121.) Adkins contended at the trial that the four steel gyros were covered by the literal language of the claims as well as under the doctrine of equivalents. Under this principle if two devices do the same work in substantially the same way and accomplish substantially the same result, they are the same, even though they differ in name, form or shape. (Graver Tank & Mfg. Co. v. Linde Air Products Co. (1950) 339 U.S. 605, 608 [94 L.Ed. 1097, 1102, 70 S.Ct. 854].) The issue of infringement is a question of fact for the jury unless there is no substantial conflict in the evidence. (See 69 C.J.S. 1016.) It is manifest from the most cursory examination of the record that the steel gyros, like Adkins’ patent, utilize a pair of bearing cups which are positioned with use of a mandrel into oppositely placed receiving holes in a frame or girnbal, and cemented into place. The 2156 gyro, to be discussed later, like the illustration of the Adkins patent application, utilizes spherical bearing cups which rest within conical receiving holes. The steel-gyros, however, utilize pairs of elements having a different configuration. The receiving holes are cylindrical in shape and the bearing cups consist of cylindrical discs which fit within the receiving holes but because of an oversized flange at the open end of the cup, cannot be inserted entirely within the hole. It is held in place not only by cement between the cylindrical surfaces within the hole, but also by cement between the flanged surface and shoulder of the hole. Our problem, basically, is to determine whether such an apparatus “infringes” the apparatus described in Adkins’ patent. The file wrapper of Adkins’ patent application shows approval and issuance of Claims 9 through 16, not to be confused with earlier claims, some of which bore the same numbers. These are all apparatus claims and differed basically from prior claims in that they related to a pair of bearing structures for the first time. Claim 9, which is the principal claim upon which the other allowed claims depend, states: “An apparatus for supporting bearings in aligned relationship which comprises a pair of bearing-receiving elements each providing means to removably support a bearing in a fixed relationship with said element, each of said bearing-' receiving elements having a mounting surface by which it may be supported, means for supporting said bearing-receiving elements at opposed relatively spaced positions, said supporting means providing supporting surfaces generally corresponding to said mounting surfaces and permitting said elements to be initially adjustably shifted relative to said supporting means into oriented positions where said bearing-supported means are in alignment with each other, and means to retain said bearing-receiving elements in said oriented positions to permit pairs of bearings to be interchangeably mounted in aligned relationship supported by said bearing supporting means.” One of the stated objects of the invention is “. . . to provide a novel method of aligning the various bearings in the gimbal mountings, whereby exact coaxial alignment and parallelism between opposed bearings may be achieved. ’ ’ The receiving holes in the gimbal are illustrated in the drawing submitted with the application as being conical in shape, and the outer contours of the bearing cups as being spherical. In this connection it is stated in the specifications: “The engaging spherical and conical surfaces permit the bearing race axis to be oriented in any direction during the adjusting period whereby coaxial alignment and parallelism of the bearing races may be easily effected. Other surfaces capable of being aligned and fixed in position such as a sphere in sphere, knife edges or other devices may be used without coming outside the scope of this invention. ...” The specifications also provide as follows: " While the use of mating conical and spherical surfaces has been mentioned, it is apparent that other similarly cooperating surfaces can be used. For example, a spherical surface can be used within another spherical surface, and other combinations can be employed. . . . Further, not only may the confronting surfaces ... be of any desired configuration, my invention also is not limited to the alignment of bearings of a gyroscope. . . . Modifications employing the principles of the present invention will occur to those skilled in the art. While a preferred form has been shown and described, it is to be understood that the invention is not to be restricted to the particular form and arrangement of parts herein described and shown, except as limited by the following claims. ’ ’ It is obvious that the literal language of Claim 9 and the specifications, which expressly make provision for ' ‘ other surfaces capable of being aligned and fixed in position, ” “ other similarly cooperating surfaces,” and “confronting surfaces . . . of any desired configuration,” intend to cover within the scope of the invention receiving holes and bearing cups having mating surfaces of any configuration which will permit the angular adjustment of the bearing cups to a common axis, unless expressly excluded by other specific language. We are unable to discern any specific or inferential language of exclusion of the configurations utilized in the steel gyros. The language of the claim itself requires only that the two surfaces “generally correspond.” We cannot infer therein any intent to limit such correspondence to the precise configuration of a spherical bearing cup and a conical receiving hole, any more than to a cylindrical hole and a cylindrical disc which partially rests within the hole, supported