This is an appeal by plaintiff Moraine Products (hereinafter Moraine) from a judgment entered on a directed verdict in favor of defendant ICI America, Inc. (hereinafter Atlas) on Counts I, II, and III of Moraine's complaint after presentation to the jury of the plaintiff's case in chief. The three counts of Moraine's second amended complaint that were tried consisted of two antitrust counts and a count for alleged tortious interference with business relationships. Moraine's claim of injury under Section 4 of the Clayton Act, 15 U.S.C. § 15, was based on its ownership of Patent No. 3,422,189 (the Rider patent), and it charged conduct violative of Section 1 of the Sherman Act, 15 U.S.C. § 1, relating to the licensing of the Feinstone patent, which was originally issued as Patent No. 2,951,011 on August 30, 1960, reissued on July 24, 1962 as Reissue No. 25,205, and which was owned at all pertinent times by Plough, Inc., a codefendant at the trial. Moraine's primary theory of recovery was that a license agreement between Plough and Atlas1 of January 19, 1961, was illegal per se. Alternatively, Moraine contended that, even if the contract was not unlawful per se, the evidence established that the defendants had conspired or combined to impose an unreasonable restraint on commerce.2

The challenged agreement involved a grant by Plough to Atlas of a license stated to be exclusive except as to Plough itself and another licensee, the Block Drug Company.3 The directed verdict in favor of the defendants, and the judgment entered thereon, were based on the trial court's conclusion that Moraine had failed to establish that the Plough-Atlas license agreement or the defendant's conduct incidental thereto had unreasonably restrained trade in violation of 15 U.S.C. § 1 or that the agreement or conduct constituted a tortious interference with any business relationship of Moraine. More specifically, the trial court directed a verdict on the "ground that (the) contract dated Jan. 19, 1961 between the defendants is valid and does not violate antitrust laws," and on the further ground that Moraine had cited no evidence in support of its contention that Plough had participated in causing defendant Atlas to rescind its pre-existing contract with Moraine on January 18, 1961.

Some time during the 1950's, Dr. J. Alfred Rider, a gastroenterologist in San Francisco, California, began research work with a composition containing methylpolysiloxane (MPS) and finely divided silica4 for the relief of gas or flatulence in human beings. Also during the 1950's, Dr. Wolffe Harry Feinstone filed a patent application for a simethicone-containing composition for the relief of flatulence in humans.

In 1958, Moraine Products was incorporated for the purpose, among others, of promoting Rider's invention. In August 1958, Stuart approached Rider and expressed an interest in marketing deflatulent products embodying his invention. On March 11, 1960, Stuart entered into an agreement with Rider and Moraine. Under that agreement, Stuart was to pay a 5% royalty on its sales of simethicone-containing products in return for an assignment of Rider's alleged invention, including his United States and foreign patent applications thereon. The agreement was to be effective whether or not a U. S. patent was obtained.

Early in 1960, Stuart began marketing a deflatulent product called Mylicon, which contained MPS plus finely divided silica, i. e., simethicone, in accordance with Rider's invention. On August 30, 1960, the Feinstone patent was issued. This patent had both method and product claims covering the treatment of gastrointestinal distress through the use of any antacid material coated with an organo-poly-siloxane. While the claims of the Feinstone patent were not directed to the identical invention covered by the Rider application, Stuart became concerned about its marketing position. Feinstone and Plough, with which corporation he had become associated, advised Stuart that its Mylicon infringed Feinstone's recently issued patent.

After months of correspondence and meetings, both with Rider and Plough, Stuart rescinded its 1960 agreement with Moraine on January 18, 1961. On the next day, January 19, 1961, Stuart signed the previously mentioned patent licensing agreement with Plough. Moraine subsequently brought suit against Stuart (Atlas) for its termination of the 1960 agreement. In June 1962 that suit was settled upon terms which, among other things, granted Atlas an option to obtain exclusive licenses for itself or others under any Rider patent that subsequently issued on his then-pending patent application.

After ten years of patent prosecution, the Rider patent issued on January 14, 1969. Within the thirty days provided in the settlement agreement of June 28, 1962, Atlas exercised its right to obtain an exclusive license under the Rider patent. Shortly thereafter, upon receipt of a letter of demand from Atlas, Moraine sued Plough and Block for patent infringement, the suits eventually being consolidated in the Northern District of Illinois. Thereafter in mid 1970 Atlas sued Moraine in the Eastern District of Michigan seeking, among other things, a declaratory judgment that the Rider patent was invalid. Since the patent suits were filed at approximately the same time as this antitrust suit, pretrial discovery proceeded essentially on a consolidated basis. The evidence developed for purpose of the patent suits was utilized in the present antitrust action.

The patent litigation has proceeded to a final judgment. In Atlas Chemical Industries, Inc. v. Moraine Products, Inc., 350 F. Supp. 353 (E.D.Mich.1972), aff'd in part, rev'd in part, 509 F.2d 1 (6th Cir. 1974), Atlas secured a final adjudication that the Rider patent for the use of simethicone for the relief of gastrointestinal gas and flatulence in humans was invalid. The reported opinions of the Michigan district court and of the Sixth Circuit set forth the essential narrative details and provide a contextual background for the present litigation and need not be repeated here.

The issues in this appeal are numerous and, at least as presented, are complex. Although Atlas filed a cross-appeal, this was withdrawn two days before its answering brief was filed. Some of the issues raised in the answer brief suggest the possibility that as a tactical matter the appellee has determined to present contentions which might more properly have been raised by the cross-appeal resulting from the trial court's striking of certain defenses and counterclaims. We regard the basic issue in this appeal as being whether the district court should have granted a directed verdict; and, being mindful that in considering the correctness of trial court action on review we are not necessarily confined to the reasons advanced in the trial court for the purpose of sustaining the action, we will consider the various contentions of the appellee only to the extent that they would justify the directed verdict. We find little reason for becoming involved in the eristic briefing pertaining to the scope of the decision of the litigation in the Sixth Circuit.

