This diversity action1 in contract concerns the ownership of patent No. 2,945,666 (patent '666), which was developed by Howard G. Freeman, president of The Jamesbury Corporation. Jamesbury, the appellee, initially brought suit for infringement of the '666 patent against Worcester Valve Company, Inc. E. W. Bliss Company, the appellant, intervened in that suit, claiming ownership of the '666 patent. At trial below, reported at 318 F. Supp. 1 (D.Mass.1970), Worcester Valve did not participate; the sole issue was that of ownership of the patent rights.

In 1940, Freeman went to work as an inventor for Rockwood Sprinkler Company, then an independent concern which has since been acquired by Bliss. Among other products, Rockwood manufactured ball valves. Rockwood's ball valves utilized only a single seal or seat, could control the flow of fluid in only one direction, and tended to leak. Rockwood's customers occasionally requested a so-called double-seal ball valve, but Rockwood never developed one during Freeman's employment. In his employment contract with Rockwood, Freeman agreed

"without further consideration to give to Rockwood the full benefit and enjoyment of any and all inventions or improvements which he may make while in the employ of Rockwood relating to methods, apparatus, chemical substances, or methods of producing which are being used, manufactured or developed by Rockwood, or the use, manufacture and development of which was at the time of said invention or inventions in contemplation by Rockwood, and all inventions which are made or worked out on the time and at the expense of Rockwood * * *. Said Freeman further agrees that he will without additional consideration disclose promptly to Rockwood all of the above-described inventions or improvements which he may make while in the employ of Rockwood. * * *"

The major question is whether Freeman's admitted close-to-the-line actions, about to be described, breached this undertaking under the applicable law. Freeman had risen to become Rockwood's director of research, and had developed nineteen patents for Rockwood. By the fall of 1953 he had the confidence that he could develop a double-seal ball valve and profitably market it through a company of his own. He proceeded to read up on techniques applicable to a double-seal ball valve. Through a friend he sought out investors, and set the date of February 2, 1954 as that to which checks should be post-dated. He arranged his timely extrication from Rockwood by demanding a large salary increase on January 13, 1954, predictably not expeditiously granted, and resigned as of January 25. The organization meeting of his new corporation, Jamesbury, was held on January 29.

Up to this point nothing apparently had been put on paper. Not until February 1 or 2 had Freeman begun to make drawings or sketches of the '666 patent concepts.2 Although Freeman has argued that his ideas were not even partly conceived until he left Rockwood, the district court found that before leaving Rockwood's employment he had "virtually" but not "completely" conceived the invention — that he "had gotten to the point where no more than a few additional days or perhaps few hours of thinking was required for him to put his ideas on paper in a form substantially the same as his later patent application." 318 F. Supp. at 7.

In mid-February, Freeman began testing his invention. In April, he encountered a mechanical difficulty in implementing his idea and he was assisted by Oscar R. Vaudreuil, who ran a machine shop. When the invention was finally perfected, a patent application naming Freeman and Vaudreuil as co-inventors was filed. The district court concluded that Vaudreuil was a spurious joint inventor whose name was added to the application to shield Jamesbury, to which the patent was assigned, from a claim that Freeman had made the invention while still at Rockwood. On appeal, Jamesbury strenuously argued that this finding was erroneous.3 The finding is clearly dictum and not binding in other litigation. 1B Moore's Federal Practice ¶ 0.402[2], at 111 (2d ed. 1965). There is, however, evidence to support it, and we cannot say it is clearly erroneous.

Patent '666 was finally issued in 1960. In 1966, Bliss came across testimony given by Freeman in December 1965 and January 1966 in which he claimed to have first conceived of the idea for his invention in the first week of February 1954 and to have begun his drawings on February 1 or 2. These statements caused Bliss to believe that Freeman had actually invented the ball valve while he was still with Rockwood. Under the 1940 employment contract, the '666 patent would then have been Rockwood's, and now Bliss' property. Bliss promptly raised this claim as an intervenor in the Jamesbury-Worcester Valve suit.

The initial issue that confronts us is whether Bliss is barred by the Massachusetts statute of limitations from raising the issue of ownership of patent '666 in 1966. Massachusetts' statute of limitations requires that contract actions be commenced within six years after the cause of action accrued. Mass.Gen.L.Ann. ch. 260 § 2. There is, however, an exception:

"If a person liable to a personal action fraudulently conceals the cause of such action from the knowledge of the person entitled to bring it, the period prior to the discovery of his cause of action by the person so entitled shall be excluded in determining the time limited for the commencement of the action." Mass.Gen.L.Ann. ch. 260 § 12.

Bliss argues that it did not discover evidence that Freeman had violated his contract until 1966 because Freeman had fraudulently concealed the time of his development of the patent.

