This Missouri diversity action involves a claim by A.L. Laboratories, Inc., against North American Philips Corp. and one of North American’s then wholly owned subsidiaries, Philips Roxane, Inc., for misappropriation of proprietary data. A.L. Labs *380appeals from the district court’s1 grant of judgment notwithstanding the verdict reducing the award of compensatory damages against Philips Roxane from $340,000 to $1 and overturning the $1,570,000 award of punitive damages against North American. In addition, A.L. Labs argues that the injunctive relief entered in its favor was inadequate and that it was entitled to attorney fees. Philips Roxane on cross-appeal seeks to avoid all liability on the basis that there was no misappropriation and in the alternative challenges the district court’s refusal to reverse an award of $785,000 punitive damages against it. We affirm.
The dispute between A.L. Labs and North American and Philips Roxane concerns the development and submission of scientific support data needed to gain the Food and Drug Administration (FDA) approval which is prerequisite to the marketing in the United States of new animal drugs. The FDA places scientific data submitted to it in a file bearing the name of the party who “owns” such data. A subsequent applicant for drug approval who wishes to rely on data already in the FDA’s possession may do so only with permission from the “owner” of the relevant file. See 21 C.F.R. § 514.1(a) (1985). Data thus do not become “general knowledge” after development and submission; rather, each subsequent applicant must obtain the data at its own expense.
In the early 1970’s A.L. Labs2 began seeking the help of an American company regarding the approval and marketing in the United States of its animal drug zinc bacitracin. In November 1973, A.L. Labs signed a marketing agreement with Thompson-Hayward Chemical Co., a wholly owned subsidiary of North American, under which Thompson-Hayward was to receive exclusive U.S. distribution rights for A.L. Labs’ zinc bacitracin in exchange for Thompson-Hayward’s best efforts in helping A.L. Labs obtain FDA approval for the drug.
Certain scientific data necessary to the FDA approval subsequently were developed through a joint study coordinated by the Animal Health Institute (AHI), an industry trade association. Results of the study were to be available only to sponsors who paid a share of the research costs, and amounts billed by AHI were paid by Thompson-Hayward and reimbursed by A.L. Labs. Philips Roxane, however, initially had been listed in name as the participant in the study; and when the data were submitted to the FDA, that agency placed them in a file previously set up in Philips Roxane’s name.
A.L. Labs’ zinc bacitracin was approved in April 1976, but that company and Thompson-Hayward within five months terminated their marketing agreement. The termination agreement provided that Thompson-Hayward would “take such steps as are necessary to transfer or assign to AL * * * any other FDA submissions made by or on behalf of TH relating to AL's Zinc Bacitracin.” Joint appendix at 568. Philips Roxane subsequently on three occasions referred to the AHI study data in seeking FDA approval for the zinc bacitracin of another foreign company. After A.L. Labs filed this suit alleging misappropriation of its data, Philips Roxane in an attempt to moot the controversy paid $30,-000 to another AHI study participant for the right to refer to that company’s data. The FDA allowed the drug approval originally obtained with the data claimed by A.L. Labs (the other applications apparently are still pending) to stand based on Philips Roxane’s substitution of the equivalent data from the different source.
*381In reviewing the contentions of the parties, we are mindful that the jury by special verdict found for A.L. Labs in all respects. Thus, we must resolve factual conflicts in favor of A.L. Labs and give A.L. Labs the benefit of all reasonable inferences. See Craft v. Metromedia, Inc., 766 F.2d 1205, 1218 (8th Cir.1985), cert. denied, --- U.S. ---, 106 S.Ct. 1285, 89 L.Ed.2d 592 (1986). A jury’s verdict may not be overturned if the evidence when viewed in the manner most favorable to the prevailing party would allow reasonable jurors to differ regarding the conclusions that could be drawn. See Thomas v. Booker, 784 F.2d 299, 305 (8th Cir.), cert. denied, --- U.S. ---, 106 S.Ct. 1975, 90 L.Ed.2d 659 (1986); Craft, 766 F.2d at 1218; see also Easley v. Empire Inc., 757 F.2d 923, 928 (8th Cir. 1985) (jury verdict at Missouri law may be overturned only in the “complete absence of probative facts”).