The first aspect of the main issue is found in a matter which appears to be of first impression and is whether it is a per se violation of Section 1 of the Sherman Act, 15 U.S.C. § 1, for the assignee of a patent in licensing a competitor to agree with that competitor that with the exception of another competitor the assignee will not license anyone else although retaining the right to use the patent itself and the first licensee will not grant any sub-licenses. The effect of the Plough-Atlas agreement which embodies the above commitments is that further licensing under the patent can only be accomplished by joint agreement between the two signatories of the licensing agreement.

Moraine contends, that under United States v. Krasnov, 143 F. Supp. 184 (E.D.Pa.1956), aff'd. per curiam, 355 U.S. 5, 78 S. Ct. 34, 2 L. Ed. 2d 21 (1957), petition for rehearing denied, 355 U.S. 908, 78 S. Ct. 328, 2 L. Ed. 2d 262, the January 19, 1961 agreement between Plough and Atlas was illegal. Undergirding the present question, in part at least, are the competing philosophies of patent and antitrust champions. The speeches and articles of governmental antitrust officials have expressed the view that the days of benign neglect towards restrictions on competition by patentees should be numbered. See M. Adelman & F. Juenger, Patent-Antitrust: Patent Dynamics and Field of Use Licensing, 50 N.Y.U.L.Rev. 273, 274 (1975).

At a panel discussion on patent licensing in August 1973 Bruce B. Wilson, then Deputy Assistant Attorney General, Antitrust Division, stated quite candidly:

. . . (T)he Department believes it to be unlawful for a patentee to agree with his licensee that he will not, without the licensee's consent, grant further licenses to any other person. That is the old Krasnov principle. 42 Antitrust L.J. 683 (1973).

At the same meeting, George Frost, Director of the Patent Section of General Motors, while presenting his own views on matters of patent policy, noted the stringent applications of 15 U.S.C. § 1 and observed:

Another class of cases involves restrictive license provisions for the benefit of the licensee. For example, a license veto power over the grant of future licenses has been held illegal. (citing United States v. Besser Mfg. Co., 96 F. Supp. 304 (E.D.Mich.1951), aff'd, 343 U.S. 444 (72 S. Ct. 838, 96 L. Ed. 1063) (1952); and Krasnov, supra.) Such restrictions on the patentee's conduct can hardly be considered a part of the right of the patentee to exclude others.

Address by George E. Frost, Restrictions on Fields of Use and Territories, Annual Meeting of the Antitrust Section of the American Bar Association, Aug. 7, 1973, in 42 id. 633, 636-37.

Although officials of the Antitrust Division have indicated the view stated above, recent filings of civil actions challenging the validity of licensing practices apparently reflect diffuse approaches, so that commentators find difficulty in uncovering "any consistent theories that might guide patentees, licensees and their counsel in structuring licensing programs." Adelman & Juenger, supra at 300-01.

In sum, patent and antitrust champions have debated for a number of years the reach of such cases as Krasnov and Besser. While Moraine argues that these cases require a conclusion that the Plough-Atlas agreement was illegal, Atlas contends that such cases as United States v. Singer Mfg. Co., 374 U.S. 174, 83 S. Ct. 1773, 10 L. Ed. 2d 823 (1963); Virtue v. Creamery Package Mfg. Co., 227 U.S. 8, 33 S. Ct. 202, 57 L. Ed. 393 (1913); Bement v. National Harrow Co., 186 U.S. 70, 22 S. Ct. 747, 46 L. Ed. 1058 (1902); LaSalle St. Press, Inc. v. McCormick & Henderson, Inc., 445 F.2d 84 (7th Cir. 1971); Bela Seating Co. v. Poloron Products, Inc., 438 F.2d 733 (7th Cir. 1971), cert. denied, 403 U.S. 922, 91 S. Ct. 2228, 29 L. Ed. 2d 701; Cataphote Corp. v. DeSoto Chemical Coatings, Inc., 450 F.2d 769 (9th Cir. 1971), cert. denied, 408 U.S. 929, 92 S. Ct. 2497, 33 L. Ed. 2d 341 (1972); and Mechanical Ice Tray Corp. v. General Motors Corp., 144 F.2d 720 (2d Cir. 1944), cert. denied, 324 U.S. 844, 65 S. Ct. 679, 89 L. Ed. 1406 (1945), support its position that the challenged agreement was perfectly legal.

Although Atlas so argues to this court, its officers gave testimony concerning their awareness that there was an "antitrust problem" regarding the challenged agreement. Atlas now presses its view that the January 19, 1961, agreement was an exclusive but not sole license. It correctly notes that this court has recognized, in Flexwood Co. v. Faussner & Co., 145 F.2d 528, 539 (7th Cir. 1944), that an exclusive license may involve more than a single licensee. It points to the above cited cases where the courts have approved licenses at least analogous to that under challenge in this case.

We think that the implicit premise of the Atlas argument is that it should not be penalized for entering an agreement in earlier years when the decisional law upheld its validity. As so construed, there is some persuasive force to the argument. An examination of the cases cited by Moraine and Atlas causes us to conclude that none of them is controlling on the question of per se illegality. Krasnov involved pooling of patents, price fixing and the institution of punitive infringement suits for the purpose of harassment rather than with the intention that they be adjudicated on the merits. In Besser, the conspiracy involved pooling of patents, an agreement to pool future improvement patents, interlocking directorates, the use of improper influence on suppliers or competitors to curtail their output, and recklessly made threats of suit. Each of the leading cases cited by Moraine in support of its theory of per se illegality involved a broad spectrum of anticompetitive conduct on the part of the defendants.