It is true that Freeman's concealment took the form merely of keeping silent, but silence can be fraudulent concealment by a person, such as a fiduciary, who has a duty to disclose. Stetson v. French, 321 Mass. 195, 72 N.E.2d 410 (1947). Under his contract with Rockwood, Freeman had a duty to disclose any inventions. Moreover, the district court's findings indicate that Freeman took positive steps to conceal his possible liability to Rockwood by listing Vaudreuil as a spurious co-inventor.4 See, e. g., Baker v. Simmons Co., 325 F.2d 580, 582 (1st Cir. 1963), cert. denied, 382 U.S. 820, 86 S. Ct. 49, 15 L. Ed. 2d 67 (1965).

Jamesbury argues, however, that the 1966 testimony did not bring to light facts which revealed fraudulent concealment but only a suspicion of concealment. These suspicions, it argues, are insufficient to demonstrate fraudulent concealment, and hence the statute of limitations was never tolled. It cites two decisions of this court, Burns v. Massachusetts Institute of Technology, 394 F.2d 416, 420 (1st Cir. 1968), and Tracerlab, Inc. v. Industrial Nucleonics Corp., 313 F.2d 97 (1st Cir. 1963), which held that a tolled limitation period begins to run after facts are discovered which reveal fraudulent concealment, but that a suspicion of concealment is insufficient to start the statute running. These opinions were not concerned with the extent of knowledge a plaintiff must have to bring a delayed suit based on fraudulent concealment of a cause of action, but with the extent of knowledge he must have obtained as to the existence of his cause of action to be compelled to bring suit within the limitations period following the acquisition of such knowledge. The facts Bliss became aware of in 1966 were not sufficient to start the period running, but they were sufficient for Bliss to allege fraudulent concealment. This allegation is sufficient for Bliss to temporarily overcome the statute of limitations in order to attempt to prove at trial that Freeman had in fact fraudulently concealed the date of his invention.5

If the 1966 discovery was sufficient basis for suit, Jamesbury argues further, Bliss should have filed earlier because the testimony revealed no new facts to Bliss. Under Burns v. Massachusetts Institute of Technology, supra, a tolled statute begins to run if a plaintiff has facts which should give him knowledge of fraudulent concealment. But the district court found that Bliss had no such knowledge before 1966. Bliss had, of course, seen Jamesbury valves ever since the fall of 1954, and it knew of the patent application and issuance. But there is no evidence that Bliss knew before 1966 that Freeman claimed to have started his invention only one week after leaving Rockwood.

Alternatively, Jamesbury argues that Bliss is precluded by laches from seeking relief after such a passage of time. Laches requires not only a passage of time but also acquiescence in the alleged wrong by the tardy plaintiff. Tracerlab, Inc. v. Industrial Nucleonics Corp., supra at 102. Acquiescence would require that Bliss knew Freeman had invented the '666 ball valve when at Rockwood but that it did not act on its knowledge. The district court having found that Bliss had no knowledge, until 1966, of exactly when the invention was made, laches is no bar to this suit.

The central issue in this case is the interpretation of the word "invention" as it is used in the 1940 employment contract between Freeman and Rockwood.6 Bliss claims an undisclosed idea is an invention; Jamesbury claims an idea must be written down or otherwise converted into tangible form before it becomes an invention. The district court sided with Jamesbury and concluded that the definition of "invention" is "well settled and specific". 318 F. Supp. at 9. In support of this proposition, the court relied on cases which defined "invention". It is this contract interpretation which Bliss appeals.

In interpreting contracts, "[t]here is in fact no `one correct' meaning of an expression; and the party choosing the expression may have no clear and conscious meaning of his own." 3 Corbin on Contracts, § 535, at 16 (1970).7 If there is a plain, ordinary, and proper meaning of a term, that is evidence of how the parties intended to use that term. Similarly, trade or business usage is also relevant. 3 Corbin, supra, § 542, at 108; § 538, at 68-9. The district court concluded that "invention" did have a definite meaning in patent law. Since patent concepts are used in conjunction with the word "invention" in the employment contract, we find the patent definition to have much weight. 3 Corbin, supra, § 552, at 203-204.

There is, in Massachusetts case law, a definition of "invention" which coincides with the usage of the word in patent law. We cannot distinguish the relevant facts of Lamson v. Martin, 159 Mass. 557, 35 N.E. 78 (1893), from this case. Lamson also concerned an invention, made by the key man in a corporation, who was claiming a later date for his invention than that claimed by the other party. The issue was whether an invention had been made by the defendant before or after the company for which he worked, and to whom he had assigned his inventions, had sold its rights to all inventions to the plaintiff. The trial court found that the defendant had conceived of the idea before the sale but had not made a model or drawing until after the sale. It also found that a mechanic would have been competent to convert the pre-sale idea into practice.8 The Massachusetts Supreme Judicial Court held that an idea did not become an invention until it was put into practice or embodied in some tangible form.9 Its theory was that before an idea became a workable reality, "there often lie severe and long-continued labor and repeated failures, and that success is not always achieved by the one who first strikes out the idea." 159 Mass. at 561-562, 35 N.E. at 80.