by a flanged part of the disc. Moreover, the specifications do not purport to limit the configurations to any one or more combinations of shapes, but instead contemplate all configurations which reasonably might perform the described function. (See. Reiner v. I. Leon Co. (2d Cir. 1960) 285 F.2d 501, 504 ; Filtex Corp. v. Amen Atiyeh (9th Cir. 1954) 216 F.2d 443, 446-447 ; Chicago Pneumatic Tool Co. v. Hughes Tool Co. (10th Cir. 1938) 97 F.2d 945, 946-947.) Relying on the doctrine of file-wrapper estoppel, Lear claims that Adkins is estopped to assert that the allowed claims cover the particular application of the bearing cup and receiving holes utilized in the steel gyros. The record does not support Lear. As stated, Claim 9 allows for bearing cups and receiving holes having mating surfaces of any configuration permitting angular adjustment. Here the accused apparatus falls within the literal language of the claim, thus determining the issue of coverage. (Graver Tank & Mfg. Co. v. Linde Air Products Co. (1950) supra, 339 U.S. 605, 607 [94 L.Ed. 1097, 1101].) Moreover, there is no basis for an estoppel because the particular claims at issue had not been previously rejected and thus had not been limited by Adkins as the result of a prior rejection, and this is also true with respect to the configurations claimed by Lear to differentiate its steel gyros from Adkins' patent. Lear’s further claim that Adkins’ patent describes an apparatus “different from the invention contemplated by the parties and set forth in the patent application at the time of the execution of the license agreement,” is not borne out by the file wrapper. Consistent with the application developments heretofore noted, the patented apparatus falls within the scope of the initial application and, because Lear has adapted it to its steel gyros, we must conclude that the patented apparatus was within the contemplation of the parties. It is also argued that because the spherical cups come to rest in the conical receiving holes and, while thus bearing on the conical surfaces, are cemented into permanent place, Adkins’ patent is limited to those configurations which provide for contact or ball joint action between the elements. There is nothing which supports this contention. Although such contact would normally be made, the scheme of the patent does not require it and the apparatus would be functional with or without contact while the elements were adjusted and cemented into position. Conversely, the elements used in the steel gyros, although normally not in contact, would nevertheless be functional although contact was made. The question of contact, accordingly, is not material to the issues. There is likewise no merit to the assertion that the steel gyros do not infringe upon Adkins’ invention for the reason that alignment may be accomplished in the steel gyros by the lateral movement of the bearing cups, whereas Adkins’ invention does not provide for such movement. Although it is possible that lateral adjustment may accomplish some other desirable function or adjustment in the steel gyros, the alignment of the bearing cups in either the Adkins or steel gyros can be accomplished only by the angular displacement of the two cups to the same axis; that is, the axis of the mandrel, with or without lateral movement of the bearing cups. The added feature of lateral movement in the steel gyros, if it can be deemed as such, does not operate to negate an infringement of the Adkins patent if the steel gyros incorporate any claim set out in the patent. (See Temco Electric Motor Co. v. Apco Mfg. Co. (1928) 275 U.S. 319, 328 [72 L.Ed. 298, 302, 48 S.Ct. 170].) Even if the literal language of Claim 9 did not cover the adaptation utilized in the steel gyros, the doctrine of equivalents would require that we conclude that Lear utilized Adkins’ invention. Here the steel gyros contain the same elements, although perhaps different in size, form and shape, as in the Adkins patent, and they accomplish substantially the same function in substantially the same way. As the Supreme Court said in Graver Tank & Mfg. Co. v. Linde Air Products Co. (1950) supra, 339 U.S. 605, at pages 607-608 [94 L.Ed. 1097, at pp. 1101-1102], “One who seeks to pirate an invention, like one who seeks to pirate a copyrighted book or play, may be expected to introduce minor variations to conceal and shelter the piracy. Outright and forthright duplication is a dull and very rare type of infringement. To prohibit no other would place the inventor at the mercy of verbalism and would be subordinating substance to form. It would deprive him of the benefit of his invention and would foster concealment rather than disclosure of inventions, which is one of the primary purposes of the patent system. The doctrine of equivalents evolved in response to this experience. The essense of the doctrine is that one may not practice a fraud on a patent. ’ ’ It remains to be seen whether the steel gyros, although falling within the claims of Adkins' patent, nevertheless can be shown to rely solely on the teachings of the prior art, and thus permit Lear to escape any liability to Adkins. As stated above, we are required to examine the prior art not only to determine the scope of Adkins’ invention but also to determine whether what is being biiilt by Lear springs entirely therefrom. (See Casco Products Corp. v. Sinko Tool & Mfg. Co. (7th Cir. 1940) supra, 116 F.2d 119,121.) Lear claims that “if defendant’s devices, when compared with the teachings and disclosure of the prior art as construed by the court, accomplish substantially the same thing, and perform substantially the same result, in substantially the same way by substantially the same means as do the prior art patents then as a matter of law, defendant’s products cannot infringe plaintiff’s later-issued patent. Further, even if every detail of defendant’s device is not completely taught by such prior art, but if it would require only ordinary mechanical skill to come up with defendant’s devices in view of such prior art, then again defendant’s devices as a matter of law, cannot infringe plaintiff’s patent or incorporate a patentable invention of plaintiff.’’ The trial court concluded that the prior art did, in fact, anticipate the steel gyros. It and Lear rely on the prior Schwan and Moody patents. The Schwan patent discloses a device for aligning the rotor of a motor. It provides for a bearing, a bearing cup, and a means of fixing the cup to the frame of the motor. In these respects it is similar to the Adkins patent. However, Sehwan does not provide for angular tilt of the bearing cups, but rather for alignment of the bearings solely by the lateral movement of the cups. Parallelism is achieved in Sehwan solely by the built-in parallelism of the surfaces to which the bearing cups are attached, whereas Adkins teaches that parallelism as well as coaxial alignment is to be achieved through the relative angular displacement of the bearing cups. Clearly, Sehwan does not provide for the same kind or degree of cooperative relationship between the elements as Adkins, and does not accomplish the same results. Lear utilizes the Adkins distinctions and relies on the Adkins results. In the Moody patent, bearings and bearing cups are fitted onto a shaft of a rotor and mounted by means of a press fit into a frame. The press fit is loose enough to permit the bearings to assume the alignment of the shaft when tapped with a hammer. Unlike Adkins, Moody does not provide for a prealignment of the bearings prior to the mounting of the shaft, does not permanently fix the bearing cups in position, and does not allow for removal and replacement of the bearings without realignment thereafter. Again, Lear’s devices utilize all those distinctions by which Adkins’ invention differs from that of Moody. Other patents were urged by Lear as constituting prior art on which both Adkins and Lear relied. Thus, in Grenat, parallel and coaxial alignment is initially achieved, but bearing-cups are not utilized, and bearings cannot be interchanged without disturbing the alignment. In Sperry, Carlson, and Herr, a “universally mounted” bearing is utilized which constantly adjusts itself to maintain alignment while in operation. It thus does not provide fixed coaxial and parallel alignment. In no respect was the prior art demonstrated to provide, by individual patent or collectively, a means of permanently fixing a pair of replaceable bearings on coaxial and parallel alignment. We cannot sajr, as urged by Lear, that the prior art disclosed a means by which it could accomplish substantially the same result, in substantially the same way by substantially the same means, as does the Adkins patent. Nor can we say that by the application of ordinary mechanical skills to the prior art, the same result may be accomplished. The prior art fails to anticipate the Adkins patent, or the utilization made thereof by Lear. Although the basic elements of the patent—bearing cups, and a means of attaching the cups— have heretofore been utilized in cooperation with each other, Adkins nevertheless makes a significant step forward in the innovation and utilization of the cooperative relationship by which he employs these elements. (See Bates v. Coe (1878) 98 U.S. 31, 48 [25 L.Ed. 68, 74] ; International Manufacturing Co. v. London, Inc. (9th Cir. 1964) 336 F.2d 723, 726.) In Bates, supra, at page 48 of 98 U.S. [25 L.Ed. at p. 74], the United States Supreme Court pointed out that where the thing patented consists of a combination of old elements, incapable of division or separate use, the alleged infringer cannot escape liability by showing that separate parts of the entire invention are found in various prior patents because such evidence does not warrant the conclusion that the plaintiff is not the original inventor of the patented improvement asserted to be infringed. We conclude that the record demonstrates, without substantial conflict, that Lear utilized the apparatus patented by Adkins throughout the period in question. This conclusion is significantly supported by Lear’s admission, hereinafter considered, that the 2156 gyro infringes the Adkins patent. The record shows that the 2156 gyro differs from the steel gyros only by the configuration of the mating surfaces of the bearing and receiving holes. Accordingly, the order granting the motion for a new trial cannot be affirmed on the ground of insufficiency of the evidence. Indeed, when the claims of the patent are interpreted in the manner set forth above, the conclusion that Lear utilized Adkins’ invention in its steel gyros is compelled by the record. The motion for a new trial was also granted on the ground that the jury awarded excessive damages under the influence of passion or prejudice. A witness for Adkins testified that the royalties due under the provisions of the contract were $904,474.49, which was the exact amount awarded by the jury. This figure was based upon data supplied by Lear as to sales of products assertedly incorporating Adkins’ invention, as well as upon Adkins’ views as to which products were royalty-bearing. Adkins’ contention, which the jury accepted, was that under the agreement Lear was required to pay royalties not only upon the gyros which utilized his apparatus but also upon certain other com