Resort to the writings of both patent and antitrust specialists provides a framework for judicial analysis but does not effectively resolve the basic question here presented. The precise contours of the Antitrust Division's belief as indicated above have not been traced either in test litigation or in a series of decided cases. A former chief of the Antitrust Division, the late Judge Richard McLaren, once stated his conclusions regarding Krasnov :

One of my predecessors suggested and I think rightly that the rule of Krasnov should be extended to make unlawful any sole or exclusive license without sublicensing rights. 5 CCH Trade Reg.Rep. P 50,106, at 55,123 (1972).

Of course, the suggestion that the rule should be extended impliedly reflects a discerning awareness that the decision itself did not have unlimited reach. The absence of any exacting analysis of the precise nature of the views of the Division, which this court although not bound thereby would respectfully consider, is but one of the factors which makes this court hesitate in accepting Moraine's reading of the outer reaches of the Supreme Court's per curiam opinion.

Whatever may have been the view of Division officials at an earlier time, the current legal literature contains at least one hint that the present view of the Antitrust Division is not consistent with a conclusion of per se illegality. The Department of Justice has recently announced that it will not oppose a proposed patent licensing agreement between the Salk Institute for Biological Studies and several pharmaceutical companies for the clinical testing and commercial development of somatostatin, a drug which Salk believes may be useful in treating certain forms of diabetes and other diseases. See "Antitrust Division Okays Salk Patent Licensing Plan," 745 ATRR at A-16 to A-17 (January 6, 1976).

The Salk Institute is a public, nonprofit foundation which performs basic biological research in its laboratories. It is supported primarily by grants from United States Government agencies and private charitable contributions. The Institute owns a patent and two pending patent applications for somatostatin and analogs thereof, but it lacks the necessary facilities and financial resources for clinical testing. It is estimated that the average cost of bringing a new drug to the market exceeds 10 million dollars.

In order to induce qualified pharmaceutical firms to conduct the necessary clinical testing of somatostatin and to obtain royalty income to fund current as well as new research, the Institute proposed to license commercial pharmaceutical firms to manufacture and sell somatostatin. One of the terms of the option agreements is a commitment by Salk to limit its initial licensing to five worldwide nonexclusive licensees, no more than two of which shall be granted to United States firms and no more than three to European firms. These limitations would apply with respect to each country until three years after the first sale in that country by a licensee to the general public of any product covered by the patent, except that Salk would be free to grant further non-exclusive licenses anywhere in the world three years after the first such sale in the United States.

In a letter dated December 16, 1975, Acting Assistant Attorney General Bruce B. Wilson explained that although the Antitrust Division generally believes that an agreement among the licensor and its prospective licensees to limit the number of licenses is probably anticompetitive, the Institute has shown that such a limitation is the least anticompetitive way to develop somatostatin on reasonable terms. Wilson also stated that the temporary limitation on the number of licensees appeared to be reasonable because Salk had been unable to obtain license agreements with qualified and interested firms without such a limitation. Id. at A-17.

Even if this court assumes that Salk's use of nonexclusive licenses distinguishes the proposed licensing arrangement from that involved in the marketing of simethicone-containing deflatulents by Atlas and Block as exclusive licensees, it is difficult to reconcile the Division's approval of the Salk plan with an inflexible reading of either Besser or Krasnov, since the latter case involved the grant of a nonexclusive license to make, use and sell slip covers. 143 F. Supp. at 192.

While we may at this point conclude that the reach of Krasnov should not have an unlimited extension, argument for some extension retains strength. An analysis of the situation appears in R. Nordhaus & E. Jurow, Patent-Antitrust Law § 122, at 434-35 (2d ed. rev. 1972), which is summarized in the following three paragraphs of this opinion.

An absolute exclusive license prevents the licensor from granting any license other than the immediate one. The grant of a license accompanied by a covenant that the licensor will not grant any other licenses without the consent of the immediate licensee permits the licensor to grant other licenses under the specified conditions. Arguably, a license of the latter character has a less severe effect on competition than the absolute exclusive license. The frailty of this argument is that it overlooks the practical operation of this type of limited exclusive license.

Where the patentee or his assignee has granted an absolute exclusive license, the exclusive licensee usually has the authority to grant sublicenses. An applicant for a sublicense can secure one by obtaining the consent of the exclusive licensee. Where the patentee or his assignee has granted a nonexclusive license, under which the licensor retains the power to grant other licenses, an applicant for a license can secure one by obtaining the consent of the licensor. In either case, therefore, an applicant may obtain a license under the patent by meeting the requirements of only a single party.

The case of a limited exclusive license is strikingly different. Here, the applicant must secure the consent of the two parties, the licensor and its immediate licensee. This arrangement, which requires joint action of two parties before an applicant can secure a license, permits undesirable applicants to be boycotted through concerted action. Thus,

(u)nder the general principle that, although a patent owner acting alone may refuse to license anyone, it can not refuse to grant a license when the refusal is pursuant to a conspiracy or combination, it would appear that a limited exclusive license of the type in question is objectionable under the antitrust laws. Nordhaus & Jurow, supra at § 122, at 434-35. (Footnote omitted.)

The objectionable nature of the limited exclusive license rests basically upon an analysis of the anticompetitive possibilities that such a licensing scheme presents. The authors whose analysis we have paraphrased go on to observe, citing, inter alia, Krasnov and Besser, that the issue remains unsettled.

It would appear, of course, that the issue had been settled if, as Atlas contends, and the district court apparently thought, the challenged agreement was perfectly legal. We have earlier herein cited the cases on which Atlas relies in this regard. In our opinion none of the cited cases are directly dispositive of the present issue, but because of the breadth of language in the cases we deem it necessary to discuss the principal ones.

The seminal case is Bement v. National Harrow Co., 186 U.S. 70, 22 S. Ct. 747, 46 L. Ed. 1058 (1902). In that case, the Supreme Court held that there was nothing which violated the Sherman Act in the agreement between the parties that National would not license any other person than Bement to manufacture or sell any harrow of the peculiar style and construction then used or sold by Bement. At first reading, the case seems to support the argument of Atlas that the Supreme Court has ruled on the legality of its license agreement. There can be no question that the Court stated that, apart from some exceptions, "the general rule is absolute freedom in the use or sale of rights under the patent laws of the United States." Id. at 91, 22 S.Ct. at 755.