There are other cases which define "invention" similarly. See National Development Co. v. Gray, 316 Mass. 240, 249, 55 N.E.2d 783 (1944); United States v. Dubilier Condenser Corp., 289 U.S. 178, 188, 53 S. Ct. 554, 77 L. Ed. 1114 (1933); T. H. Symington Co. v. National Malleable Casting Co., 250 U.S. 383, 386, 39 S. Ct. 542, 63 L. Ed. 1045 (1919). Bliss tries to distinguish these cases as priority cases: cases in which each of two individuals claims to have been first in making an invention. If a non-disclosed idea could constitute an invention, courts would have to read minds to determine priority. Therefore, courts will not consider an idea to be an "invention" until it is made tangible. This does not mean, Bliss says, that a non-disclosed idea cannot be an "invention"; it is merely an arbitrary definition created because of difficult problems of proof.

Those same problems of proof which Bliss claims militate in favor of a requirement of tangible evidence of an invention are also present in this case. Here, one party is attempting to prove that an invention was actually made at a point in time prior to its being reduced to drawings. Moreover, the rationale put forth by the Massachusetts court in Lamson v. Martin, supra, is not one of evidentiary convenience; it is an acknowledgement that an "invention" is generally thought to be more than a thought in the inventor's head. Finally, Lamson and United States v. Dubilier Condenser Corp., supra, were not priority disputes between two would-be patent-holders; they were both cases in which the patent-holder attempted to prove he made the invention at a point in time later than that at which another party, asserting rights to the patent, claimed the invention was made. In sum, it would seem that the Massachusetts cases and the patent law cases hold that an idea becomes an "invention" only when it is reduced to some tangible form. No such "invention" existed when Freeman left Rockwood's employ. Cf. Goodyear Tire & Rubber Co. of Akron, Ohio v. Miller, 22 F.2d 353 (9th Cir. 1927).

Bliss cites two cases from other jurisdictions in support of its definition of "invention". New Jersey Zinc Co. v. Singmaster, 71 F.2d 277 (2d Cir. 1934), cert. denied, 293 U.S. 591, 55 S. Ct. 106, 79 L. Ed. 685 (1934), involved an employee who contracted to assign all patentable "ideas" to his employer. The employee conceived the idea in dispute in July, disclosed it to others in September, made a written description in November, and left his job the following April. The court ruled that the invention had been conceived ten months before the employee left work and hence belonged to the employer. The case is distinguishable because it concerned a contract that assigned ideas, not inventions. The focus was on when the idea was conceived as opposed to when the invention was made. The issue of when an idea becomes an invention was not discussed, but the facts suggest that the idea had been reduced to more tangible form in the ten-month period following its conception.

Winston Research Corp. v. Minnesota Mining & Mfg. Co., 350 F.2d 134, 145 (9th Cir. 1965), contains only a brief discussion of a contract provision which required an employee to assign "inventions conceived". The contract itself is not quoted in the opinion; "inventions conceived" is the court's term. The court merely upheld as not clearly erroneous a district court's finding that three inventions had been conceived during the inventor's employment. "Inventions conceived" is neither defined nor distinguished from "inventions made", the term in Freeman's contract, which Bliss argues should be the focal point of our interpretive analysis.

Although the generally understood definition in the trade and the patent law definition support Jamesbury's interpretation that an idea becomes an invention only when put in tangible form, that support is only evidence — but strong evidence — that Jamesbury offers the correct interpretation. Indeed, patent law contains an exception to its own definitional rule so that in some cases the rival inventor who conceives the idea first, but puts it in tangible form after someone else, may secure the patent by proving earlier conception and reasonable diligence in reducing his conception to practice. 35 U.S.C. § 102(g). See Note, Date of Invention: The Varying Standards of Proof, 57 Georgetown L.J. 162, 162-63 (1968). It is still possible that Freeman and Rockwood had a different definition in mind.

The only other evidence put forth by either party is Bliss' assertion that Freeman's practice was to inform Rockwood of possible patentable inventions when they were merely ideas. If true, this would be evidence that the parties interpreted "invention" to mean "idea". The parties' own application of a contract provision is clearly relevant to interpretation, 3 Corbin, supra, § 558, but the evidence at trial does not support Bliss' assertion. In the over 1300 pages of trial transcript and 500 pages of documents, Bliss points to only five pages of testimony as relevant to how the parties applied the contract requirement. Bliss' counsel never asked Freeman, who was on the stand for extensive periods of time, what the practice actually was.