I.
Philips Roxane argues that A.L. Labs was entitled to no relief because the scientific data from the AHI study do not constitute trade secrets and because Philips Roxane acquired the results of the study in a lawful manner. The data, Philips Roxane contends, were known throughout the industry, or at least to the other sponsors of the study, with no measures taken to ensure that those sponsors would maintain secrecy. The district court, however, instructed the jury that
[t]he fact that more than one company or enterprise has access to information does not necessarily prevent the information from being a trade secret. All that is required is that the information is not publicly known, and that it would be difficult or expensive for a competitor not lawfully possessed of the information to acquire it by fair means.
******
Also, the fact that information or data is developed in cooperation with other companies or joint venturers, or through a consultant or other party assisting in its development, does not mean that such information or data is not a trade secret. It may still be a confidential trade secret, provided that, in fact, it is known only to the venturers or consultants and is not generally known in the industry.
Joint appendix at 749-50.
These instructions, which Philips Roxane does not challenge, are in keeping with Restatement of Torts § 757 (1939), adopted by Missouri courts in determining the existence of trade secrets. National Rejectors v. Trieman, 409 S.W.2d 1, 18 (Mo.1966). As the Restatement comments make clear, a trade secret requires not absolute secrecy but “a substantial element of secrecy * * * so that, except by the use of improper means, there would be difficulty in acquiring the information.” Restatement, supra, § 757 comment b. The evidence is undisputed that it would have taken Philips Roxane several years and several hundred thousand dollars to duplicate the AHI studies, and even in subsequently obtaining the right to the data from another study sponsor Philips Roxane was willing to pay $30,-000. One characteristic of a trade secret is that it is vendible, with its sale value depending on its secrecy. Id. comment c. National Rejectors, relied on by Philips Roxane, stands only for the proposition that development costs and value alone cannot justify limitations on the use of information already freely presented to the general public. See 409 S.W.2d at 20. The requirement of the FDA that an applicant for drug approval have permission to reference scientific data previously submitted by another company, plus the AHI’s limits on the availability of the joint study data, plus the ultimate willingness of Philips Roxane to buy a right of reference all suggest that this is not a case involving information of free and general circulation. A.L. Labs presented ample evidence from which a reasonable jury could have found that the AHI study data constituted trade secrets.
Philips Roxane next argues that, even assuming trade secrets, it could not have been guilty of misappropriation because it properly acquired the data in question when it was designated by the FDA as *382the owner of a file in which the AHI study results were placed. This argument, however, may be summarily rejected. The relative rights of A.L. Labs, Thompson-Hayward, and Philips Roxane to the AHI study data are governed by the relationships among those companies, and an error by the FDA is not sufficient to alter those rights.
The record shows testimony suggesting that both A.L. Labs and Thompson-Hayward contemplated that Thompson-Hayward would rely on the regulatory expertise of Philips Roxane in obtaining FDA approval for the zinc bacitracin and that A.L. Labs financed a visit by a Philips Roxane scientist and gave confidential data to Philips Roxane prior to the signing of the marketing agreement. Joint appendix at 147-53, 178, 233, 264, 267, 469-72, 552. While Philips Roxane in July 1973 submitted its name as a potential sponsor of the AHI study, it at the same time wrote Thompson-Hayward that its “main purpose was to get us listed so that Thompson-Hayward could be included as a cooperator.” Id. at 574. Then, in March 1974, when it came time for prospective sponsors to sign letters agreeing to pay proportionate shares of the costs of the research, Philips Roxane wrote Thompson-Hayward that if Thompson-Hayward was interested, Thompson-Hayward should contact the AHI directly because Philips Roxane was “not in this area of marketing” and “personally [was] not going to participate in th[e] study.” Id. at 602.