The holding in Bement, however, must be read within the context of the procedural status of that case. The Court was limited in its review by the findings of fact made by the referee, who had omitted to find from the evidence that there were in fact other manufacturers of harrows who had entered into the same kind of contracts with National. Because its review was limited to the findings of fact which had actually been made, the Court concluded its opinion:

If such similar agreements had been made, and if, when executed, they would have formed an illegal combination within the act of Congress, we cannot presume for the purpose of reversing this judgment, in the absence of any finding to that effect, that they were made and became effective as an illegal combination. As between these parties, we hold that the agreements A and B actually entered into were not a violation of the act. We are not called upon to express an opinion upon a state of facts not found. Id. at 95, 22 S.Ct. at 757.

In view of this language, Bement cannot be read as expressly sanctioning the agreement or conduct of either Plough or Atlas. In the present case, there is the additional fact, supported by the evidence, that Plough licensed Block Drug Company, but did not allow Block (through its wholly owned subsidiary Reed & Carnrick) "to sublicense either in the United States or in foreign countries . . .," and that both Plough and Atlas contemplated, at the time of their agreement, that only one other limited license would be granted under the Feinstone patent.

Moreover, the Supreme Court's language expressly recognized that there were exceptions to the general rule of absolute freedom in the use or sale of patent rights. It begs the question raised in this appeal to assert that one of this court's decisions, viz., Bela Seating Co. v. Poloron Products, 438 F.2d at 739, recognized a general rule of freedom. This court's decision did paraphrase the language of Bement, but it did so in the context of a claim of discrimination in royalty rates charged to undifferentiable competitors under a patent. That case was much different from the present litigation, and it did not confront the question of whether the assignee of a patent could agree with a competitor that no further licenses would be granted without its consent.

Similarly, there is nothing in Virtue v. Creamery Package Mfg. Co., 227 U.S. 8, 33 S. Ct. 202, 57 L. Ed. 393 (1913), that relates directly to the present issue. In that case, the Supreme Court ruled that the Owatonna Company and the Creamery Package Manufacturing Company, its exclusive sales agent, did not violate the Sherman Act by bringing infringement suits against D. E. Virtue, one of the joint inventors of a churn and butter worker. A necessary element in Virtue's charge of an antitrust violation was the cooperation of the corporate defendants in bringing the infringement suits. Apart from the formal contracts, Virtue focused on the coincidence in time of the suits as proof of concerted action on the part of the defendants and of a conspiracy to destroy the business of the plaintiffs. 227 U.S. at 37, 33 S. Ct. 202. The Court found this contention untenable, and ruled that the provision in the contracts for the prosecution of infringement suits was but an assurance of the rights conveyed in valid contracts. The Court found it unnecessary to analyze or to discuss particular covenants in the "very elaborate and verbose" contract, id. at 26, 33 S. Ct. 202, and its statement that patent rights might be conveyed partially or entirely and that the monopoly of use, of manufacture or of sale was not condemned by law must be read in the light of the actual Court analysis.

The district court although recognizing that the holding in United States v. Singer Mfg. Co., 374 U.S. 174, 83 S. Ct. 1773, 10 L. Ed. 2d 823 (1963), had no bearing on the issue before it nevertheless found support in a paragraph of the Court's opinion specifying what was not involved in the case before it. We find no basis of disposition of the issue before us in this negative paragraph other than the general proposition, with which we certainly do not quarrel, that the very purpose of patenting inventions is to furnish substantial exclusive rights to the holder of the patent. Nor do we find any of the other cases relied upon by Atlas determinative of the issue before us. While the cases in varying degrees contain language pointing to a conclusion that patent holders have broad freedom in licensing, we are not aware of any language specifically creating a theory of unswerving supremacy of patent law over antitrust law nor establishing in a patent licensing situation an absolute immunity from antitrust law.

Likewise the relevant statute provides no definitive answer. 35 U.S.C. § 261 provides in relevant part:

The applicant, patentee, or his assigns or legal representatives may in like manner (by an instrument in writing) grant and convey an exclusive right under his application for patent, or patents, to the whole or any specified part of the United States.

From a historical standpoint, it appears that the quoted language was originally intended to apply to neither assignments nor licenses, but rather to what were classified as "grants" in the nineteenth century. There is a technical distinction between a grant, an assignment, and a license. A historical and literal reading of the statute points to the conclusion that the section does not really apply at all to patent licenses.5 Or, as it has been expressed more recently, "Patent licenses are not governed by the Patent Act, Section 261 being inapplicable to licensees." P. Rosenberg, Patent Law Fundamentals 264 (1975).

Even though the statutory section is not directly applicable, this court recognizes that the difference between a "grant" and an exclusive license is merely formal in nature. The litigation has been complicated by a continuing dispute over the nature of the license granted by Plough to Atlas. Moraine characterizes the challenged agreement as creating a non-exclusive license, probably because the leading cases undergirding its per se theory have involved a non-exclusive licensee agreeing not to sublicense. Atlas contends that it is indeed an exclusive licensee, subject to a reserved right in the licensor to grant a non-exclusive license to a third party. We hold that Atlas is an exclusive licensee.

However, this holding does not help in the resolution of the issue in this appeal. Nothing in 35 U.S.C. § 261 or the decided cases sanctions an arrangement wherein competitors agree with one another that they will not grant further sublicenses. The bare language of § 261 does allow a patentee or his assignee to grant an exclusive license to make, use, or sell the patented invention. But the statutory language must be construed in connection with antitrust law.