He did ask Freeman about his practice of consulting Rockwood's patent attorney if he felt he or a member of his department had come up with a patentable idea.10 Freeman said he would usually bring the Rockwood patent attorneys "the ideas, usually with a sketch or description if possible, discuss it with them personally, possibly on random occasions send them a letter, send them the idea by mail and receive comments." From this testimony concerning Freeman's practice in informing the patent attorneys, Bliss concludes that Freeman's practice was to inform Rockwood as soon as he conceived a novel idea, even if the idea had not been reduced to drawings or to a model. But Freeman was testifying about how he informed the patent attorneys not about when he informed Rockwood. Moreover, while he implied that he sometimes informed them without using drawings, he did not say that he ever informed them before any drawings were made. It is quite plausible that Freeman could initially consult the attorneys without furnishing a sketch of the proposed product. The implication that no sketches existed at a time when an idea had developed to a point where the patent attorneys were brought in is not only absent from Freeman's testimony, but is also contrary to Bliss' argument, made elsewhere, that such ideas without drawings were a virtual engineering impossibility. Freeman's testimony will simply not support the conclusion that his practice was to inform Rockwood of his ideas before he had put anything on paper.

The interpretive evidence in favor of Bliss' definition of "invention" would seem non-existent. But even if we were to say that there remained some doubt about the definition of that term, we would resolve that doubt in favor of the party who did not choose the wording of the agreement. Marston v. American Employers Insurance Co., 439 F.2d 1035 (1st Cir., 1971); 3 Corbin, supra, § 559, at 262. This agreement was a standard form contract drawn up by Rockwood and signed by Freeman shortly after being graduated from college. Rockwood drew up the contract and had superior bargaining power. See Stedman, Rights and Responsibilities of the Employed Inventor, 45 Indiana L.J. 254, 258 (1970). If the term "invention", after the use of all analytical aids, retained its ambiguity, we would still adopt the Freeman-Jamesbury definition.

Our interpretation is that Freeman had not made an invention, within the meaning of the employment contract when he left Rockwood, because he had not put any of his ideas down in any tangible form. The district court found that Freeman deliberately refrained from making any drawings in order to circumvent the contract requirements. Bliss argues that it would be sound policy to frustrate the success of such bad faith. Perhaps so, but the contract could easily have provided that Rockwood was entitled to any inventions patented by Freeman and relating to Rockwood's business, within a specified number of years after termination of employment.11 To rule as Bliss would have us rule would require the court to attempt to read inventors' minds in order to determine when the essential idea behind an invention was conceived. Such an interpretation is perhaps facially appealing in close cases such as Lamson, where there was some evidence that the critical ideas were conceived before the termination of the contract period, or the instant case, where reduction of ideas to paper followed quickly. But such an approach would also, in cases where ideas were not reduced to practice for many years, involve the courts in something close to retrospective telepathy. It would also require us to circumvent established doctrines of contract interpretation. We are not prepared to face any of those consequences.

As a final argument, Bliss argues that Freeman had a fiduciary duty under Massachusetts law to give Rockwood the benefit of any inventions resulting from ideas conceived on Rockwood's time. This argument relies on three cases, Hoyt v. Corporon, 268 Mass. 544, 168 N.E. 94 (1929), Wireless Specialty Apparatus Co. v. Mica Condenser Co., Ltd., 239 Mass., 158, 131 N.E. 307 (1921), and United States v. Dubilier Condenser Corp., supra. None of these cases supports Bliss' position. They all concern employees hired without written contracts, who had quit and taken inventions with them. The unwritten agreements were construed to require employees who had been hired as inventors, as opposed to designers, to turn over the patents to their former employers. In all cases the employment contracts were oral; in none were there merely ideas that had not taken tangible form. These cases simply do not stand for the proposition that absent agreement, written or unwritten, an employee has a fiduciary duty to turn over ideas to his employer.

There is a common law doctrine of "shop right" which entitles an employer to rights in an employee's invention to which the employer had made some contribution, such as the use of equipment. But the ambiguity of this doctrine and the inability to predict how courts would apply it led to the use of written contracts to allocate inventive rights. Where those rights are allocated by contract, the common law doctrine is superseded. Stedman, supra, at 256-58; Employer's and Employee's Rights, note 11 supra, at 828-30.

The district court's interpretation of "invention" as being an idea reduced to tangible form is supported by the evidence and is not clearly erroneous. Bliss has been unable to prove that Freeman made his invention while still in the employ of Rockwood.

Affirmed.