The record further includes correspondence from Thompson-Hayward to the AHI explaining that Thompson-Hayward would be sharing in the costs of the joint study and that the Philips Roxane address had been used only because Philips Roxane personnel were providing the scientific expertise and certain correspondence was to be addressed to them. Id. at 584. Subsequently, when the AHI submitted the study results to the FDA, it listed Thompson-Hayward, and not Philips Roxane, among the companies entitled to use the data. Id. at 589-90; see also id. at 183, 212-14. Philips Roxane did not take any part in actually conducting the studies, id. at 241, 450; and while Philips Roxane claimed to have invested 400 to 700 hours of time and expertise overall (i.e., not limited to time spent relating to the AHI study) in connection with gaining FDA approval for the A.L. Labs zinc bacitracin, other testimony suggested that the $60,000 cash paid by A.L. Labs to Thompson-Hayward under the termination agreement was to reimburse expenses of Thompson-Hayward and Philips Roxane. Id. at 109-10. From this evidence a reasonable jury certainly could have found that Philips Roxane participated in the AHI study only on behalf of A.L. Labs and not in its own right and thus had no entitlement to the data thereby generated.
Finally, Philips Roxane argues that its use of the AHI study results was not tortious because it did not acquire the information through a confidential relationship. A “confidential relationship,” Philips Roxane asserts, is synonymous with a fiduciary relationship, with the necessary dominance or special influence lacking here because A.L. Labs, Thompson-Hayward, and Philips Roxane engaged in an arm’s length business transaction. As revealed by the cases discussed by the Missouri court in National Rejectors, however, the tort of misappropriation addresses as “confidential relationships” those instances where a company discloses its confidential information to a second company for some specific purpose and the second company uses the information other than for that purpose. 409 S.W.2d at 34-36; e.g., Restatement, supra, § 757 comment c. Furthermore, no particular form is required for a confidential relationship; the question, rather, “is simply whether in the circumstances B knows or should know that the information is A’s trade secret and that its disclosure is made in confidence.” Id. § 757 comment j. Certain disclosures thus may be considered confidential by general implication. E.g., Sandlin v. Johnson, 152 F.2d 8, 11 (8th Cir.1945) (licensing arrangement); see also Burten v. Milton Bradley Co., 763 F.2d 461, 463 (1st Cir.1985) (“Where the facts *383demonstrate that a disclosure was made in order to promote a specific relationship * * the parties will be bound to receive the information in confidence.”).
A.L. Labs’ position is that the AHI study data were revealed to Philips Roxane by A.L. Labs for Philips Roxane’s use in obtaining FDA approval for A.L. Labs’ zinc bacitracin. The information thus, by general implication as well as in accordance with the general practices of the parties, including Philips Roxane, see joint appendix at 234-35, was to be treated as confidential and used only for A.L. Labs’ benefit. Philips Roxane argues that a confidential relationship could not have existed because the marketing agreement ran only between A.L. Labs and Thompson-Hayward. In addition, Philips Roxane contends that references to confidentiality in, for example, the marketing agreement applied only to data then in existence and in the possession of A.L. Labs and not to the subsequently developed AHI study data. These arguments, however, are inconsistent with the factual scenario outlined above regarding Philips Roxane’s role in representing A.L. Labs in the joint study, and a reasonable jury similarly could have found that a confidential relationship existed within the meaning of Missouri trade secrets law.
II.