The provision of the Sherman Act prohibiting every contract, combination or conspiracy in restraint of trade or commerce among the several states is literally all encompassing, but courts have construed it as precluding only those contracts or combinations which "unreasonably" restrain competition. Northern Pac. Ry. Co. v. United States, 356 U.S. 1, 78 S. Ct. 514, 2 L. Ed. 2d 545 (1958). Some business arrangements, such as price-fixing agreements, division of markets among competitors, tying arrangements, and group boycotts, have such an obvious effect on competition that they are conclusively presumed to be illegal; i. e., they are per se violations of the Act. Mogul v. General Motors Corp., 391 F. Supp. 1305 (D.C.Pa.1975). Where activity allegedly in violation of antitrust law has not been held per se violative of Section 1 of the Sherman Act, the traditional "Rule of Reason" analysis has been held applicable. Generally speaking, such analysis involves, inter alia, a study of the consequences of the conduct on the affected market before imposition of antitrust sanctions. See George R. Whitten, Jr., Inc. v. Paddock Pool Builders, Inc., 508 F.2d 547, 559 (1st Cir. 1974), cert. denied, 421 U.S. 1004, 95 S. Ct. 2407, 44 L. Ed. 2d 673 (1975).

The patent structure of the drug industry is peculiar in several respects, H. Miller, Patent License Restrictions in the Prescription Drug Industry, 53 Va.L.Rev. 1283, 1292 (1967), and license restraints of many defined types are quite common in the pharmaceutical industry. Adelman & Juenger, supra at 299. A considerable body of literature has developed regarding field-of-use restrictions, and expert commentators have suggested that a desirable policy of enforced patent insecurity can be effectuated, and competition restimulated, by adopting a limited number of rules that would classify patent licensing practices in the drug industry, at least for antitrust purposes, in accordance with their propensity to suppress patent attack by licensees. Id. at 305.

The record in this appeal does not give the court the information it would need in order to undertake the kind of economic and policy analysis which should properly undergird any determination to add to the pantheon of prohibited per se practices. "The correct economic evaluation of patent-licensing arrangements under patent law or antitrust law, as with similar arrangements when patents are not involved, is a complicated process and deserves careful analysis. . . . " W. Bowman, Patent and Antitrust Law: A Legal and Economic Appraisal xi (1973). The substantial acceptance, by both the legal and economic professions, of proscriptive judicial opinions relating to licensor-licensee contracts under both patent and antitrust law cannot substitute for critical analysis in search of better answers to the relevant policy questions. Cf. id. at 255-56.

We cannot fault the district court for its apparent unwillingness to find a per se violation. The strict per se rules of modern antitrust law establish a conclusive presumption that a particular kind of action has improper anticompetitive effects, and this presumption governs regardless of whether the particular conduct in its actual context has been proved to have those consequences. Carter-Wallace, Inc. v. United States, 449 F.2d 1374, 1381, 196 Ct. Cl. 35 (1971). Courts impose these presumptions where they have had sufficient prior experience with a type of action to feel justified in concluding that previously discerned anticompetitive effects will always, or nearly always, be present once the existence of a practice is shown or where, even without that actual experience, the court reaches the same conclusion through analysis. Id. In sum, per se rules are judicially created to deter agreements or practices which constitute unreasonable restraints on trade or which have such a pernicious effect on competition and so lack any redeeming virtue that a declaration of illegality can be made without elaborate inquiry into the precise harm they cause or the business excuse for their use. E.g., United States v. Topco Associates, Inc., 405 U.S. 596, 607, 92 S. Ct. 1126, 31 L. Ed. 2d 515 (1972); White Motor Co. v. United States, 372 U.S. 253, 262, 83 S. Ct. 696, 9 L. Ed. 2d 738 (1963); Northern Pacific, supra, 356 U.S. at 5, 78 S. Ct. 514.

No doubt it is possible for manufacturers to draft specific patent license terms so as to insure the cooperation of licensees in effecting anticompetitive ends. See Marquis, supra at 50. Where a challenge is made to a particular form of restrictive licensing, the analytic problem is to uncover a readily accessible and reliable indicator of the anticompetitive tendencies of particular license agreements See id. at 93. Many types of license agreements can have an adverse effect upon competition, but it is questionable whether the traditional judicial focus on specific license restrictions rather than on the entire competitive setting surrounding the licensing of patents represents the preferable method of balancing antitrust and patent policies. See id. at 90.

Where licensing practices suppress incentives to attack patent validity, to invent around the patent, or to employ available technology, they can hardly be considered a legitimate exercise of patent rights immune from antitrust scrutiny. Adelman & Juenger, supra at 296. Economic analysis suggests that the licensor's business interest is often not advanced by licensing all applicants. Antitrust decisions have sometimes stringently regulated patent licensing contracts which are efficient not only in the economic but also in the social sense. Cf. Bowman, supra at 241. A licensor is not always able to recoup the profits surrendered to competing licensees, Adelman & Juenger, supra at 298, and it is not difficult to see that the apprehension of such a possibility would exert influence upon the decisions of sophisticated businessmen.

Where a patent license is used to protect the licensee in addition to the patentee or is used to allow the licensees to divide a market among themselves, thus enabling them jointly to regiment an industry under the guise of a patent license, there is good reason to declare such a restrictive scheme illegal. Miller, supra at 1301. Although not left unchallenged, there is evidence in the record pointing to the conclusion that Atlas, Plough, and Block conspired to divide the market for simethicone-containing antacids among themselves. Similarly, there is an evidentiary basis for Atlas' assertions of sound commercial grounds for its desiring the challenged patent license and consistently pressing Plough for its literal enforcement.

(11) On such a record, the erection of a judicial presumption that the challenged license agreement was pernicious or totally lacking any redeeming economic virtue is unwarranted. The continuing debate between patent and antitrust champions has not so conclusively established the anticompetitive purpose or effect of a mutual agreement not to grant sublicenses that this, or any other court, can properly eschew economic analysis and rule a priori that, upon execution, the January 19, 1961, license contract was illegal per se. Such a narrow focus on the specific terms of a licensing arrangement would ignore the body of sophisticated legal and economic literature to the effect that truncated judicial vision has frequently upset desirable commercial practices. It would draw a line differentiating legal and illegal practices upon a presumption not fairly drawn from the factual realities of drug research and marketing. Thus, the district court's refusal to find per se illegality in the challenged agreement was not erroneous.