Turning to the monetary relief assessed against Philips Roxane, A.L. Labs argues that the district court erroneously rejected the unjust enrichment measure of recovery when it reduced the jury award of $340,000 compensatory damages — approximately the amount of research costs avoided by Philips Roxane through its misappropriation — to $1 nominal damages. The district court in the essence of its opinion, however, accepted the unjust enrichment theory; it found, rather, that no enrichment had been shown because Philips Roxane properly acquired the AHI data when it purchased a valid right of reference from another study participant and A.L. Labs had offered insufficient evidence to create a jury issue regarding the value of the information to Philips Roxane during the twenty-eight months of improper possession. A.L. Labs on appeal does not challenge the insufficiency holding but does argue that the question regarding the ability of other study participants to sell rights of reference to the AHI data should have been submitted to the jury rather than decided by the court in ruling on equitable relief. Because, however, the FDA had agreed that if the right of reference were invalidated it would void any drug approvals Philips Roxane had obtained through use of the A.L. Labs data, Philips Roxane still would not have been unjustly enriched and there would have been no compensatory damages issue for the jury. Accordingly, the district court did not err in reducing the compensatory award to nominal damages.
Philips Roxane argues that there was also no punitive damages issue for the jury because there was no evidence that Philips Roxane in referencing the AHI study results in submissions to the FDA acted other than in good faith reliance on the representation of its counsel. Sam Musser, an executive vice president with Philips Roxane, testified by deposition, however, that Philips Roxane relied not specifically on the advice of counsel but on “the people in North American Philips,” joint appendix at 285; see also id. at 289; moreover, the counsel on whose advice Philips Roxane relied was Kleon LeFever, an employee of North American. Furthermore, Musser was unable to pinpoint when and to whom LeFever allegedly gave the legal opinion that Philips Roxane had the right to reference the AHI data for its own uses. Id. at 285. LeFever in his deposition testimony denied having given any such advice to Philips Roxane prior to the institution of the present litigation, id. at 423-24, but during his in-court testimony Le-Fever produced a memorandum showing that he had discussed the matter by phone with Musser on October 3, 1979.3 Finally, *384in keeping with Musser’s testimony, see id. at 285, Philips Roxane’s argument is that it consulted LeFever to learn the details of the A.L. Labs termination agreement with Thompson-Hayward and to learn whether that agreement affected Philips Roxane’s rights. In the face of memoranda showing that Philips Roxane knew that A.L. Labs paid for the AHI study, id. at 596, 599, 601, and the documentary evidence discussed earlier suggesting that Philips Roxane participated in the AHI study only on behalf of A.L. Labs, the jury could have found bad faith in Philips Roxane’s failure to even question its initial “acquisition” of the information.4 Evidence is not so lacking as to justify overturning the jury’s award of $785,000 punitive damages against Philips Roxane. See Easley, 757 F.2d at 929.
III.
In addition to the damages awarded by the jury, A.L. Labs obtained equitable relief from the district court in the form of an injunction restraining Philips Roxane until August 27, 1985, from marketing or permitting others to market any bacitracin product for which FDA approval was gained through reference to the disputed data.5 A.L. Labs contends, however, that it was entitled to a permanent injunction or, in the alternative, that the fifty-one-month period of the restraint should have run only from the time of judgment and not retroactively from the time of the misappropriation.
The entitlement to a permanent injunction, A.L. Labs argues, follows under Missouri law from Philips Roxane’s bad faith and willful conduct. In support of this position A.L. Labs cites Smith v. Dravo Corp., 203 F.2d 369 (7th Cir.1953), a case it claims was adopted by the Missouri Court of Appeals in Reddi-Wip, Inc. v. Lemay Valve Co., 354 S.W.2d 913, 918 (Mo.Ct.App. 1962). The district court, however, relied on the fairly explicit language of two subsequent Missouri Supreme Court cases to conclude that state law would not allow entry of a permanent injunction. See Carboline Co. v. Jarboe, 454 S.W.2d 540, 552 (Mo.1970); National Rejectors, 409 S.W.2d at 43; cf. Sigma Chemical Co. v. Harris, 794 F.2d 371, 374-75 (8th Cir.1986) (conclusion of district court that employee had been under a “temporally unlimited” duty not to disclose trade secrets was in conflict with the opinion of the Missouri Supreme Court in National Rejectors). In a diversity case we accord great deference to the district court’s interpretation of state law, McAninch v. Traders National Bank, 779 F.2d 466, 469 (8th Cir.1985), cert. denied, --- U.S. ---, 106 S.Ct. 2917, 91 L.Ed.2d 545 (1986), and we therefore affirm the district court’s denial of a permanent injunction.
The alternative to permanent injunctive relief, a “head start” injunction, compensates for the time saved by a defendant by misappropriating a trade secret rather *385than reproducing the information on its own.6 A.L. Labs does not contest the district court’s conclusion that the head start gained by Philips Roxane was fifty-one months — thirty-six months to replicate the AHI study plus fifteen months to gain an FDA drug approval based thereon — but instead argues that Philips Roxane would be deprived of its head start only if it were enjoined from marketing bacitracin for fifty-one months from the time of judgment (i.e., from May 6, 1985, through early August 1989). A.L. Labs’ argument, however, assumes that absent the misappropriation Philips Roxane would not have commenced to replicate the AHI study until May 6, 1985.
An analysis of the district court’s remedy shows that its injunction put Philips Roxane in the same position as if, on May 28, 1981, instead of referencing the AHI study data to the FDA, Philips Roxane had begun replicating the research independently. Philips Roxane then would have been able to submit its own data to the FDA on May 28, 1984, and a drug application filed at that time in reliance on that data would have been approved by the FDA by approximately August 27, 1985. A.L. Labs has offered insufficient proof of any necessary postapproval development that would have delayed Philips Roxane’s marketing of bacitracin beyond that time. The early approval of the bacitracin was the only benefit to Philips Roxane shown by A.L. Labs, and the district court’s injunction removed that benefit by in essence delaying the date of approval to August 27, 1985.
IV.
Turning to the judgment notwithstanding the verdict in favor of North American, A.L. Labs argues that there was sufficient evidence to support the award of $1,570,000 punitive damages on the theory either that by reason of North American’s dominance and control over Philips Roxane corporate forms should be disregarded and the parent held responsible for the acts of the subsidiary, or that by reason of North American’s role in the decision to reference the AHI study data North American should share in the direct liability for the misappropriation.
Corporate forms may be disregarded and separate entities treated as one only where a two-prong test is met: “Not only must the corporation be controlled and influenced by one or a few persons, in addition, the evidence must establish that the corporate cloak was used as a subterfuge to defeat public convenience, to justify wrong or to perpetuate fraud.” Fairbanks v. Chambers, 665 S.W.2d 33, 37 (Mo.Ct.App. 1984). Thus, full ownership, with concomitant dominance and control, is insufficient to justify “piercing the corporate veil”; rather, a plaintiff must prove that equity requires that the parent and subsidiary be treated as one. Camelot Carpets v. Metro Distributing Co., 607 S.W.2d 746, 749-50 (Mo.Ct.App.1980) (quoting Lawton-Byrne-Bruner Insurance Agency Co. v. Stiers Brothers Construction Co., 186 S.W.2d 480, 484-85 (Mo.Ct.App.1945)); see also Community Title Co. v. Roosevelt Federal Savings & Loan Association, 670 S.W.2d 895, 905 n. 8 (Mo.Ct.App.1984) (dicta).