On the other hand, this court is not prepared to take the additional step and to conclude, as Atlas insists it should, that the 1961 licensing arrangement was entirely lawful. Such a conclusion would circumvent completely an application of the traditional "Rule of Reason" analysis. Accordingly, it is necessary to examine Moraine's argument that the evidence it presented established, at least on the record, sufficient facts to present a jury question as to an antitrust violation by Atlas. It is not necessary for this court itself to conduct the "Rule of Reason" analysis if it determines that the trial court erred in its entry of judgment. Upon a finding of trial court error in this regard, a remand for further proceedings would provide the mechanism for the proper reconciliation of patent and antitrust policies.

Only unreasonable restraints of trade or commerce are within the prohibition of the Sherman Act. Standard Oil v. United States, 221 U.S. 1, 31 S. Ct. 502, 55 L. Ed. 619 (1911). Since not all contracts among potential competitors are prohibited, at some time in the course of litigation in the absence of a per se violation there must be applied a "Rule of Reason" analysis, which includes consideration of the facts peculiar to the business in which the restraint is applied, the nature of the restraint and its effects, and the history of the restraint and the reasons for its adoption. See Topco Associates, supra, 405 U.S. at 607, 92 S. Ct. 1126. It is well established that only those acts, contracts or agreements which unduly obstruct the due course of trade, or which injuriously restrain trade because of their inherent nature or effect, or because of their evident purpose are unlawful under the "Rule of Reason." E.g., Kestenbaum v. Falstaff Brewing Corp., 514 F.2d 690 (5th Cir. 1975), cert. denied, --- U.S. ----, 96 S. Ct. 1412, 47 L. Ed. 2d 349, 44 U.S.L.W. 3493 (1976).

The function of this court in reviewing the propriety of a directed verdict is to examine the entire record to determine whether there were any jury questions. Stief v. J. A. Sexauer Mfg. Co., 380 F.2d 453 (2d Cir. 1967), cert. denied, 389 U.S. 897, 88 S. Ct. 220, 19 L. Ed. 2d 216. There is no dispute between the parties concerning the test applicable in this case. The rule that, upon motion for a directed verdict, the evidence must be viewed in the light most favorable to the party against whom the motion is made and that the opposing party must be given the benefit of all inferences which the evidence fairly supports, even though contrary inferences might reasonably be drawn, governs rulings upon motions for directed verdict in treble damage suits under the antitrust laws. Continental Ore Co. v. Union Carbide & Carbon Corp., 370 U.S. 690, 696 n. 6, 82 S. Ct. 1404, 8 L. Ed. 2d 777 (1962).

While Atlas argues that the trial court correctly determined that there was no unreasonable restraint of trade under any circumstances, per se or otherwise, we do not read the decision under review as sufficiently addressing the "otherwise" aspect. In essence, it appears to us that the district court went no further than determining that the authorities hold that an owner of a patent has the right to grant an exclusive license if it so desires and, having engaged in such legal conduct, it has not violated the antitrust laws. Although the trial court recognized the unsettled state of the law and concluded, as have we, that Moraine had not established a per se violation, we do not find any basis for determining that there was any "Rule of Reason" analysis. No written opinion was provided by the trial court, and we have had to reach the conclusion that we have on the basis of bench pronouncements.

We turn then to an analysis of the evidence in discharge of our function to determine if a jury, properly instructed in the law, could properly determine from the evidence, including the drawing of permissible inferences, that Atlas had violated the prohibitions of the Sherman Act.

We note at the outset that the focus of analysis need not be exclusively on the execution of the January 19, 1961 license agreement. Every patent license agreement is literally a contract and combination, but no formal agreement need be shown in order to establish an antitrust violation. Indeed, well counselled restraint-combiners could scarcely be expected to put or leave in their files detailed written documentation of their illegal purposes which, of course, does not mean that they might not resort to documentation of the " window-dressing" type. On a review of the instant record, we conclude there was evidence from which a jury could infer that, even prior to the formal execution of the challenged contract, there already was some form of agreement the effect of which would be to limit in numbers those catering to the needs of the substantial market of customers plagued with flatulopetic symptoms.

Many months before the execution of the license, Plough became aware of Stuart's activity in marketing its Mylicon deflatulent. Shortly thereafter, a meeting was held between the presidents of the two companies. During the course of the fall and the early winter of 1960-61, there were a number of meetings and the exchange of letters. For commercial reasons, Stuart wanted an exclusive license under the Feinstone patent. Both Plough and Stuart felt that the license should be limited to the two companies. In a letter of October 25, 1960, Stuart wrote Plough, stating:

We hope that in view of your and our plans to retain such products exclusively within Plough and Stuart effective measures can be taken to stop Reed & Carnrick as well as any others. (Emphasis added.)

The factual background of this letter was that Stuart had become aware that Reed was marketing a simethicone-containing deflatulent called Phazyme. The expression of assertedly mutual plans to gain exclusive rights in simethicone-containing antacids raises the possible inference that the two companies were already combining for an anti-competitive purpose, even though the formal instrument constituting the license had not yet been executed.6

Also on October 25, 1960, Plough wrote Stuart regarding its draft of a licensing agreement. The letter permits an inference that there was already a tentative, inchoate agreement which left room for future changes:

However, if there is anything in it that you feel requires rewriting or modification, please do not hesitate to make such modifications so that we can eventually come up with a mutually satisfactory instrument.

Although no instrument meeting the strict demands of the common law of contracts was yet in existence, the evidence of record permits an inference that the two companies were already combining to serve not merely the interests of careful license draftsmanship but the anticompetitive end of excluding others.