The record before us is devoid of any evidence that North American had other than a lawful purpose in establishing Philips Roxane as a subsidiary rather than maintaining Philips Roxane’s operations within the parent corporation. And while North American at times may have exercised great control over certain functions of Philips Roxane, neither is there any sug*386gestión that North American ever manipulated the corporate distinction between itself and Philips Roxane to perpetuate fraud. A.L. Labs’ argument, rather, relies on a blurring of the corporate lines between Philips Roxane and Thompson-Hayward in regard to performance of Thompson-Hayward’s obligations under the A.L. Labs marketing agreement. For example, a representative of A.L. Labs testified that when the original marketing agreement was being negotiated, North American represented that both Thompson-Hayward and Philips Roxane would be involved in the project and that North American determined that that arrangement would be formalized through a contract with Thompson-Hayward only and then circulated the proposed contract to its two subsidiaries. Joint appendix at 146-47, 151-52. North American also negotiated the termination agreement, id. at 73, 156, and the contract with the foreign company on whose behalf the A.L. Labs data subsequently were referenced. Id. at 600 (in fact, it was not even determined at the outset whether this contract would be with Thompson-Hayward or Philips Roxane, id. at 311). A memorandum in the record places a North American representative (plus a Thompson-Hayward representative) at at least one of an apparent series of meetings at which the questionable morality and legality of using the A.L. Labs data were at some point raised, and the availability of the data was identified as an “undetermined cost factor” at a time when the question of which party would bear the costs of FDA approval seemed to be a major issue in the contract negotiations. Id. at 600-01. Finally, North American admitted that when Philips Roxane submitted to the FDA an application for new animal drug approval on behalf of the second foreign corporation, North American participated in the decision or gave advice or was consulted. The district court apparently concluded that Missouri would impose liability on North American not for merely creating and perpetuating the blurring, between its subsidiaries but only for directly making or influencing the decision to use the challenged data. The district court however ruled that it would be mere speculation to conclude on the basis of the foregoing evidence that North American had overridden the qualms of Philips Roxane on the specific detail of use of the A.L. Labs data. Although it is a close question, we again defer to the district court’s determination of the requirements of Missouri law, see McAninch, 779 F.2d at 469, and affirm its decision that the punitive damages award against North American should be set aside.
V.
Because we uphold the district court’s orders regarding actual and punitive damages, we need not reach its conditional grant of new trials for both Philips Roxane and North American. We agree with the district court’s well-reasoned opinion denying A.L. Labs’ request for attorney fees.
The judgment is affirmed.
. The Honorable Howard F. Sachs, United States District Judge for the Western District of Missouri.
. A.L. Labs actually did not exist until December 1975, when it was created as the U.S. subsidiary of A/S Apothekernes Laboratorium for Specialpraeparater of Oslo, Norway. Since the district court found that A.L. Labs succeeded to all the rights of A/S Apothekernes in this matter, and that finding is not challenged on appeal, we shall use "A.L. Labs" on all references for the sake of clarity.
. Even assuming the conversation took place as LeFever alleged, LeFever’s own testimo*384ny suggests that he did nothing more than pass information from a Thompson-Hayward executive to Musser, with no enhancement as a result of his own legal knowledge. Joint appendix at 529. Furthermore, the Thompson-Hayward executive was involved in the project for which the A.L. Labs data was used and possibly even attended a meeting at which Philips Roxane’s right to the data was questioned, id. at 600-01, thus making the channeling of the executive’s knowledge through corporate counsel seem more like a "laundering” of the information or an after-the-fact rationalization and diminishing the credibility of the argument of good faith reliance on counsel.
. Philips Roxane makes much of a statement of the district court that Philips Roxane’s "acquisition” of the AHI data was "innocent.” See Memorandum and Order of July 23, 1985, at 9 n. 4. We read the district court’s opinion as referring merely to the manner in which the data came to be in an FDA file in Philips Roxane’s name, an occurrence which, as we stated earlier, could not increase Philips Roxane’s rights in the study results. There is no inconsistency in a finding that Philips Roxane subsequently was willfully blind as to the source of its right to the data in its name.
. The date stated in the original order was August 13, 1985, but the district court apparently subsequently recalculated the running of the fifty-one-month period.
. Carboline also suggests that on the facts of a given case the proper head start period instead would be the length of time between the misappropriation and the point when the trade secret otherwise became public. 454 S.W.2d at 552. Thus, A.L. Labs arguably would have been entitled to an injunction of only twenty-eight months, that being the period of time between when Philips Roxane first referenced the AHI study data to the FDA and when Philips Roxane purchased the right to use the data from another study participant. Since this alternative is even less favorable to A.L. Labs than the measure used by the district court, we need not consider it in resolving the issue before us.