Nearly a month later, the representatives of the two companies met once again. The parties there agreed that, if Plough could not force Reed & Carnrick out of the market and if court action was the only alternative, Plough would be permitted to give Reed a license. Further, it was agreed that all other companies, and products of Reed other than Phazyme, would be contested in court. A proposed general agreement was reached covering the essential terms of the eventual 1961 license on November 22, 1960. A Plough letter of December 7, 1960, confirms that some understandings had been reached by late November:

It was my impression (Dr. Feinstone) and Harry Solmson's (a Plough executive) that we had resolved the few items that required modification during the course of your visit as was finally discussed and agreed upon in Harry's office. Consequently, we have revised that draft to conform with our mutual understanding as of that time. (Emphasis added.)

As antitrust specialists have observed, courts often focus on a specific license restriction rather than on the entire competitive setting surrounding the licensing of patents. See Marquis, supra at 90 n. 120. It is difficult to avoid concluding that the trial court failed to follow the salutary advice that

(a) key point to keep in mind in investigating competition is that the patent license is often only one of many agreements between the licensor and licensee, ranging from sales contracts to other patent licenses. In spite of the importance of patent licenses, therefore, a more accurate appraisal of competition is obtained by considering all other agreements and dealings between the firms in conjunction with the patent license. Although the patent license may frequently make the greatest contribution, these other agreements are important not only for the specific restrictions they impose, but for their cumulative effect upon the behavior of competitors. Id. at 90.

On the present record, this court cannot confidently assert that a jury could not reasonably infer that there were oral agreements between Plough and Atlas prior to the formal execution of the patent license. Nor can this court proscribe, as a matter of law, the drawing of jural inferences that the mutual understandings aimed at anticompetitive ends.

Plough and Atlas acted diligently in preventing the entry of other competitors (except Reed & Carnrick) into the market for human deflatulent products. From the execution of the license agreement until the approximate date on which the present litigation was commenced, well over forty drug companies applied for licenses under the Feinstone patent. All were refused. Other than the original two licenses to Atlas and Block, no licenses were granted to any other company. A jury could reasonably infer that Atlas combined with Plough in forcing other simethicone-containing products off the market.

At all pertinent times, the only simethicone-containing products which remained on the market and were available to the public throughout the United States were those of Plough, Atlas, and Block, the alleged co-conspirators. Expert testimony based upon economic data was used by Moraine to show that the license restriction had a significant effect upon the relevant market. We are persuaded that a jury could infer that there was an actual restraint of trade.

The important question, of course, was whether any restraint of trade was unreasonable or whether, if a restraint existed, it was sanctioned by the federal law applicable to patents. This court holds that the combination of the license agreement impliedly calling for the joint consent of Atlas and Plough before additional licenses would be granted, when taken in conjunction with the repeated refusals of either Atlas or Plough to grant any such sublicenses, represents conduct for which the patent statute provides no direct sanction. In the absence of any direct case law directly proscribing either the license terms or the conduct of Atlas and Plough, the ultimate question of the reasonableness of any trade restraints disclosed by the evidence was for the jury to determine.

Under Fortner Enterprises, Inc. v. United States Steel Corp., 394 U.S. 495, 500, 89 S. Ct. 1252, 1257, 22 L. Ed. 2d 495 (1969), a plaintiff who cannot meet the standards necessary to bring into play the doctrine of per se illegality can "still prevail on the merits whenever he can prove, on the basis of a more thorough examination of the purposes and effects of the practices involved, that the general standards of the Sherman Act have been violated." In that case, the district court had granted summary judgment in favor of the defendants in a private antitrust action. Here, the district court directed a verdict for the defendants, and entered judgment thereon. In so doing, the trial court acted upon a mistaken reading of the case law and upon an unformulated analysis of the evidence. Since Moraine never abandoned its "Rule of Reason" basis for recovery, the action of the trial court was in error. Unless there are alternate grounds upon which the judgment in favor of Atlas can stand, this court must reverse the judgment as to the antitrust counts and remand for a new trial. We will in the next section of this opinion address ourselves to alternative grounds advanced by Atlas for affirmance.

A different situation exists with regard to Count II of Moraine's complaint. The Plough-Atlas negotiations of late 1960 took place within the context of Rider's application for a patent. During the negotiations, Stuart advised Plough that it was considering a rescission of the Moraine-Stuart agreement of March 11, 1960. On the basis of our review of the record, we conclude that the district court properly ruled that Moraine had not produced sufficient evidence that Plough participated in the decision to rescind the Moraine-Stuart agreement. Inasmuch as Moraine and Atlas settled their controversy regarding this matter in the June 28, 1962, option agreement, the trial court's grant of a directed verdict on Count II was correct.IV. Alternative Grounds for Affirmance

On appeal, a litigant is entitled to rely on any ground for affirmance, whether or not passed upon by the trial court. Dandridge v. Williams, 397 U.S. 471, 475, 90 S. Ct. 1153, 25 L. Ed. 2d 491 (1970). Atlas argues that there are many additional sound grounds upon which the district court could have either dismissed the complaint or granted a directed verdict against Moraine. We shall not discuss all eight of the grounds upon which Atlas relies. However, we think that at least three of the asserted grounds merit discussion.

A troublesome aspect of the continuance of this litigation arises out of the fact that the essence of Moraine's damage claim is a deprivation of royalties from the Rider patent during the several year period from the time of its granting until its judicially declared invalidity. Although there is some basis to an argument that the underlying policy of Lear, Inc. v. Adkins, 395 U.S. 653, 89 S. Ct. 1902, 23 L. Ed. 2d 610 (1969), points to a conclusion that Moraine, the holder of a patent subsequently determined to be invalid, should not be able to recover damages predicated in effect upon licensing royalties, we are not persuaded that such policy considerations override Moraine's rights under the antitrust laws. While there are paradoxical aspects of allowing recovery to arise from illegal interference with the sale of something which ultimately was proven to have no sales value, it cannot be said that there was no such value during the period of the presumptive validity of the patent. We are assuming, of course, the nonexistence of facts indicating knowledge on the part of the patent holder of patent invalidity or bad faith otherwise. While Moraine was the holder of a presumptively valid patent, it could legally entertain the expectation, unless it had in some manner deprived itself thereof, of receiving royalties from licensing arrangements which in final analysis are agreements in which the licensee is purchasing the right to be free from infringement litigation, which Moraine did have to sell during the period of validity.

B. Moraine-Atlas Settlement of June 28, 1962

Moraine's suit against Atlas for damages arising from the January 18, 1961 rescission of the 1960 agreement was settled, and Atlas was granted an option to take an exclusive license under the Rider patent. When the Rider patent issued in 1969, Atlas exercised its right to take such a license. At that time, Atlas was already manufacturing and selling in the United States products covered by the Rider patent.

The gravamen of Moraine's complaint of injury is that, because potential licensees under the Rider patent also required a license under the Feinstone patent in order to practice the Rider invention, the combination unlawfully limiting licenses under the Feinstone patent necessarily limited the number of potential licensees under the Rider patent and deprived Moraine of royalty income. Atlas seeks to sustain the judgment in its favor by contending that the illegal agreement or conspiracy, if any, could not possibly have damaged Moraine. It rests this argument on a claim that, except for several narrow exceptions, the terms of its license from Moraine precluded Atlas from nominating additional licensees.

One thrust of the Atlas argument is that the 1962 agreement gave it the right to license itself or others and that once it had taken the license for itself it no longer had any legal right to license others. This approach puts great significance on the use of the disjunctive "or," a semantical significance that we are not persuaded the parties ever intended. While ordinarily the construction of documents is a legal question, the language of the 1962 agreement sufficiently lacks clarity as to the claimed restrictive limitation as to justify the admission of testimony as to the actual intent of the parties. Here there was testimony from which the jury could infer that Atlas did have sublicensing rights.

In respect to the agreement, Atlas also contends that in any event it was limited to claimed infringers in nominating others for a license. We decline to accept this strained construction of the agreement.

Atlas also argues that Moraine's claims for loss of royalties in Counts I and III of its amended complaint were inappropriately brought under the antitrust laws because Moraine lacks standing to sue. This same argument was made on several pretrial motions to dismiss, but the district court denied the motions upon the authority of Congress Building Corp. v. Loew's, Inc., 246 F.2d 587 (7th Cir. 1957).

We are not persuaded that Congress Building is precisely analogous to this case. In that decision, a theatre lessor was allowed to bring antitrust charges against conspirators whose action allegedly reduced the plaintiff's revenues from a percentage lease of the theatre. Atlas rightly observes of the facts of Congress Building that the plaintiff there had a reasonable business expectancy to protect, i. e., rental payments for its theatre lease. Here, it is contended that Moraine had not reserved any reasonable business expectancy for such royalties.

The resolution of the standing question necessarily interacts with the question of the precise legal effect of the 1962 agreement. If Moraine's agreement with Stuart contemplated additional licensees other than Atlas, the lack of nomination of licensees by Atlas, taken in conjunction with its Plough agreement, could be the basis for the determination of injury to the plaintiff who would thereby have standing to sue. What we have said under the previous subsection of this opinion is applicable to the matter of the 1962 agreement relationship to the present question. Upon remand, the ramifications of the effect of the 1962 agreement upon the present and allied questions may have to be the subject of further factual exploration. Although we think Congress Building is factually distinguishable, we decline to hold on the record before us that Moraine had no standing.

We have examined the other grounds which Atlas claims were sufficient to justify either a dismissal of the complaint or a directed verdict in its favor. We are not persuaded that Rider's supposed fraud in the procurement of his patent represents a sufficient ground for affirmance. We think there was evidence from which the jury could infer that Rider perpetrated no fraud. In the related patent litigation to which this opinion has earlier referred, the courts suggested that Moraine and its principals satisfied "the highest degree of candor and good faith" required of those having dealings with the Patent Office. Atlas Chemical Industries, 509 F.2d at 8, quoting Kingsland v. Dorsey, 338 U.S. 318, 319, 70 S. Ct. 123, 94 L. Ed. 123 (1949). The parties disagree over the findings actually made by the Michigan court on this question, but the error of the trial court in taking this case from the jury makes it unnecessary for this court to set forth at this time the principles of collateral estoppel applicable to this asserted ground for affirmance.

The record does not give this court sufficient information with which to undertake the critical legal and economic analysis which should precede the establishment of a new per se antitrust violation. We find no error in the district court's unwillingness to accept Moraine's theory that the Plough-Atlas license agreement of January 19, 1961, was a per se violation of the Sherman Act. However, the record establishes that the trial court never attempted the flexible "Rule of Reason" analysis. Nothing in 35 U.S.C. § 261 or in the decided cases expressly sanctions an arrangement wherein two competitors join together and agree that no further competitors may be licensed under a patent without the joint consent of the first two. The district court erred in directing a verdict on the antitrust counts in favor of Atlas, and in entering judgment thereon, because of its conclusion that the challenged license agreement was valid.

The district court correctly ruled that Moraine had not produced sufficient evidence to prove that Plough participated in the decision by Stuart to terminate its 1960 agreement with Moraine, and its directed verdict in favor of Atlas on the tortious interference with business relationships count (Count II) was not erroneous. The alternative grounds argued by Atlas to support the judgment below depend on the literal language of the June 28, 1962 option agreement, on policy arguments drawn from Lear, Inc. v. Adkins, supra, and on charges of fraud by Moraine and Rider. These asserted grounds commingle questions of fact, law, and policy. On this record, and viewed in the light of the applicable standard for the direction of a verdict, this court cannot ground an affirmance upon them.

The judgment of the district court is affirmed as to Count II of the amended complaint. As to Counts I and III, the judgment of the district court is reversed, and the cause is remanded for further proceedings not inconsistent with this opinion.

AFFIRMED IN PART, REVERSED IN PART AND REMANDED.